diff --git a/block.go b/block.go
index 48f768223..4ef58e913 100644
--- a/block.go
+++ b/block.go
@@ -23,10 +23,101 @@ import (
 
 	"github.com/oklog/ulid"
 	"github.com/pkg/errors"
+	"github.com/prometheus/tsdb/chunkenc"
 	"github.com/prometheus/tsdb/chunks"
+	"github.com/prometheus/tsdb/index"
 	"github.com/prometheus/tsdb/labels"
 )
 
+// IndexWriter serializes the index for a block of series data.
+// The methods must be called in the order they are specified in.
+type IndexWriter interface {
+	// AddSymbols registers all string symbols that are encountered in series
+	// and other indices.
+	AddSymbols(sym map[string]struct{}) error
+
+	// AddSeries populates the index writer with a series and its offsets
+	// of chunks that the index can reference.
+	// Implementations may require series to be insert in increasing order by
+	// their labels.
+	// The reference numbers are used to resolve entries in postings lists that
+	// are added later.
+	AddSeries(ref uint64, l labels.Labels, chunks ...chunks.Meta) error
+
+	// WriteLabelIndex serializes an index from label names to values.
+	// The passed in values chained tuples of strings of the length of names.
+	WriteLabelIndex(names []string, values []string) error
+
+	// WritePostings writes a postings list for a single label pair.
+	// The Postings here contain refs to the series that were added.
+	WritePostings(name, value string, it index.Postings) error
+
+	// Close writes any finalization and closes the resources associated with
+	// the underlying writer.
+	Close() error
+}
+
+// IndexReader provides reading access of serialized index data.
+type IndexReader interface {
+	// Symbols returns a set of string symbols that may occur in series' labels
+	// and indices.
+	Symbols() (map[string]struct{}, error)
+
+	// LabelValues returns the possible label values
+	LabelValues(names ...string) (index.StringTuples, error)
+
+	// Postings returns the postings list iterator for the label pair.
+	// The Postings here contain the offsets to the series inside the index.
+	// Found IDs are not strictly required to point to a valid Series, e.g. during
+	// background garbage collections.
+	Postings(name, value string) (index.Postings, error)
+
+	// SortedPostings returns a postings list that is reordered to be sorted
+	// by the label set of the underlying series.
+	SortedPostings(index.Postings) index.Postings
+
+	// Series populates the given labels and chunk metas for the series identified
+	// by the reference.
+	// Returns ErrNotFound if the ref does not resolve to a known series.
+	Series(ref uint64, lset *labels.Labels, chks *[]chunks.Meta) error
+
+	// LabelIndices returns the label pairs for which indices exist.
+	LabelIndices() ([][]string, error)
+
+	// Close released the underlying resources of the reader.
+	Close() error
+}
+
+// StringTuples provides access to a sorted list of string tuples.
+type StringTuples interface {
+	// Total number of tuples in the list.
+	Len() int
+	// At returns the tuple at position i.
+	At(i int) ([]string, error)
+}
+
+// ChunkWriter serializes a time block of chunked series data.
+type ChunkWriter interface {
+	// WriteChunks writes several chunks. The Chunk field of the ChunkMetas
+	// must be populated.
+	// After returning successfully, the Ref fields in the ChunkMetas
+	// are set and can be used to retrieve the chunks from the written data.
+	WriteChunks(chunks ...chunks.Meta) error
+
+	// Close writes any required finalization and closes the resources
+	// associated with the underlying writer.
+	Close() error
+}
+
+// ChunkReader provides reading access of serialized time series data.
+type ChunkReader interface {
+	// Chunk returns the series data chunk with the given reference.
+	Chunk(ref uint64) (chunkenc.Chunk, error)
+
+	// Close releases all underlying resources of the reader.
+	Close() error
+}
+
 // BlockReader provides reading access to a data block.
 type BlockReader interface {
 	// Index returns an IndexReader over the block's data.
@@ -91,8 +182,12 @@ type blockMeta struct {
 	*BlockMeta
 }
 
+const indexFilename = "index"
 const metaFilename = "meta.json"
 
+func chunkDir(dir string) string { return filepath.Join(dir, "chunks") }
+func walDir(dir string) string   { return filepath.Join(dir, "wal") }
+
 func readMetaFile(dir string) (*BlockMeta, error) {
 	b, err := ioutil.ReadFile(filepath.Join(dir, metaFilename))
 	if err != nil {
@@ -150,17 +245,17 @@ type Block struct {
 
 // OpenBlock opens the block in the directory. It can be passed a chunk pool, which is used
 // to instantiate chunk structs.
-func OpenBlock(dir string, pool chunks.Pool) (*Block, error) {
+func OpenBlock(dir string, pool chunkenc.Pool) (*Block, error) {
 	meta, err := readMetaFile(dir)
 	if err != nil {
 		return nil, err
 	}
 
-	cr, err := NewDirChunkReader(chunkDir(dir), pool)
+	cr, err := chunks.NewDirReader(chunkDir(dir), pool)
 	if err != nil {
 		return nil, err
 	}
-	ir, err := NewFileIndexReader(filepath.Join(dir, "index"))
+	ir, err := index.NewFileReader(filepath.Join(dir, "index"))
 	if err != nil {
 		return nil, err
 	}
@@ -300,7 +395,7 @@ func (pb *Block) Delete(mint, maxt int64, ms ...labels.Matcher) error {
 	stones := memTombstones{}
 
 	var lset labels.Labels
-	var chks []ChunkMeta
+	var chks []chunks.Meta
 
 Outer:
 	for p.Next() {
@@ -405,9 +500,6 @@ func (pb *Block) Snapshot(dir string) error {
 	return nil
 }
 
-func chunkDir(dir string) string { return filepath.Join(dir, "chunks") }
-func walDir(dir string) string   { return filepath.Join(dir, "wal") }
-
 func clampInterval(a, b, mint, maxt int64) (int64, int64) {
 	if a < mint {
 		a = mint
@@ -417,36 +509,3 @@ func clampInterval(a, b, mint, maxt int64) (int64, int64) {
 	}
 	return a, b
 }
-
-type mmapFile struct {
-	f *os.File
-	b []byte
-}
-
-func openMmapFile(path string) (*mmapFile, error) {
-	f, err := os.Open(path)
-	if err != nil {
-		return nil, errors.Wrap(err, "try lock file")
-	}
-	info, err := f.Stat()
-	if err != nil {
-		return nil, errors.Wrap(err, "stat")
-	}
-
-	b, err := mmap(f, int(info.Size()))
-	if err != nil {
-		return nil, errors.Wrap(err, "mmap")
-	}
-
-	return &mmapFile{f: f, b: b}, nil
-}
-
-func (f *mmapFile) Close() error {
-	err0 := munmap(f.b)
-	err1 := f.f.Close()
-
-	if err0 != nil {
-		return err0
-	}
-	return err1
-}
diff --git a/block_test.go b/block_test.go
index be6a88090..60d9fe54f 100644
--- a/block_test.go
+++ b/block_test.go
@@ -16,8 +16,10 @@ package tsdb
 import (
 	"io/ioutil"
 	"os"
+	"path/filepath"
 	"testing"
 
+	"github.com/prometheus/tsdb/index"
 	"github.com/prometheus/tsdb/testutil"
 )
 
@@ -42,9 +44,9 @@ func createEmptyBlock(t *testing.T, dir string) *Block {
 
 	testutil.Ok(t, writeMetaFile(dir, &BlockMeta{}))
 
-	ir, err := newIndexWriter(dir)
-	testutil.Ok(t, err)
-	testutil.Ok(t, ir.Close())
+	ir, err := index.NewWriter(filepath.Join(dir, indexFilename))
+	Ok(t, err)
+	Ok(t, ir.Close())
 
 	testutil.Ok(t, os.MkdirAll(chunkDir(dir), 0777))
 
diff --git a/chunks/bstream.go b/chunkenc/bstream.go
similarity index 99%
rename from chunks/bstream.go
rename to chunkenc/bstream.go
index ed651db23..a352f078e 100644
--- a/chunks/bstream.go
+++ b/chunkenc/bstream.go
@@ -39,7 +39,7 @@
 // OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
-package chunks
+package chunkenc
 
 import "io"
 
diff --git a/chunks/chunk.go b/chunkenc/chunk.go
similarity index 99%
rename from chunks/chunk.go
rename to chunkenc/chunk.go
index 4d298041e..4c85fa054 100644
--- a/chunks/chunk.go
+++ b/chunkenc/chunk.go
@@ -11,7 +11,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-package chunks
+package chunkenc
 
 import (
 	"fmt"
diff --git a/chunks/chunk_test.go b/chunkenc/chunk_test.go
similarity index 99%
rename from chunks/chunk_test.go
rename to chunkenc/chunk_test.go
index b03e52819..edabd749f 100644
--- a/chunks/chunk_test.go
+++ b/chunkenc/chunk_test.go
@@ -11,7 +11,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-package chunks
+package chunkenc
 
 import (
 	"fmt"
diff --git a/chunks/xor.go b/chunkenc/xor.go
similarity index 99%
rename from chunks/xor.go
rename to chunkenc/xor.go
index dcee466f7..b3f5b5135 100644
--- a/chunks/xor.go
+++ b/chunkenc/xor.go
@@ -41,7 +41,7 @@
 // OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
-package chunks
+package chunkenc
 
 import (
 	"encoding/binary"
diff --git a/chunks.go b/chunks/chunks.go
similarity index 63%
rename from chunks.go
rename to chunks/chunks.go
index 30ab93f80..d07cf11a6 100644
--- a/chunks.go
+++ b/chunks/chunks.go
@@ -11,18 +11,22 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-package tsdb
+package chunks
 
 import (
 	"bufio"
 	"encoding/binary"
 	"fmt"
 	"hash"
+	"hash/crc32"
 	"io"
+	"io/ioutil"
 	"os"
+	"path/filepath"
+	"strconv"
 
 	"github.com/pkg/errors"
-	"github.com/prometheus/tsdb/chunks"
+	"github.com/prometheus/tsdb/chunkenc"
 	"github.com/prometheus/tsdb/fileutil"
 )
 
@@ -31,19 +35,19 @@ const (
 	MagicChunks = 0x85BD40DD
 )
 
-// ChunkMeta holds information about a chunk of data.
-type ChunkMeta struct {
+// Meta holds information about a chunk of data.
+type Meta struct {
 	// Ref and Chunk hold either a reference that can be used to retrieve
 	// chunk data or the data itself.
 	// Generally, only one of them is set.
 	Ref   uint64
-	Chunk chunks.Chunk
+	Chunk chunkenc.Chunk
 
 	MinTime, MaxTime int64 // time range the data covers
 }
 
 // writeHash writes the chunk encoding and raw data into the provided hash.
-func (cm *ChunkMeta) writeHash(h hash.Hash) error {
+func (cm *Meta) writeHash(h hash.Hash) error {
 	if _, err := h.Write([]byte{byte(cm.Chunk.Encoding())}); err != nil {
 		return err
 	}
@@ -53,62 +57,30 @@ func (cm *ChunkMeta) writeHash(h hash.Hash) error {
 	return nil
 }
 
-// deletedIterator wraps an Iterator and makes sure any deleted metrics are not
-// returned.
-type deletedIterator struct {
-	it chunks.Iterator
+var (
+	errInvalidSize     = fmt.Errorf("invalid size")
+	errInvalidFlag     = fmt.Errorf("invalid flag")
+	errInvalidChecksum = fmt.Errorf("invalid checksum")
+)
 
-	intervals Intervals
+// The table gets initialized with sync.Once but may still cause a race
+// with any other use of the crc32 package anywhere. Thus we initialize it
+// before.
+var castagnoliTable *crc32.Table
+
+func init() {
+	castagnoliTable = crc32.MakeTable(crc32.Castagnoli)
 }
 
-func (it *deletedIterator) At() (int64, float64) {
-	return it.it.At()
+// newCRC32 initializes a CRC32 hash with a preconfigured polynomial, so the
+// polynomial may be easily changed in one location at a later time, if necessary.
+func newCRC32() hash.Hash32 {
+	return crc32.New(castagnoliTable)
 }
 
-func (it *deletedIterator) Next() bool {
-Outer:
-	for it.it.Next() {
-		ts, _ := it.it.At()
-
-		for _, tr := range it.intervals {
-			if tr.inBounds(ts) {
-				continue Outer
-			}
-
-			if ts > tr.Maxt {
-				it.intervals = it.intervals[1:]
-				continue
-			}
-
-			return true
-		}
-
-		return true
-	}
-
-	return false
-}
-
-func (it *deletedIterator) Err() error {
-	return it.it.Err()
-}
-
-// ChunkWriter serializes a time block of chunked series data.
-type ChunkWriter interface {
-	// WriteChunks writes several chunks. The Chunk field of the ChunkMetas
-	// must be populated.
-	// After returning successfully, the Ref fields in the ChunkMetas
-	// are set and can be used to retrieve the chunks from the written data.
-	WriteChunks(chunks ...ChunkMeta) error
-
-	// Close writes any required finalization and closes the resources
-	// associated with the underlying writer.
-	Close() error
-}
-
-// chunkWriter implements the ChunkWriter interface for the standard
+// Writer implements the ChunkWriter interface for the standard
 // serialization format.
-type chunkWriter struct {
+type Writer struct {
 	dirFile *os.File
 	files   []*os.File
 	wbuf    *bufio.Writer
@@ -124,7 +96,8 @@ const (
 	chunksFormatV1 = 1
 )
 
-func newChunkWriter(dir string) (*chunkWriter, error) {
+// NewWriter returns a new writer against the given directory.
+func NewWriter(dir string) (*Writer, error) {
 	if err := os.MkdirAll(dir, 0777); err != nil {
 		return nil, err
 	}
@@ -132,7 +105,7 @@ func newChunkWriter(dir string) (*chunkWriter, error) {
 	if err != nil {
 		return nil, err
 	}
-	cw := &chunkWriter{
+	cw := &Writer{
 		dirFile:     dirFile,
 		n:           0,
 		crc32:       newCRC32(),
@@ -141,7 +114,7 @@ func newChunkWriter(dir string) (*chunkWriter, error) {
 	return cw, nil
 }
 
-func (w *chunkWriter) tail() *os.File {
+func (w *Writer) tail() *os.File {
 	if len(w.files) == 0 {
 		return nil
 	}
@@ -150,7 +123,7 @@ func (w *chunkWriter) tail() *os.File {
 
 // finalizeTail writes all pending data to the current tail file,
 // truncates its size, and closes it.
-func (w *chunkWriter) finalizeTail() error {
+func (w *Writer) finalizeTail() error {
 	tf := w.tail()
 	if tf == nil {
 		return nil
@@ -174,7 +147,7 @@ func (w *chunkWriter) finalizeTail() error {
 	return tf.Close()
 }
 
-func (w *chunkWriter) cut() error {
+func (w *Writer) cut() error {
 	// Sync current tail to disk and close.
 	if err := w.finalizeTail(); err != nil {
 		return err
@@ -216,13 +189,13 @@ func (w *chunkWriter) cut() error {
 	return nil
 }
 
-func (w *chunkWriter) write(b []byte) error {
+func (w *Writer) write(b []byte) error {
 	n, err := w.wbuf.Write(b)
 	w.n += int64(n)
 	return err
 }
 
-func (w *chunkWriter) WriteChunks(chks ...ChunkMeta) error {
+func (w *Writer) WriteChunks(chks ...Meta) error {
 	// Calculate maximum space we need and cut a new segment in case
 	// we don't fit into the current one.
 	maxLen := int64(binary.MaxVarintLen32) // The number of chunks.
@@ -272,11 +245,11 @@ func (w *chunkWriter) WriteChunks(chks ...ChunkMeta) error {
 	return nil
 }
 
-func (w *chunkWriter) seq() int {
+func (w *Writer) seq() int {
 	return len(w.files) - 1
 }
 
-func (w *chunkWriter) Close() error {
+func (w *Writer) Close() error {
 	if err := w.finalizeTail(); err != nil {
 		return err
 	}
@@ -285,29 +258,40 @@ func (w *chunkWriter) Close() error {
 	return w.dirFile.Close()
 }
 
-// ChunkReader provides reading access of serialized time series data.
-type ChunkReader interface {
-	// Chunk returns the series data chunk with the given reference.
-	Chunk(ref uint64) (chunks.Chunk, error)
-
-	// Close releases all underlying resources of the reader.
-	Close() error
+// ByteSlice abstracts a byte slice.
+type ByteSlice interface {
+	Len() int
+	Range(start, end int) []byte
 }
 
-// chunkReader implements a SeriesReader for a serialized byte stream
+type realByteSlice []byte
+
+func (b realByteSlice) Len() int {
+	return len(b)
+}
+
+func (b realByteSlice) Range(start, end int) []byte {
+	return b[start:end]
+}
+
+func (b realByteSlice) Sub(start, end int) ByteSlice {
+	return b[start:end]
+}
+
+// Reader implements a SeriesReader for a serialized byte stream
 // of series data.
-type chunkReader struct {
+type Reader struct {
 	// The underlying bytes holding the encoded series data.
 	bs []ByteSlice
 
 	// Closers for resources behind the byte slices.
 	cs []io.Closer
 
-	pool chunks.Pool
+	pool chunkenc.Pool
 }
 
-func newChunkReader(bs []ByteSlice, cs []io.Closer, pool chunks.Pool) (*chunkReader, error) {
-	cr := chunkReader{pool: pool, bs: bs, cs: cs}
+func newReader(bs []ByteSlice, cs []io.Closer, pool chunkenc.Pool) (*Reader, error) {
+	cr := Reader{pool: pool, bs: bs, cs: cs}
 
 	for i, b := range cr.bs {
 		if b.Len() < 4 {
@@ -321,44 +305,44 @@ func newChunkReader(bs []ByteSlice, cs []io.Closer, pool chunks.Pool) (*chunkRea
 	return &cr, nil
 }
 
-// NewChunkReader returns a new chunk reader against the given byte slices.
-func NewChunkReader(bs []ByteSlice, pool chunks.Pool) (ChunkReader, error) {
+// NewReader returns a new chunk reader against the given byte slices.
+func NewReader(bs []ByteSlice, pool chunkenc.Pool) (*Reader, error) {
 	if pool == nil {
-		pool = chunks.NewPool()
+		pool = chunkenc.NewPool()
 	}
-	return newChunkReader(bs, nil, pool)
+	return newReader(bs, nil, pool)
 }
 
-// NewDirChunkReader returns a new ChunkReader against sequentially numbered files in the
+// NewDirReader returns a new Reader against sequentially numbered files in the
 // given directory.
-func NewDirChunkReader(dir string, pool chunks.Pool) (ChunkReader, error) {
+func NewDirReader(dir string, pool chunkenc.Pool) (*Reader, error) {
 	files, err := sequenceFiles(dir)
 	if err != nil {
 		return nil, err
 	}
 	if pool == nil {
-		pool = chunks.NewPool()
+		pool = chunkenc.NewPool()
 	}
 
 	var bs []ByteSlice
 	var cs []io.Closer
 
 	for _, fn := range files {
-		f, err := openMmapFile(fn)
+		f, err := fileutil.OpenMmapFile(fn)
 		if err != nil {
 			return nil, errors.Wrapf(err, "mmap files")
 		}
 		cs = append(cs, f)
-		bs = append(bs, realByteSlice(f.b))
+		bs = append(bs, realByteSlice(f.Bytes()))
 	}
-	return newChunkReader(bs, cs, pool)
+	return newReader(bs, cs, pool)
 }
 
-func (s *chunkReader) Close() error {
+func (s *Reader) Close() error {
 	return closeAll(s.cs...)
 }
 
-func (s *chunkReader) Chunk(ref uint64) (chunks.Chunk, error) {
+func (s *Reader) Chunk(ref uint64) (chunkenc.Chunk, error) {
 	var (
 		seq = int(ref >> 32)
 		off = int((ref << 32) >> 32)
@@ -381,5 +365,47 @@ func (s *chunkReader) Chunk(ref uint64) (chunks.Chunk, error) {
 	}
 	r = b.Range(off+n, off+n+int(l))
 
-	return s.pool.Get(chunks.Encoding(r[0]), r[1:1+l])
+	return s.pool.Get(chunkenc.Encoding(r[0]), r[1:1+l])
+}
+
+func nextSequenceFile(dir string) (string, int, error) {
+	names, err := fileutil.ReadDir(dir)
+	if err != nil {
+		return "", 0, err
+	}
+
+	i := uint64(0)
+	for _, n := range names {
+		j, err := strconv.ParseUint(n, 10, 64)
+		if err != nil {
+			continue
+		}
+		i = j
+	}
+	return filepath.Join(dir, fmt.Sprintf("%0.6d", i+1)), int(i + 1), nil
+}
+
+func sequenceFiles(dir string) ([]string, error) {
+	files, err := ioutil.ReadDir(dir)
+	if err != nil {
+		return nil, err
+	}
+	var res []string
+
+	for _, fi := range files {
+		if _, err := strconv.ParseUint(fi.Name(), 10, 64); err != nil {
+			continue
+		}
+		res = append(res, filepath.Join(dir, fi.Name()))
+	}
+	return res, nil
+}
+
+func closeAll(cs ...io.Closer) (err error) {
+	for _, c := range cs {
+		if e := c.Close(); e != nil {
+			err = e
+		}
+	}
+	return err
 }
diff --git a/chunks_test.go b/chunks_test.go
deleted file mode 100644
index 40a282814..000000000
--- a/chunks_test.go
+++ /dev/null
@@ -1,98 +0,0 @@
-// Copyright 2017 The Prometheus Authors
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-// http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-package tsdb
-
-import (
-	"math/rand"
-	"testing"
-
-	"github.com/pkg/errors"
-	"github.com/prometheus/tsdb/chunks"
-	"github.com/prometheus/tsdb/testutil"
-)
-
-type mockChunkReader map[uint64]chunks.Chunk
-
-func (cr mockChunkReader) Chunk(ref uint64) (chunks.Chunk, error) {
-	chk, ok := cr[ref]
-	if ok {
-		return chk, nil
-	}
-
-	return nil, errors.New("Chunk with ref not found")
-}
-
-func (cr mockChunkReader) Close() error {
-	return nil
-}
-
-func TestDeletedIterator(t *testing.T) {
-	chk := chunks.NewXORChunk()
-	app, err := chk.Appender()
-	testutil.Ok(t, err)
-	// Insert random stuff from (0, 1000).
-	act := make([]sample, 1000)
-	for i := 0; i < 1000; i++ {
-		act[i].t = int64(i)
-		act[i].v = rand.Float64()
-		app.Append(act[i].t, act[i].v)
-	}
-
-	cases := []struct {
-		r Intervals
-	}{
-		{r: Intervals{{1, 20}}},
-		{r: Intervals{{1, 10}, {12, 20}, {21, 23}, {25, 30}}},
-		{r: Intervals{{1, 10}, {12, 20}, {20, 30}}},
-		{r: Intervals{{1, 10}, {12, 23}, {25, 30}}},
-		{r: Intervals{{1, 23}, {12, 20}, {25, 30}}},
-		{r: Intervals{{1, 23}, {12, 20}, {25, 3000}}},
-		{r: Intervals{{0, 2000}}},
-		{r: Intervals{{500, 2000}}},
-		{r: Intervals{{0, 200}}},
-		{r: Intervals{{1000, 20000}}},
-	}
-
-	for _, c := range cases {
-		i := int64(-1)
-		it := &deletedIterator{it: chk.Iterator(), intervals: c.r[:]}
-		ranges := c.r[:]
-		for it.Next() {
-			i++
-			for _, tr := range ranges {
-				if tr.inBounds(i) {
-					i = tr.Maxt + 1
-					ranges = ranges[1:]
-				}
-			}
-
-			testutil.Assert(t, i < 1000 == true, "")
-
-			ts, v := it.At()
-			testutil.Equals(t, act[i].t, ts)
-			testutil.Equals(t, act[i].v, v)
-		}
-		// There has been an extra call to Next().
-		i++
-		for _, tr := range ranges {
-			if tr.inBounds(i) {
-				i = tr.Maxt + 1
-				ranges = ranges[1:]
-			}
-		}
-
-		testutil.Assert(t, i < 1000 == false, "")
-		testutil.Ok(t, it.Err())
-	}
-}
diff --git a/compact.go b/compact.go
index 5bb328aa3..989756bdb 100644
--- a/compact.go
+++ b/compact.go
@@ -26,8 +26,10 @@ import (
 	"github.com/oklog/ulid"
 	"github.com/pkg/errors"
 	"github.com/prometheus/client_golang/prometheus"
+	"github.com/prometheus/tsdb/chunkenc"
 	"github.com/prometheus/tsdb/chunks"
 	"github.com/prometheus/tsdb/fileutil"
+	"github.com/prometheus/tsdb/index"
 	"github.com/prometheus/tsdb/labels"
 )
 
@@ -65,7 +67,7 @@ type LeveledCompactor struct {
 	metrics   *compactorMetrics
 	logger    log.Logger
 	ranges    []int64
-	chunkPool chunks.Pool
+	chunkPool chunkenc.Pool
 }
 
 type compactorMetrics struct {
@@ -123,12 +125,12 @@ func newCompactorMetrics(r prometheus.Registerer) *compactorMetrics {
 }
 
 // NewLeveledCompactor returns a LeveledCompactor.
-func NewLeveledCompactor(r prometheus.Registerer, l log.Logger, ranges []int64, pool chunks.Pool) (*LeveledCompactor, error) {
+func NewLeveledCompactor(r prometheus.Registerer, l log.Logger, ranges []int64, pool chunkenc.Pool) (*LeveledCompactor, error) {
 	if len(ranges) == 0 {
 		return nil, errors.Errorf("at least one range must be provided")
 	}
 	if pool == nil {
-		pool = chunks.NewPool()
+		pool = chunkenc.NewPool()
 	}
 	return &LeveledCompactor{
 		ranges:    ranges,
@@ -370,7 +372,7 @@ type instrumentedChunkWriter struct {
 	trange  prometheus.Histogram
 }
 
-func (w *instrumentedChunkWriter) WriteChunks(chunks ...ChunkMeta) error {
+func (w *instrumentedChunkWriter) WriteChunks(chunks ...chunks.Meta) error {
 	for _, c := range chunks {
 		w.size.Observe(float64(len(c.Chunk.Bytes())))
 		w.samples.Observe(float64(c.Chunk.NumSamples()))
@@ -411,7 +413,7 @@ func (c *LeveledCompactor) write(dest string, meta *BlockMeta, blocks ...BlockRe
 	// data of all blocks.
 	var chunkw ChunkWriter
 
-	chunkw, err = newChunkWriter(chunkDir(tmp))
+	chunkw, err = chunks.NewWriter(chunkDir(tmp))
 	if err != nil {
 		return errors.Wrap(err, "open chunk writer")
 	}
@@ -425,7 +427,7 @@ func (c *LeveledCompactor) write(dest string, meta *BlockMeta, blocks ...BlockRe
 		}
 	}
 
-	indexw, err := newIndexWriter(tmp)
+	indexw, err := index.NewWriter(filepath.Join(tmp, indexFilename))
 	if err != nil {
 		return errors.Wrap(err, "open index writer")
 	}
@@ -514,7 +516,7 @@ func (c *LeveledCompactor) populateBlock(blocks []BlockReader, meta *BlockMeta,
 			allSymbols[s] = struct{}{}
 		}
 
-		all, err := indexr.Postings(allPostingsKey.Name, allPostingsKey.Value)
+		all, err := indexr.Postings("", "")
 		if err != nil {
 			return err
 		}
@@ -534,7 +536,7 @@ func (c *LeveledCompactor) populateBlock(blocks []BlockReader, meta *BlockMeta,
 
 	// We fully rebuild the postings list index from merged series.
 	var (
-		postings = newMemPostings()
+		postings = index.NewMemPostings()
 		values   = map[string]stringset{}
 		i        = uint64(0)
 	)
@@ -558,7 +560,7 @@ func (c *LeveledCompactor) populateBlock(blocks []BlockReader, meta *BlockMeta,
 					continue
 				}
 
-				newChunk := chunks.NewXORChunk()
+				newChunk := chunkenc.NewXORChunk()
 				app, err := newChunk.Appender()
 				if err != nil {
 					return err
@@ -599,7 +601,7 @@ func (c *LeveledCompactor) populateBlock(blocks []BlockReader, meta *BlockMeta,
 			}
 			valset.set(l.Value)
 		}
-		postings.add(i, lset)
+		postings.Add(i, lset)
 
 		i++
 	}
@@ -619,8 +621,8 @@ func (c *LeveledCompactor) populateBlock(blocks []BlockReader, meta *BlockMeta,
 		}
 	}
 
-	for _, l := range postings.sortedKeys() {
-		if err := indexw.WritePostings(l.Name, l.Value, postings.get(l.Name, l.Value)); err != nil {
+	for _, l := range postings.SortedKeys() {
+		if err := indexw.WritePostings(l.Name, l.Value, postings.Get(l.Name, l.Value)); err != nil {
 			return errors.Wrap(err, "write postings")
 		}
 	}
@@ -628,18 +630,18 @@ func (c *LeveledCompactor) populateBlock(blocks []BlockReader, meta *BlockMeta,
 }
 
 type compactionSeriesSet struct {
-	p          Postings
+	p          index.Postings
 	index      IndexReader
 	chunks     ChunkReader
 	tombstones TombstoneReader
 
 	l         labels.Labels
-	c         []ChunkMeta
+	c         []chunks.Meta
 	intervals Intervals
 	err       error
 }
 
-func newCompactionSeriesSet(i IndexReader, c ChunkReader, t TombstoneReader, p Postings) *compactionSeriesSet {
+func newCompactionSeriesSet(i IndexReader, c ChunkReader, t TombstoneReader, p index.Postings) *compactionSeriesSet {
 	return &compactionSeriesSet{
 		index:      i,
 		chunks:     c,
@@ -667,7 +669,7 @@ func (c *compactionSeriesSet) Next() bool {
 
 	// Remove completely deleted chunks.
 	if len(c.intervals) > 0 {
-		chks := make([]ChunkMeta, 0, len(c.c))
+		chks := make([]chunks.Meta, 0, len(c.c))
 		for _, chk := range c.c {
 			if !(Interval{chk.MinTime, chk.MaxTime}.isSubrange(c.intervals)) {
 				chks = append(chks, chk)
@@ -697,7 +699,7 @@ func (c *compactionSeriesSet) Err() error {
 	return c.p.Err()
 }
 
-func (c *compactionSeriesSet) At() (labels.Labels, []ChunkMeta, Intervals) {
+func (c *compactionSeriesSet) At() (labels.Labels, []chunks.Meta, Intervals) {
 	return c.l, c.c, c.intervals
 }
 
@@ -706,13 +708,13 @@ type compactionMerger struct {
 
 	aok, bok  bool
 	l         labels.Labels
-	c         []ChunkMeta
+	c         []chunks.Meta
 	intervals Intervals
 }
 
 type compactionSeries struct {
 	labels labels.Labels
-	chunks []*ChunkMeta
+	chunks []*chunks.Meta
 }
 
 func newCompactionMerger(a, b ChunkSeriesSet) (*compactionMerger, error) {
@@ -747,7 +749,7 @@ func (c *compactionMerger) Next() bool {
 	// While advancing child iterators the memory used for labels and chunks
 	// may be reused. When picking a series we have to store the result.
 	var lset labels.Labels
-	var chks []ChunkMeta
+	var chks []chunks.Meta
 
 	d := c.compare()
 	// Both sets contain the current series. Chain them into a single one.
@@ -788,7 +790,7 @@ func (c *compactionMerger) Err() error {
 	return c.b.Err()
 }
 
-func (c *compactionMerger) At() (labels.Labels, []ChunkMeta, Intervals) {
+func (c *compactionMerger) At() (labels.Labels, []chunks.Meta, Intervals) {
 	return c.l, c.c, c.intervals
 }
 
diff --git a/db.go b/db.go
index eeb81e8d6..522330b63 100644
--- a/db.go
+++ b/db.go
@@ -36,7 +36,7 @@ import (
 	"github.com/oklog/ulid"
 	"github.com/pkg/errors"
 	"github.com/prometheus/client_golang/prometheus"
-	"github.com/prometheus/tsdb/chunks"
+	"github.com/prometheus/tsdb/chunkenc"
 	"github.com/prometheus/tsdb/fileutil"
 	"github.com/prometheus/tsdb/labels"
 )
@@ -99,7 +99,7 @@ type DB struct {
 	logger    log.Logger
 	metrics   *dbMetrics
 	opts      *Options
-	chunkPool chunks.Pool
+	chunkPool chunkenc.Pool
 	compactor Compactor
 
 	// Mutex for that must be held when modifying the general block layout.
@@ -185,7 +185,7 @@ func Open(dir string, l log.Logger, r prometheus.Registerer, opts *Options) (db
 		donec:              make(chan struct{}),
 		stopc:              make(chan struct{}),
 		compactionsEnabled: true,
-		chunkPool:          chunks.NewPool(),
+		chunkPool:          chunkenc.NewPool(),
 	}
 	db.metrics = newDBMetrics(db, r)
 
diff --git a/encoding_helpers.go b/encoding_helpers.go
index b55c7fda9..d0805fe07 100644
--- a/encoding_helpers.go
+++ b/encoding_helpers.go
@@ -5,8 +5,12 @@ import (
 	"hash"
 	"hash/crc32"
 	"unsafe"
+
+	"github.com/pkg/errors"
 )
 
+var errInvalidSize = errors.New("invalid size")
+
 // enbuf is a helper type to populate a byte slice with various types.
 type encbuf struct {
 	b []byte
diff --git a/fileutil/mmap.go b/fileutil/mmap.go
new file mode 100644
index 000000000..a0c598254
--- /dev/null
+++ b/fileutil/mmap.go
@@ -0,0 +1,48 @@
+package fileutil
+
+import (
+	"os"
+
+	"github.com/pkg/errors"
+)
+
+type MmapFile struct {
+	f *os.File
+	b []byte
+}
+
+func OpenMmapFile(path string) (*MmapFile, error) {
+	f, err := os.Open(path)
+	if err != nil {
+		return nil, errors.Wrap(err, "try lock file")
+	}
+	info, err := f.Stat()
+	if err != nil {
+		return nil, errors.Wrap(err, "stat")
+	}
+
+	b, err := mmap(f, int(info.Size()))
+	if err != nil {
+		return nil, errors.Wrap(err, "mmap")
+	}
+
+	return &MmapFile{f: f, b: b}, nil
+}
+
+func (f *MmapFile) Close() error {
+	err0 := munmap(f.b)
+	err1 := f.f.Close()
+
+	if err0 != nil {
+		return err0
+	}
+	return err1
+}
+
+func (f *MmapFile) File() *os.File {
+	return f.f
+}
+
+func (f *MmapFile) Bytes() []byte {
+	return f.b
+}
diff --git a/db_unix.go b/fileutil/mmap_unix.go
similarity index 98%
rename from db_unix.go
rename to fileutil/mmap_unix.go
index 02c411d7f..043f4d408 100644
--- a/db_unix.go
+++ b/fileutil/mmap_unix.go
@@ -13,7 +13,7 @@
 
 // +build !windows,!plan9
 
-package tsdb
+package fileutil
 
 import (
 	"os"
diff --git a/db_windows.go b/fileutil/mmap_windows.go
similarity index 98%
rename from db_windows.go
rename to fileutil/mmap_windows.go
index 444bf4103..3bee807c2 100644
--- a/db_windows.go
+++ b/fileutil/mmap_windows.go
@@ -11,7 +11,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-package tsdb
+package fileutil
 
 import (
 	"os"
diff --git a/head.go b/head.go
index 2c1468aca..4652c9579 100644
--- a/head.go
+++ b/head.go
@@ -17,6 +17,7 @@ import (
 	"math"
 	"runtime"
 	"sort"
+	"strings"
 	"sync"
 	"sync/atomic"
 	"time"
@@ -25,7 +26,9 @@ import (
 	"github.com/go-kit/kit/log/level"
 	"github.com/pkg/errors"
 	"github.com/prometheus/client_golang/prometheus"
+	"github.com/prometheus/tsdb/chunkenc"
 	"github.com/prometheus/tsdb/chunks"
+	"github.com/prometheus/tsdb/index"
 	"github.com/prometheus/tsdb/labels"
 )
 
@@ -64,7 +67,7 @@ type Head struct {
 	symbols map[string]struct{}
 	values  map[string]stringset // label names to possible values
 
-	postings *memPostings // postings lists for terms
+	postings *index.MemPostings // postings lists for terms
 
 	tombstones memTombstones
 }
@@ -185,7 +188,7 @@ func NewHead(r prometheus.Registerer, l log.Logger, wal WAL, chunkRange int64) (
 		series:     newStripeSeries(),
 		values:     map[string]stringset{},
 		symbols:    map[string]struct{}{},
-		postings:   newUnorderedMemPostings(),
+		postings:   index.NewUnorderedMemPostings(),
 		tombstones: memTombstones{},
 	}
 	h.metrics = newHeadMetrics(h, r)
@@ -226,7 +229,7 @@ func (h *Head) processWALSamples(
 
 // ReadWAL initializes the head by consuming the write ahead log.
 func (h *Head) ReadWAL() error {
-	defer h.postings.ensureOrder()
+	defer h.postings.EnsureOrder()
 
 	r := h.wal.Reader()
 	mint := h.MinTime()
@@ -616,64 +619,14 @@ func (h *Head) gc() {
 	h.metrics.chunksRemoved.Add(float64(chunksRemoved))
 	h.metrics.chunks.Sub(float64(chunksRemoved))
 
-	// Remove deleted series IDs from the postings lists. First do a collection
-	// run where we rebuild all postings that have something to delete
-	h.postings.mtx.RLock()
-
-	type replEntry struct {
-		idx int
-		l   []uint64
-	}
-	collected := map[labels.Label]replEntry{}
-
-	for t, p := range h.postings.m {
-		repl := replEntry{idx: len(p)}
-
-		for i, id := range p {
-			if _, ok := deleted[id]; ok {
-				// First ID that got deleted, initialize replacement with
-				// all remaining IDs so far.
-				if repl.l == nil {
-					repl.l = make([]uint64, 0, len(p))
-					repl.l = append(repl.l, p[:i]...)
-				}
-				continue
-			}
-			// Only add to the replacement once we know we have to do it.
-			if repl.l != nil {
-				repl.l = append(repl.l, id)
-			}
-		}
-		if repl.l != nil {
-			collected[t] = repl
-		}
-	}
-
-	h.postings.mtx.RUnlock()
-
-	// Replace all postings that have changed. Append all IDs that may have
-	// been added while we switched locks.
-	h.postings.mtx.Lock()
-
-	for t, repl := range collected {
-		l := append(repl.l, h.postings.m[t][repl.idx:]...)
-
-		if len(l) > 0 {
-			h.postings.m[t] = l
-		} else {
-			delete(h.postings.m, t)
-		}
-	}
-
-	h.postings.mtx.Unlock()
+	// Remove deleted series IDs from the postings lists.
+	h.postings.Delete(deleted)
 
 	// Rebuild symbols and label value indices from what is left in the postings terms.
-	h.postings.mtx.RLock()
-
 	symbols := make(map[string]struct{})
 	values := make(map[string]stringset, len(h.values))
 
-	for t := range h.postings.m {
+	h.postings.Iter(func(t labels.Label, _ index.Postings) error {
 		symbols[t.Name] = struct{}{}
 		symbols[t.Value] = struct{}{}
 
@@ -683,9 +636,8 @@ func (h *Head) gc() {
 			values[t.Name] = ss
 		}
 		ss.set(t.Value)
-	}
-
-	h.postings.mtx.RUnlock()
+		return nil
+	})
 
 	h.symMtx.Lock()
 
@@ -765,7 +717,7 @@ func unpackChunkID(id uint64) (seriesID, chunkID uint64) {
 }
 
 // Chunk returns the chunk for the reference number.
-func (h *headChunkReader) Chunk(ref uint64) (chunks.Chunk, error) {
+func (h *headChunkReader) Chunk(ref uint64) (chunkenc.Chunk, error) {
 	sid, cid := unpackChunkID(ref)
 
 	s := h.head.series.getByID(sid)
@@ -798,12 +750,12 @@ func (h *headChunkReader) Chunk(ref uint64) (chunks.Chunk, error) {
 }
 
 type safeChunk struct {
-	chunks.Chunk
+	chunkenc.Chunk
 	s   *memSeries
 	cid int
 }
 
-func (c *safeChunk) Iterator() chunks.Iterator {
+func (c *safeChunk) Iterator() chunkenc.Iterator {
 	c.s.Lock()
 	it := c.s.iterator(c.cid)
 	c.s.Unlock()
@@ -836,7 +788,7 @@ func (h *headIndexReader) Symbols() (map[string]struct{}, error) {
 }
 
 // LabelValues returns the possible label values
-func (h *headIndexReader) LabelValues(names ...string) (StringTuples, error) {
+func (h *headIndexReader) LabelValues(names ...string) (index.StringTuples, error) {
 	if len(names) != 1 {
 		return nil, errInvalidSize
 	}
@@ -850,22 +802,22 @@ func (h *headIndexReader) LabelValues(names ...string) (StringTuples, error) {
 	}
 	sort.Strings(sl)
 
-	return &stringTuples{l: len(names), s: sl}, nil
+	return index.NewStringTuples(sl, len(names))
 }
 
 // Postings returns the postings list iterator for the label pair.
-func (h *headIndexReader) Postings(name, value string) (Postings, error) {
-	return h.head.postings.get(name, value), nil
+func (h *headIndexReader) Postings(name, value string) (index.Postings, error) {
+	return h.head.postings.Get(name, value), nil
 }
 
-func (h *headIndexReader) SortedPostings(p Postings) Postings {
+func (h *headIndexReader) SortedPostings(p index.Postings) index.Postings {
 	ep := make([]uint64, 0, 128)
 
 	for p.Next() {
 		ep = append(ep, p.At())
 	}
 	if err := p.Err(); err != nil {
-		return errPostings{err: errors.Wrap(err, "expand postings")}
+		return index.ErrPostings(errors.Wrap(err, "expand postings"))
 	}
 
 	sort.Slice(ep, func(i, j int) bool {
@@ -878,11 +830,11 @@ func (h *headIndexReader) SortedPostings(p Postings) Postings {
 		}
 		return labels.Compare(a.lset, b.lset) < 0
 	})
-	return newListPostings(ep)
+	return index.NewListPostings(ep)
 }
 
 // Series returns the series for the given reference.
-func (h *headIndexReader) Series(ref uint64, lbls *labels.Labels, chks *[]ChunkMeta) error {
+func (h *headIndexReader) Series(ref uint64, lbls *labels.Labels, chks *[]chunks.Meta) error {
 	s := h.head.series.getByID(ref)
 
 	if s == nil {
@@ -901,7 +853,7 @@ func (h *headIndexReader) Series(ref uint64, lbls *labels.Labels, chks *[]ChunkM
 		if !intervalOverlap(c.minTime, c.maxTime, h.mint, h.maxt) {
 			continue
 		}
-		*chks = append(*chks, ChunkMeta{
+		*chks = append(*chks, chunks.Meta{
 			MinTime: c.minTime,
 			MaxTime: c.maxTime,
 			Ref:     packChunkID(s.ref, uint64(s.chunkID(i))),
@@ -949,7 +901,7 @@ func (h *Head) getOrCreateWithID(id, hash uint64, lset labels.Labels) (*memSerie
 	h.metrics.series.Inc()
 	h.metrics.seriesCreated.Inc()
 
-	h.postings.add(id, lset)
+	h.postings.Add(id, lset)
 
 	h.symMtx.Lock()
 	defer h.symMtx.Unlock()
@@ -1154,7 +1106,7 @@ type memSeries struct {
 	lastValue float64
 	sampleBuf [4]sample
 
-	app chunks.Appender // Current appender for the chunk.
+	app chunkenc.Appender // Current appender for the chunk.
 }
 
 func (s *memSeries) minTime() int64 {
@@ -1167,7 +1119,7 @@ func (s *memSeries) maxTime() int64 {
 
 func (s *memSeries) cut(mint int64) *memChunk {
 	c := &memChunk{
-		chunk:   chunks.NewXORChunk(),
+		chunk:   chunkenc.NewXORChunk(),
 		minTime: mint,
 		maxTime: math.MinInt64,
 	}
@@ -1295,13 +1247,13 @@ func computeChunkEndTime(start, cur, max int64) int64 {
 	return start + (max-start)/a
 }
 
-func (s *memSeries) iterator(id int) chunks.Iterator {
+func (s *memSeries) iterator(id int) chunkenc.Iterator {
 	c := s.chunk(id)
 	// TODO(fabxc): Work around! A querier may have retrieved a pointer to a series' chunk,
 	// which got then garbage collected before it got accessed.
 	// We must ensure to not garbage collect as long as any readers still hold a reference.
 	if c == nil {
-		return chunks.NewNopIterator()
+		return chunkenc.NewNopIterator()
 	}
 
 	if id-s.firstChunkID < len(s.chunks)-1 {
@@ -1326,12 +1278,12 @@ func (s *memSeries) head() *memChunk {
 }
 
 type memChunk struct {
-	chunk            chunks.Chunk
+	chunk            chunkenc.Chunk
 	minTime, maxTime int64
 }
 
 type memSafeIterator struct {
-	chunks.Iterator
+	chunkenc.Iterator
 
 	i     int
 	total int
@@ -1356,3 +1308,27 @@ func (it *memSafeIterator) At() (int64, float64) {
 	s := it.buf[4-(it.total-it.i)]
 	return s.t, s.v
 }
+
+type stringset map[string]struct{}
+
+func (ss stringset) set(s string) {
+	ss[s] = struct{}{}
+}
+
+func (ss stringset) has(s string) bool {
+	_, ok := ss[s]
+	return ok
+}
+
+func (ss stringset) String() string {
+	return strings.Join(ss.slice(), ",")
+}
+
+func (ss stringset) slice() []string {
+	slice := make([]string, 0, len(ss))
+	for k := range ss {
+		slice = append(slice, k)
+	}
+	sort.Strings(slice)
+	return slice
+}
diff --git a/head_test.go b/head_test.go
index 4cc675442..b88f34a1e 100644
--- a/head_test.go
+++ b/head_test.go
@@ -22,7 +22,8 @@ import (
 	"testing"
 
 	"github.com/pkg/errors"
-	"github.com/prometheus/tsdb/chunks"
+	"github.com/prometheus/tsdb/chunkenc"
+	"github.com/prometheus/tsdb/index"
 	"github.com/prometheus/tsdb/labels"
 	"github.com/prometheus/tsdb/testutil"
 )
@@ -150,7 +151,7 @@ func TestHead_ReadWAL(t *testing.T) {
 	testutil.Equals(t, labels.FromStrings("a", "4"), s50.lset)
 	testutil.Equals(t, labels.FromStrings("a", "3"), s100.lset)
 
-	expandChunk := func(c chunks.Iterator) (x []sample) {
+	expandChunk := func(c chunkenc.Iterator) (x []sample) {
 		for c.Next() {
 			t, v := c.At()
 			x = append(x, sample{t: t, v: v})
@@ -210,12 +211,12 @@ func TestHead_Truncate(t *testing.T) {
 	testutil.Assert(t, h.series.getByID(s3.ref) == nil, "")
 	testutil.Assert(t, h.series.getByID(s4.ref) == nil, "")
 
-	postingsA1, _ := expandPostings(h.postings.get("a", "1"))
-	postingsA2, _ := expandPostings(h.postings.get("a", "2"))
-	postingsB1, _ := expandPostings(h.postings.get("b", "1"))
-	postingsB2, _ := expandPostings(h.postings.get("b", "2"))
-	postingsC1, _ := expandPostings(h.postings.get("c", "1"))
-	postingsAll, _ := expandPostings(h.postings.get("", ""))
+	postingsA1, _ := index.ExpandPostings(h.postings.Get("a", "1"))
+	postingsA2, _ := index.ExpandPostings(h.postings.Get("a", "2"))
+	postingsB1, _ := index.ExpandPostings(h.postings.Get("b", "1"))
+	postingsB2, _ := index.ExpandPostings(h.postings.Get("b", "2"))
+	postingsC1, _ := index.ExpandPostings(h.postings.Get("c", "1"))
+	postingsAll, _ := index.ExpandPostings(h.postings.Get("", ""))
 
 	testutil.Equals(t, []uint64{s1.ref}, postingsA1)
 	testutil.Equals(t, []uint64{s2.ref}, postingsA2)
diff --git a/index/encoding_helpers.go b/index/encoding_helpers.go
new file mode 100644
index 000000000..69e729791
--- /dev/null
+++ b/index/encoding_helpers.go
@@ -0,0 +1,179 @@
+package index
+
+import (
+	"encoding/binary"
+	"hash"
+	"hash/crc32"
+	"unsafe"
+)
+
+// enbuf is a helper type to populate a byte slice with various types.
+type encbuf struct {
+	b []byte
+	c [binary.MaxVarintLen64]byte
+}
+
+func (e *encbuf) reset()      { e.b = e.b[:0] }
+func (e *encbuf) get() []byte { return e.b }
+func (e *encbuf) len() int    { return len(e.b) }
+
+func (e *encbuf) putString(s string) { e.b = append(e.b, s...) }
+func (e *encbuf) putBytes(b []byte)  { e.b = append(e.b, b...) }
+func (e *encbuf) putByte(c byte)     { e.b = append(e.b, c) }
+
+func (e *encbuf) putBE32int(x int)      { e.putBE32(uint32(x)) }
+func (e *encbuf) putBE64int(x int)      { e.putBE64(uint64(x)) }
+func (e *encbuf) putBE64int64(x int64)  { e.putBE64(uint64(x)) }
+func (e *encbuf) putUvarint32(x uint32) { e.putUvarint64(uint64(x)) }
+func (e *encbuf) putUvarint(x int)      { e.putUvarint64(uint64(x)) }
+
+func (e *encbuf) putBE32(x uint32) {
+	binary.BigEndian.PutUint32(e.c[:], x)
+	e.b = append(e.b, e.c[:4]...)
+}
+
+func (e *encbuf) putBE64(x uint64) {
+	binary.BigEndian.PutUint64(e.c[:], x)
+	e.b = append(e.b, e.c[:8]...)
+}
+
+func (e *encbuf) putUvarint64(x uint64) {
+	n := binary.PutUvarint(e.c[:], x)
+	e.b = append(e.b, e.c[:n]...)
+}
+
+func (e *encbuf) putVarint64(x int64) {
+	n := binary.PutVarint(e.c[:], x)
+	e.b = append(e.b, e.c[:n]...)
+}
+
+// putVarintStr writes a string to the buffer prefixed by its varint length (in bytes!).
+func (e *encbuf) putUvarintStr(s string) {
+	b := *(*[]byte)(unsafe.Pointer(&s))
+	e.putUvarint(len(b))
+	e.putString(s)
+}
+
+// putHash appends a hash over the buffers current contents to the buffer.
+func (e *encbuf) putHash(h hash.Hash) {
+	h.Reset()
+	_, err := h.Write(e.b)
+	if err != nil {
+		panic(err) // The CRC32 implementation does not error
+	}
+	e.b = h.Sum(e.b)
+}
+
+// decbuf provides safe methods to extract data from a byte slice. It does all
+// necessary bounds checking and advancing of the byte slice.
+// Several datums can be extracted without checking for errors. However, before using
+// any datum, the err() method must be checked.
+type decbuf struct {
+	b []byte
+	e error
+}
+
+func (d *decbuf) uvarint() int      { return int(d.uvarint64()) }
+func (d *decbuf) uvarint32() uint32 { return uint32(d.uvarint64()) }
+func (d *decbuf) be32int() int      { return int(d.be32()) }
+func (d *decbuf) be64int64() int64  { return int64(d.be64()) }
+
+// crc32 returns a CRC32 checksum over the remaining bytes.
+func (d *decbuf) crc32() uint32 {
+	return crc32.Checksum(d.b, castagnoliTable)
+}
+
+func (d *decbuf) uvarintStr() string {
+	l := d.uvarint64()
+	if d.e != nil {
+		return ""
+	}
+	if len(d.b) < int(l) {
+		d.e = errInvalidSize
+		return ""
+	}
+	s := string(d.b[:l])
+	d.b = d.b[l:]
+	return s
+}
+
+func (d *decbuf) varint64() int64 {
+	if d.e != nil {
+		return 0
+	}
+	x, n := binary.Varint(d.b)
+	if n < 1 {
+		d.e = errInvalidSize
+		return 0
+	}
+	d.b = d.b[n:]
+	return x
+}
+
+func (d *decbuf) uvarint64() uint64 {
+	if d.e != nil {
+		return 0
+	}
+	x, n := binary.Uvarint(d.b)
+	if n < 1 {
+		d.e = errInvalidSize
+		return 0
+	}
+	d.b = d.b[n:]
+	return x
+}
+
+func (d *decbuf) be64() uint64 {
+	if d.e != nil {
+		return 0
+	}
+	if len(d.b) < 4 {
+		d.e = errInvalidSize
+		return 0
+	}
+	x := binary.BigEndian.Uint64(d.b)
+	d.b = d.b[8:]
+	return x
+}
+
+func (d *decbuf) be32() uint32 {
+	if d.e != nil {
+		return 0
+	}
+	if len(d.b) < 4 {
+		d.e = errInvalidSize
+		return 0
+	}
+	x := binary.BigEndian.Uint32(d.b)
+	d.b = d.b[4:]
+	return x
+}
+
+func (d *decbuf) byte() byte {
+	if d.e != nil {
+		return 0
+	}
+	if len(d.b) < 1 {
+		d.e = errInvalidSize
+		return 0
+	}
+	x := d.b[0]
+	d.b = d.b[1:]
+	return x
+}
+
+func (d *decbuf) decbuf(l int) decbuf {
+	if d.e != nil {
+		return decbuf{e: d.e}
+	}
+	if l > len(d.b) {
+		return decbuf{e: errInvalidSize}
+	}
+	r := decbuf{b: d.b[:l]}
+	d.b = d.b[l:]
+	return r
+}
+
+func (d *decbuf) err() error  { return d.e }
+func (d *decbuf) len() int    { return len(d.b) }
+func (d *decbuf) get() []byte { return d.b }
diff --git a/index.go b/index/index.go
similarity index 79%
rename from index.go
rename to index/index.go
index 6895c16f4..dc84386bd 100644
--- a/index.go
+++ b/index/index.go
@@ -11,13 +11,14 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-package tsdb
+package index
 
 import (
 	"bufio"
 	"encoding/binary"
 	"fmt"
 	"hash"
+	"hash/crc32"
 	"io"
 	"math"
 	"os"
@@ -26,6 +27,7 @@ import (
 	"strings"
 
 	"github.com/pkg/errors"
+	"github.com/prometheus/tsdb/chunks"
 	"github.com/prometheus/tsdb/fileutil"
 	"github.com/prometheus/tsdb/labels"
 )
@@ -35,18 +37,12 @@ const (
 	MagicIndex = 0xBAAAD700
 
 	indexFormatV1 = 1
-
-	size_unit = 4
 )
 
-const indexFilename = "index"
-
-const compactionPageBytes = minSectorSize * 64
-
 type indexWriterSeries struct {
 	labels labels.Labels
-	chunks []ChunkMeta // series file offset of chunks
-	offset uint32      // index file offset of series reference
+	chunks []chunks.Meta // series file offset of chunks
+	offset uint32        // index file offset of series reference
 }
 
 type indexWriterSeriesSlice []*indexWriterSeries
@@ -87,37 +83,24 @@ func (s indexWriterStage) String() string {
 	return "<unknown>"
 }
 
-// IndexWriter serializes the index for a block of series data.
-// The methods must be called in the order they are specified in.
-type IndexWriter interface {
-	// AddSymbols registers all string symbols that are encountered in series
-	// and other indices.
-	AddSymbols(sym map[string]struct{}) error
+// The table gets initialized with sync.Once but may still cause a race
+// with any other use of the crc32 package anywhere. Thus we initialize it
+// before.
+var castagnoliTable *crc32.Table
 
-	// AddSeries populates the index writer with a series and its offsets
-	// of chunks that the index can reference.
-	// Implementations may require series to be insert in increasing order by
-	// their labels.
-	// The reference numbers are used to resolve entries in postings lists that
-	// are added later.
-	AddSeries(ref uint64, l labels.Labels, chunks ...ChunkMeta) error
+func init() {
+	castagnoliTable = crc32.MakeTable(crc32.Castagnoli)
+}
 
-	// WriteLabelIndex serializes an index from label names to values.
-	// The passed in values chained tuples of strings of the length of names.
-	WriteLabelIndex(names []string, values []string) error
-
-	// WritePostings writes a postings list for a single label pair.
-	// The Postings here contain refs to the series that were added.
-	WritePostings(name, value string, it Postings) error
-
-	// Close writes any finalization and closes the resources associated with
-	// the underlying writer.
-	Close() error
+// newCRC32 initializes a CRC32 hash with a preconfigured polynomial, so the
+// polynomial may be easily changed in one location at a later time, if necessary.
+func newCRC32() hash.Hash32 {
+	return crc32.New(castagnoliTable)
 }
 
 // indexWriter implements the IndexWriter interface for the standard
 // serialization format.
-type indexWriter struct {
+type Writer struct {
 	f    *os.File
 	fbuf *bufio.Writer
 	pos  uint64
@@ -150,14 +133,17 @@ type indexTOC struct {
 	postingsTable     uint64
 }
 
-func newIndexWriter(dir string) (*indexWriter, error) {
+// NewWriter returns a new Writer to the given filename.
+func NewWriter(fn string) (*Writer, error) {
+	dir := filepath.Dir(fn)
+
 	df, err := fileutil.OpenDir(dir)
 	if err != nil {
 		return nil, err
 	}
 	defer df.Close() // close for flatform windows
 
-	f, err := os.OpenFile(filepath.Join(dir, indexFilename), os.O_CREATE|os.O_WRONLY, 0666)
+	f, err := os.OpenFile(fn, os.O_CREATE|os.O_WRONLY, 0666)
 	if err != nil {
 		return nil, err
 	}
@@ -165,7 +151,7 @@ func newIndexWriter(dir string) (*indexWriter, error) {
 		return nil, errors.Wrap(err, "sync dir")
 	}
 
-	iw := &indexWriter{
+	iw := &Writer{
 		f:     f,
 		fbuf:  bufio.NewWriterSize(f, 1<<22),
 		pos:   0,
@@ -187,7 +173,7 @@ func newIndexWriter(dir string) (*indexWriter, error) {
 	return iw, nil
 }
 
-func (w *indexWriter) write(bufs ...[]byte) error {
+func (w *Writer) write(bufs ...[]byte) error {
 	for _, b := range bufs {
 		n, err := w.fbuf.Write(b)
 		w.pos += uint64(n)
@@ -206,18 +192,18 @@ func (w *indexWriter) write(bufs ...[]byte) error {
 }
 
 // addPadding adds zero byte padding until the file size is a multiple size_unit.
-func (w *indexWriter) addPadding() error {
-	p := w.pos % size_unit
+func (w *Writer) addPadding(size int) error {
+	p := w.pos % uint64(size)
 	if p == 0 {
 		return nil
 	}
-	p = size_unit - p
+	p = uint64(size) - p
 	return errors.Wrap(w.write(make([]byte, p)), "add padding")
 }
 
 // ensureStage handles transitions between write stages and ensures that IndexWriter
 // methods are called in an order valid for the implementation.
-func (w *indexWriter) ensureStage(s indexWriterStage) error {
+func (w *Writer) ensureStage(s indexWriterStage) error {
 	if w.stage == s {
 		return nil
 	}
@@ -256,7 +242,7 @@ func (w *indexWriter) ensureStage(s indexWriterStage) error {
 	return nil
 }
 
-func (w *indexWriter) writeMeta() error {
+func (w *Writer) writeMeta() error {
 	w.buf1.reset()
 	w.buf1.putBE32(MagicIndex)
 	w.buf1.putByte(indexFormatV1)
@@ -264,7 +250,7 @@ func (w *indexWriter) writeMeta() error {
 	return w.write(w.buf1.get())
 }
 
-func (w *indexWriter) AddSeries(ref uint64, lset labels.Labels, chunks ...ChunkMeta) error {
+func (w *Writer) AddSeries(ref uint64, lset labels.Labels, chunks ...chunks.Meta) error {
 	if err := w.ensureStage(idxStageSeries); err != nil {
 		return err
 	}
@@ -328,7 +314,7 @@ func (w *indexWriter) AddSeries(ref uint64, lset labels.Labels, chunks ...ChunkM
 	return nil
 }
 
-func (w *indexWriter) AddSymbols(sym map[string]struct{}) error {
+func (w *Writer) AddSymbols(sym map[string]struct{}) error {
 	if err := w.ensureStage(idxStageSymbols); err != nil {
 		return err
 	}
@@ -361,7 +347,7 @@ func (w *indexWriter) AddSymbols(sym map[string]struct{}) error {
 	return errors.Wrap(err, "write symbols")
 }
 
-func (w *indexWriter) WriteLabelIndex(names []string, values []string) error {
+func (w *Writer) WriteLabelIndex(names []string, values []string) error {
 	if len(values)%len(names) != 0 {
 		return errors.Errorf("invalid value list length %d for %d names", len(values), len(names))
 	}
@@ -369,14 +355,14 @@ func (w *indexWriter) WriteLabelIndex(names []string, values []string) error {
 		return errors.Wrap(err, "ensure stage")
 	}
 
-	valt, err := newStringTuples(values, len(names))
+	valt, err := NewStringTuples(values, len(names))
 	if err != nil {
 		return err
 	}
 	sort.Sort(valt)
 
 	// Align beginning to 4 bytes for more efficient index list scans.
-	if err := w.addPadding(); err != nil {
+	if err := w.addPadding(4); err != nil {
 		return err
 	}
 
@@ -407,7 +393,7 @@ func (w *indexWriter) WriteLabelIndex(names []string, values []string) error {
 }
 
 // writeOffsetTable writes a sequence of readable hash entries.
-func (w *indexWriter) writeOffsetTable(entries []hashEntry) error {
+func (w *Writer) writeOffsetTable(entries []hashEntry) error {
 	w.buf2.reset()
 	w.buf2.putBE32int(len(entries))
 
@@ -428,7 +414,7 @@ func (w *indexWriter) writeOffsetTable(entries []hashEntry) error {
 
 const indexTOCLen = 6*8 + 4
 
-func (w *indexWriter) writeTOC() error {
+func (w *Writer) writeTOC() error {
 	w.buf1.reset()
 
 	w.buf1.putBE64(w.toc.symbols)
@@ -443,13 +429,13 @@ func (w *indexWriter) writeTOC() error {
 	return w.write(w.buf1.get())
 }
 
-func (w *indexWriter) WritePostings(name, value string, it Postings) error {
+func (w *Writer) WritePostings(name, value string, it Postings) error {
 	if err := w.ensureStage(idxStagePostings); err != nil {
 		return errors.Wrap(err, "ensure stage")
 	}
 
 	// Align beginning to 4 bytes for more efficient postings list scans.
-	if err := w.addPadding(); err != nil {
+	if err := w.addPadding(4); err != nil {
 		return err
 	}
 
@@ -506,7 +492,7 @@ type hashEntry struct {
 	offset uint64
 }
 
-func (w *indexWriter) Close() error {
+func (w *Writer) Close() error {
 	if err := w.ensureStage(idxStageDone); err != nil {
 		return err
 	}
@@ -519,37 +505,6 @@ func (w *indexWriter) Close() error {
 	return w.f.Close()
 }
 
-// IndexReader provides reading access of serialized index data.
-type IndexReader interface {
-	// Symbols returns a set of string symbols that may occur in series' labels
-	// and indices.
-	Symbols() (map[string]struct{}, error)
-
-	// LabelValues returns the possible label values
-	LabelValues(names ...string) (StringTuples, error)
-
-	// Postings returns the postings list iterator for the label pair.
-	// The Postings here contain the offsets to the series inside the index.
-	// Found IDs are not strictly required to point to a valid Series, e.g. during
-	// background garbage collections.
-	Postings(name, value string) (Postings, error)
-
-	// SortedPostings returns a postings list that is reordered to be sorted
-	// by the label set of the underlying series.
-	SortedPostings(Postings) Postings
-
-	// Series populates the given labels and chunk metas for the series identified
-	// by the reference.
-	// Returns ErrNotFound if the ref does not resolve to a known series.
-	Series(ref uint64, lset *labels.Labels, chks *[]ChunkMeta) error
-
-	// LabelIndices returns the label pairs for which indices exist.
-	LabelIndices() ([][]string, error)
-
-	// Close released the underlying resources of the reader.
-	Close() error
-}
-
 // StringTuples provides access to a sorted list of string tuples.
 type StringTuples interface {
 	// Total number of tuples in the list.
@@ -558,7 +513,7 @@ type StringTuples interface {
 	At(i int) ([]string, error)
 }
 
-type indexReader struct {
+type Reader struct {
 	// The underlying byte slice holding the encoded series data.
 	b   ByteSlice
 	toc indexTOC
@@ -605,22 +560,22 @@ func (b realByteSlice) Sub(start, end int) ByteSlice {
 	return b[start:end]
 }
 
-// NewIndexReader returns a new IndexReader on the given byte slice.
-func NewIndexReader(b ByteSlice) (IndexReader, error) {
-	return newIndexReader(b, nil)
+// NewReader returns a new IndexReader on the given byte slice.
+func NewReader(b ByteSlice) (*Reader, error) {
+	return newReader(b, nil)
 }
 
-// NewFileIndexReader returns a new index reader against the given index file.
-func NewFileIndexReader(path string) (IndexReader, error) {
-	f, err := openMmapFile(path)
+// NewFileReader returns a new index reader against the given index file.
+func NewFileReader(path string) (*Reader, error) {
+	f, err := fileutil.OpenMmapFile(path)
 	if err != nil {
 		return nil, err
 	}
-	return newIndexReader(realByteSlice(f.b), f)
+	return newReader(realByteSlice(f.Bytes()), f)
 }
 
-func newIndexReader(b ByteSlice, c io.Closer) (*indexReader, error) {
-	r := &indexReader{
+func newReader(b ByteSlice, c io.Closer) (*Reader, error) {
+	r := &Reader{
 		b:       b,
 		c:       c,
 		symbols: map[uint32]string{},
@@ -650,7 +605,7 @@ func newIndexReader(b ByteSlice, c io.Closer) (*indexReader, error) {
 	return r, errors.Wrap(err, "read postings table")
 }
 
-func (r *indexReader) readTOC() error {
+func (r *Reader) readTOC() error {
 	if r.b.Len() < indexTOCLen {
 		return errInvalidSize
 	}
@@ -676,7 +631,7 @@ func (r *indexReader) readTOC() error {
 // decbufAt returns a new decoding buffer. It expects the first 4 bytes
 // after offset to hold the big endian encoded content length, followed by the contents and the expected
 // checksum.
-func (r *indexReader) decbufAt(off int) decbuf {
+func (r *Reader) decbufAt(off int) decbuf {
 	if r.b.Len() < off+4 {
 		return decbuf{e: errInvalidSize}
 	}
@@ -700,7 +655,7 @@ func (r *indexReader) decbufAt(off int) decbuf {
 // decbufUvarintAt returns a new decoding buffer. It expects the first bytes
 // after offset to hold the uvarint-encoded buffers length, followed by the contents and the expected
 // checksum.
-func (r *indexReader) decbufUvarintAt(off int) decbuf {
+func (r *Reader) decbufUvarintAt(off int) decbuf {
 	// We never have to access this method at the far end of the byte slice. Thus just checking
 	// against the MaxVarintLen32 is sufficient.
 	if r.b.Len() < off+binary.MaxVarintLen32 {
@@ -730,7 +685,7 @@ func (r *indexReader) decbufUvarintAt(off int) decbuf {
 // readSymbols reads the symbol table fully into memory and allocates proper strings for them.
 // Strings backed by the mmap'd memory would cause memory faults if applications keep using them
 // after the reader is closed.
-func (r *indexReader) readSymbols(off int) error {
+func (r *Reader) readSymbols(off int) error {
 	if off == 0 {
 		return nil
 	}
@@ -754,7 +709,7 @@ func (r *indexReader) readSymbols(off int) error {
 
 // readOffsetTable reads an offset table at the given position and returns a map
 // with the key strings concatenated by the 0xff unicode non-character.
-func (r *indexReader) readOffsetTable(off uint64) (map[string]uint32, error) {
+func (r *Reader) readOffsetTable(off uint64) (map[string]uint32, error) {
 	const sep = "\xff"
 
 	d := r.decbufAt(int(off))
@@ -776,11 +731,11 @@ func (r *indexReader) readOffsetTable(off uint64) (map[string]uint32, error) {
 	return res, d.err()
 }
 
-func (r *indexReader) Close() error {
+func (r *Reader) Close() error {
 	return r.c.Close()
 }
 
-func (r *indexReader) lookupSymbol(o uint32) (string, error) {
+func (r *Reader) lookupSymbol(o uint32) (string, error) {
 	s, ok := r.symbols[o]
 	if !ok {
 		return "", errors.Errorf("unknown symbol offset %d", o)
@@ -788,7 +743,7 @@ func (r *indexReader) lookupSymbol(o uint32) (string, error) {
 	return s, nil
 }
 
-func (r *indexReader) Symbols() (map[string]struct{}, error) {
+func (r *Reader) Symbols() (map[string]struct{}, error) {
 	res := make(map[string]struct{}, len(r.symbols))
 
 	for _, s := range r.symbols {
@@ -797,7 +752,7 @@ func (r *indexReader) Symbols() (map[string]struct{}, error) {
 	return res, nil
 }
 
-func (r *indexReader) LabelValues(names ...string) (StringTuples, error) {
+func (r *Reader) LabelValues(names ...string) (StringTuples, error) {
 	const sep = "\xff"
 
 	key := strings.Join(names, sep)
@@ -830,7 +785,7 @@ type emptyStringTuples struct{}
 func (emptyStringTuples) At(i int) ([]string, error) { return nil, nil }
 func (emptyStringTuples) Len() int                   { return 0 }
 
-func (r *indexReader) LabelIndices() ([][]string, error) {
+func (r *Reader) LabelIndices() ([][]string, error) {
 	const sep = "\xff"
 
 	res := [][]string{}
@@ -841,7 +796,7 @@ func (r *indexReader) LabelIndices() ([][]string, error) {
 	return res, nil
 }
 
-func (r *indexReader) Series(ref uint64, lbls *labels.Labels, chks *[]ChunkMeta) error {
+func (r *Reader) Series(ref uint64, lbls *labels.Labels, chks *[]chunks.Meta) error {
 	d := r.decbufUvarintAt(int(ref))
 
 	*lbls = (*lbls)[:0]
@@ -880,7 +835,7 @@ func (r *indexReader) Series(ref uint64, lbls *labels.Labels, chks *[]ChunkMeta)
 	maxt := int64(d.uvarint64()) + t0
 	ref0 := int64(d.uvarint64())
 
-	*chks = append(*chks, ChunkMeta{
+	*chks = append(*chks, chunks.Meta{
 		Ref:     uint64(ref0),
 		MinTime: t0,
 		MaxTime: maxt,
@@ -898,7 +853,7 @@ func (r *indexReader) Series(ref uint64, lbls *labels.Labels, chks *[]ChunkMeta)
 			return errors.Wrapf(d.err(), "read meta for chunk %d", i)
 		}
 
-		*chks = append(*chks, ChunkMeta{
+		*chks = append(*chks, chunks.Meta{
 			Ref:     uint64(ref0),
 			MinTime: mint,
 			MaxTime: maxt,
@@ -907,7 +862,7 @@ func (r *indexReader) Series(ref uint64, lbls *labels.Labels, chks *[]ChunkMeta)
 	return d.err()
 }
 
-func (r *indexReader) Postings(name, value string) (Postings, error) {
+func (r *Reader) Postings(name, value string) (Postings, error) {
 	const sep = "\xff"
 	key := strings.Join([]string{name, value}, sep)
 
@@ -921,7 +876,7 @@ func (r *indexReader) Postings(name, value string) (Postings, error) {
 	return newBigEndianPostings(d.get()), errors.Wrap(d.err(), "get postings bytes")
 }
 
-func (r *indexReader) SortedPostings(p Postings) Postings {
+func (r *Reader) SortedPostings(p Postings) Postings {
 	return p
 }
 
@@ -930,7 +885,7 @@ type stringTuples struct {
 	s []string // flattened tuple entries
 }
 
-func newStringTuples(s []string, l int) (*stringTuples, error) {
+func NewStringTuples(s []string, l int) (*stringTuples, error) {
 	if len(s)%l != 0 {
 		return nil, errors.Wrap(errInvalidSize, "string tuple list")
 	}
diff --git a/index_test.go b/index/index_test.go
similarity index 77%
rename from index_test.go
rename to index/index_test.go
index 4c272f897..f4ba813f3 100644
--- a/index_test.go
+++ b/index/index_test.go
@@ -11,7 +11,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-package tsdb
+package index
 
 import (
 	"io/ioutil"
@@ -20,8 +20,12 @@ import (
 	"path/filepath"
 	"sort"
 	"testing"
+	"unsafe"
 
 	"github.com/pkg/errors"
+	promlabels "github.com/prometheus/prometheus/pkg/labels"
+	"github.com/prometheus/prometheus/pkg/textparse"
+	"github.com/prometheus/tsdb/chunkenc"
 	"github.com/prometheus/tsdb/chunks"
 	"github.com/prometheus/tsdb/labels"
 	"github.com/prometheus/tsdb/testutil"
@@ -29,13 +33,13 @@ import (
 
 type series struct {
 	l      labels.Labels
-	chunks []ChunkMeta
+	chunks []chunks.Meta
 }
 
 type mockIndex struct {
 	series     map[uint64]series
 	labelIndex map[string][]string
-	postings   *memPostings
+	postings   map[labels.Label][]uint64
 	symbols    map[string]struct{}
 }
 
@@ -43,11 +47,9 @@ func newMockIndex() mockIndex {
 	ix := mockIndex{
 		series:     make(map[uint64]series),
 		labelIndex: make(map[string][]string),
-		postings:   newMemPostings(),
+		postings:   make(map[labels.Label][]uint64),
 		symbols:    make(map[string]struct{}),
 	}
-	ix.postings.ensureOrder()
-
 	return ix
 }
 
@@ -55,7 +57,7 @@ func (m mockIndex) Symbols() (map[string]struct{}, error) {
 	return m.symbols, nil
 }
 
-func (m mockIndex) AddSeries(ref uint64, l labels.Labels, chunks ...ChunkMeta) error {
+func (m mockIndex) AddSeries(ref uint64, l labels.Labels, chunks ...chunks.Meta) error {
 	if _, ok := m.series[ref]; ok {
 		return errors.Errorf("series with reference %d already added", ref)
 	}
@@ -80,23 +82,22 @@ func (m mockIndex) WriteLabelIndex(names []string, values []string) error {
 	if len(names) != 1 {
 		return errors.New("composite indexes not supported yet")
 	}
-
 	sort.Strings(values)
 	m.labelIndex[names[0]] = values
 	return nil
 }
 
 func (m mockIndex) WritePostings(name, value string, it Postings) error {
-	if _, ok := m.postings.m[labels.Label{name, value}]; ok {
-		return errors.Errorf("postings for %s=%q already added", name, value)
+	l := labels.Label{Name: name, Value: value}
+	if _, ok := m.postings[l]; ok {
+		return errors.Errorf("postings for %s already added", l)
 	}
-	ep, err := expandPostings(it)
+	ep, err := ExpandPostings(it)
 	if err != nil {
 		return err
 	}
-	m.postings.m[labels.Label{name, value}] = ep
-
-	return it.Err()
+	m.postings[l] = ep
+	return nil
 }
 
 func (m mockIndex) Close() error {
@@ -109,29 +110,30 @@ func (m mockIndex) LabelValues(names ...string) (StringTuples, error) {
 		return nil, errors.New("composite indexes not supported yet")
 	}
 
-	return newStringTuples(m.labelIndex[names[0]], 1)
+	return NewStringTuples(m.labelIndex[names[0]], 1)
 }
 
 func (m mockIndex) Postings(name, value string) (Postings, error) {
-	return m.postings.get(name, value), nil
+	l := labels.Label{Name: name, Value: value}
+	return NewListPostings(m.postings[l]), nil
 }
 
 func (m mockIndex) SortedPostings(p Postings) Postings {
-	ep, err := expandPostings(p)
+	ep, err := ExpandPostings(p)
 	if err != nil {
-		return errPostings{err: errors.Wrap(err, "expand postings")}
+		return ErrPostings(errors.Wrap(err, "expand postings"))
 	}
 
 	sort.Slice(ep, func(i, j int) bool {
 		return labels.Compare(m.series[ep[i]].l, m.series[ep[j]].l) < 0
 	})
-	return newListPostings(ep)
+	return NewListPostings(ep)
 }
 
-func (m mockIndex) Series(ref uint64, lset *labels.Labels, chks *[]ChunkMeta) error {
+func (m mockIndex) Series(ref uint64, lset *labels.Labels, chks *[]chunks.Meta) error {
 	s, ok := m.series[ref]
 	if !ok {
-		return ErrNotFound
+		return errors.New("not found")
 	}
 	*lset = append((*lset)[:0], s.l...)
 	*chks = append((*chks)[:0], s.chunks...)
@@ -154,22 +156,24 @@ func TestIndexRW_Create_Open(t *testing.T) {
 	testutil.Ok(t, err)
 	defer os.RemoveAll(dir)
 
+	fn := filepath.Join(dir, "index")
+
 	// An empty index must still result in a readable file.
-	iw, err := newIndexWriter(dir)
+	iw, err := NewWriter(fn)
 	testutil.Ok(t, err)
 	testutil.Ok(t, iw.Close())
 
-	ir, err := NewFileIndexReader(filepath.Join(dir, "index"))
+	ir, err := NewFileReader(fn)
 	testutil.Ok(t, err)
 	testutil.Ok(t, ir.Close())
 
 	// Modify magic header must cause open to fail.
-	f, err := os.OpenFile(filepath.Join(dir, "index"), os.O_WRONLY, 0666)
+	f, err := os.OpenFile(fn, os.O_WRONLY, 0666)
 	testutil.Ok(t, err)
 	_, err = f.WriteAt([]byte{0, 0}, 0)
 	testutil.Ok(t, err)
 
-	_, err = NewFileIndexReader(dir)
+	_, err = NewFileReader(dir)
 	testutil.NotOk(t, err)
 }
 
@@ -178,7 +182,9 @@ func TestIndexRW_Postings(t *testing.T) {
 	testutil.Ok(t, err)
 	defer os.RemoveAll(dir)
 
-	iw, err := newIndexWriter(dir)
+	fn := filepath.Join(dir, "index")
+
+	iw, err := NewWriter(fn)
 	testutil.Ok(t, err)
 
 	series := []labels.Labels{
@@ -210,14 +216,14 @@ func TestIndexRW_Postings(t *testing.T) {
 
 	testutil.Ok(t, iw.Close())
 
-	ir, err := NewFileIndexReader(filepath.Join(dir, "index"))
+	ir, err := NewFileReader(fn)
 	testutil.Ok(t, err)
 
 	p, err := ir.Postings("a", "1")
 	testutil.Ok(t, err)
 
 	var l labels.Labels
-	var c []ChunkMeta
+	var c []chunks.Meta
 
 	for i := 0; p.Next(); i++ {
 		err := ir.Series(p.At(), &l, &c)
@@ -254,14 +260,14 @@ func TestPersistence_index_e2e(t *testing.T) {
 
 	// Generate ChunkMetas for every label set.
 	for i, lset := range lbls {
-		var metas []ChunkMeta
+		var metas []chunks.Meta
 
 		for j := 0; j <= (i % 20); j++ {
-			metas = append(metas, ChunkMeta{
+			metas = append(metas, chunks.Meta{
 				MinTime: int64(j * 10000),
 				MaxTime: int64((j + 1) * 10000),
 				Ref:     rand.Uint64(),
-				Chunk:   chunks.NewXORChunk(),
+				Chunk:   chunkenc.NewXORChunk(),
 			})
 		}
 		input = append(input, &indexWriterSeries{
@@ -270,17 +276,16 @@ func TestPersistence_index_e2e(t *testing.T) {
 		})
 	}
 
-	iw, err := newIndexWriter(dir)
+	iw, err := NewWriter(filepath.Join(dir, "index"))
 	testutil.Ok(t, err)
 
 	testutil.Ok(t, iw.AddSymbols(symbols))
 
 	// Population procedure as done by compaction.
 	var (
-		postings = newMemPostings()
-		values   = map[string]stringset{}
+		postings = NewMemPostings()
+		values   = map[string]map[string]struct{}{}
 	)
-	postings.ensureOrder()
 
 	mi := newMockIndex()
 
@@ -292,12 +297,12 @@ func TestPersistence_index_e2e(t *testing.T) {
 		for _, l := range s.labels {
 			valset, ok := values[l.Name]
 			if !ok {
-				valset = stringset{}
+				valset = map[string]struct{}{}
 				values[l.Name] = valset
 			}
-			valset.set(l.Value)
+			valset[l.Value] = struct{}{}
 		}
-		postings.add(uint64(i), s.labels)
+		postings.Add(uint64(i), s.labels)
 		i++
 	}
 
@@ -328,14 +333,14 @@ func TestPersistence_index_e2e(t *testing.T) {
 	ir, err := NewFileIndexReader(filepath.Join(dir, "index"))
 	testutil.Ok(t, err)
 
-	for p := range mi.postings.m {
+	for p := range mi.postings {
 		gotp, err := ir.Postings(p.Name, p.Value)
 		testutil.Ok(t, err)
 
 		expp, err := mi.Postings(p.Name, p.Value)
 
 		var lset, explset labels.Labels
-		var chks, expchks []ChunkMeta
+		var chks, expchks []chunks.Meta
 
 		for gotp.Next() {
 			testutil.Assert(t, expp.Next() == true, "")
@@ -374,3 +379,41 @@ func TestPersistence_index_e2e(t *testing.T) {
 
 	testutil.Ok(t, ir.Close())
 }
+
+func readPrometheusLabels(fn string, n int) ([]labels.Labels, error) {
+	f, err := os.Open(fn)
+	if err != nil {
+		return nil, err
+	}
+	defer f.Close()
+
+	b, err := ioutil.ReadAll(f)
+	if err != nil {
+		return nil, err
+	}
+
+	p := textparse.New(b)
+	i := 0
+	var mets []labels.Labels
+	hashes := map[uint64]struct{}{}
+
+	for p.Next() && i < n {
+		m := make(labels.Labels, 0, 10)
+		p.Metric((*promlabels.Labels)(unsafe.Pointer(&m)))
+
+		h := m.Hash()
+		if _, ok := hashes[h]; ok {
+			continue
+		}
+		mets = append(mets, m)
+		hashes[h] = struct{}{}
+		i++
+	}
+	if err := p.Err(); err != nil {
+		return nil, err
+	}
+	if i != n {
+		return mets, errors.Errorf("requested %d metrics but found %d", n, i)
+	}
+	return mets, nil
+}
diff --git a/postings.go b/index/postings.go
similarity index 78%
rename from postings.go
rename to index/postings.go
index 200917e13..bc63b4dac 100644
--- a/postings.go
+++ b/index/postings.go
@@ -11,7 +11,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-package tsdb
+package index
 
 import (
 	"encoding/binary"
@@ -23,35 +23,35 @@ import (
 	"github.com/prometheus/tsdb/labels"
 )
 
-// memPostings holds postings list for series ID per label pair. They may be written
+// MemPostings holds postings list for series ID per label pair. They may be written
 // to out of order.
 // ensureOrder() must be called once before any reads are done. This allows for quick
 // unordered batch fills on startup.
-type memPostings struct {
+type MemPostings struct {
 	mtx     sync.RWMutex
 	m       map[labels.Label][]uint64
 	ordered bool
 }
 
-// newMemPoistings returns a memPostings that's ready for reads and writes.
-func newMemPostings() *memPostings {
-	return &memPostings{
+// NewMemPostings returns a memPostings that's ready for reads and writes.
+func NewMemPostings() *MemPostings {
+	return &MemPostings{
 		m:       make(map[labels.Label][]uint64, 512),
 		ordered: true,
 	}
 }
 
-// newUnorderedMemPostings returns a memPostings that is not safe to be read from
+// NewUnorderedMemPostings returns a memPostings that is not safe to be read from
 // until ensureOrder was called once.
-func newUnorderedMemPostings() *memPostings {
-	return &memPostings{
+func NewUnorderedMemPostings() *MemPostings {
+	return &MemPostings{
 		m:       make(map[labels.Label][]uint64, 512),
 		ordered: false,
 	}
 }
 
-// sortedKeys returns a list of sorted label keys of the postings.
-func (p *memPostings) sortedKeys() []labels.Label {
+// SortedKeys returns a list of sorted label keys of the postings.
+func (p *MemPostings) SortedKeys() []labels.Label {
 	p.mtx.RLock()
 	keys := make([]labels.Label, 0, len(p.m))
 
@@ -69,23 +69,28 @@ func (p *memPostings) sortedKeys() []labels.Label {
 	return keys
 }
 
-// Postings returns an iterator over the postings list for s.
-func (p *memPostings) get(name, value string) Postings {
+// Get returns a postings list for the given label pair.
+func (p *MemPostings) Get(name, value string) Postings {
 	p.mtx.RLock()
 	l := p.m[labels.Label{Name: name, Value: value}]
 	p.mtx.RUnlock()
 
 	if l == nil {
-		return emptyPostings
+		return EmptyPostings()
 	}
 	return newListPostings(l)
 }
 
+// All returns a postings list over all documents ever added.
+func (p *MemPostings) All() Postings {
+	return p.Get(allPostingsKey.Name, allPostingsKey.Value)
+}
+
 var allPostingsKey = labels.Label{}
 
-// ensurePostings ensures that all postings lists are sorted. After it returns all further
+// EnsureOrder ensures that all postings lists are sorted. After it returns all further
 // calls to add and addFor will insert new IDs in a sorted manner.
-func (p *memPostings) ensureOrder() {
+func (p *MemPostings) EnsureOrder() {
 	p.mtx.Lock()
 	defer p.mtx.Unlock()
 
@@ -117,9 +122,61 @@ func (p *memPostings) ensureOrder() {
 	p.ordered = true
 }
 
-// add adds a document to the index. The caller has to ensure that no
-// term argument appears twice.
-func (p *memPostings) add(id uint64, lset labels.Labels) {
+// Delete removes all ids in the given map from the postings lists.
+func (p *MemPostings) Delete(deleted map[uint64]struct{}) {
+	var keys []labels.Label
+
+	p.mtx.RLock()
+	for l := range p.m {
+		keys = append(keys, l)
+	}
+	p.mtx.RUnlock()
+
+	for _, l := range keys {
+		p.mtx.Lock()
+
+		found := false
+		for _, id := range p.m[l] {
+			if _, ok := deleted[id]; ok {
+				found = true
+				break
+			}
+		}
+		if !found {
+			p.mtx.Unlock()
+			continue
+		}
+		repl := make([]uint64, 0, len(p.m[l]))
+
+		for _, id := range p.m[l] {
+			if _, ok := deleted[id]; !ok {
+				repl = append(repl, id)
+			}
+		}
+		if len(repl) > 0 {
+			p.m[l] = repl
+		} else {
+			delete(p.m, l)
+		}
+		p.mtx.Unlock()
+	}
+}
+
+// Iter calls f for each postings list. It aborts if f returns an error and returns it.
+func (p *MemPostings) Iter(f func(labels.Label, Postings) error) error {
+	p.mtx.RLock()
+	defer p.mtx.RUnlock()
+
+	for l, p := range p.m {
+		if err := f(l, newListPostings(p)); err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+// Add a label set to the postings index.
+func (p *MemPostings) Add(id uint64, lset labels.Labels) {
 	p.mtx.Lock()
 
 	for _, l := range lset {
@@ -130,7 +187,7 @@ func (p *memPostings) add(id uint64, lset labels.Labels) {
 	p.mtx.Unlock()
 }
 
-func (p *memPostings) addFor(id uint64, l labels.Label) {
+func (p *MemPostings) addFor(id uint64, l labels.Label) {
 	list := append(p.m[l], id)
 	p.m[l] = list
 
@@ -149,7 +206,8 @@ func (p *memPostings) addFor(id uint64, l labels.Label) {
 	}
 }
 
-func expandPostings(p Postings) (res []uint64, err error) {
+// ExpandPostings returns the postings expanded as a slice.
+func ExpandPostings(p Postings) (res []uint64, err error) {
 	for p.Next() {
 		res = append(res, p.At())
 	}
@@ -189,6 +247,11 @@ func EmptyPostings() Postings {
 	return emptyPostings
 }
 
+// ErrPostings returns new postings that immediately error.
+func ErrPostings(err error) Postings {
+	return errPostings{err}
+}
+
 // Intersect returns a new postings list over the intersection of the
 // input postings.
 func Intersect(its ...Postings) Postings {
@@ -340,6 +403,12 @@ func (it *mergedPostings) Err() error {
 	return it.b.Err()
 }
 
+// Without returns a new postings list that contains all elements from the full list that
+// are not in the drop list
+func Without(full, drop Postings) Postings {
+	return newRemovedPostings(full, drop)
+}
+
 type removedPostings struct {
 	full, remove Postings
 
@@ -420,6 +489,10 @@ type listPostings struct {
 	cur  uint64
 }
 
+func NewListPostings(list []uint64) Postings {
+	return newListPostings(list)
+}
+
 func newListPostings(list []uint64) *listPostings {
 	return &listPostings{list: list}
 }
@@ -508,27 +581,3 @@ func (it *bigEndianPostings) Seek(x uint64) bool {
 func (it *bigEndianPostings) Err() error {
 	return nil
 }
-
-type stringset map[string]struct{}
-
-func (ss stringset) set(s string) {
-	ss[s] = struct{}{}
-}
-
-func (ss stringset) has(s string) bool {
-	_, ok := ss[s]
-	return ok
-}
-
-func (ss stringset) String() string {
-	return strings.Join(ss.slice(), ",")
-}
-
-func (ss stringset) slice() []string {
-	slice := make([]string, 0, len(ss))
-	for k := range ss {
-		slice = append(slice, k)
-	}
-	sort.Strings(slice)
-	return slice
-}
diff --git a/postings_test.go b/index/postings_test.go
similarity index 95%
rename from postings_test.go
rename to index/postings_test.go
index abaf1b054..68553e357 100644
--- a/postings_test.go
+++ b/index/postings_test.go
@@ -11,7 +11,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-package tsdb
+package index
 
 import (
 	"encoding/binary"
@@ -25,7 +25,7 @@ import (
 )
 
 func TestMemPostings_addFor(t *testing.T) {
-	p := newMemPostings()
+	p := NewMemPostings()
 	p.m[allPostingsKey] = []uint64{1, 2, 3, 4, 6, 7, 8}
 
 	p.addFor(5, allPostingsKey)
@@ -34,7 +34,7 @@ func TestMemPostings_addFor(t *testing.T) {
 }
 
 func TestMemPostings_ensureOrder(t *testing.T) {
-	p := newUnorderedMemPostings()
+	p := NewUnorderedMemPostings()
 
 	for i := 0; i < 100; i++ {
 		l := make([]uint64, 100)
@@ -46,7 +46,7 @@ func TestMemPostings_ensureOrder(t *testing.T) {
 		p.m[labels.Label{"a", v}] = l
 	}
 
-	p.ensureOrder()
+	p.EnsureOrder()
 
 	for _, l := range p.m {
 		ok := sort.SliceIsSorted(l, func(i, j int) bool {
@@ -100,7 +100,7 @@ func TestIntersect(t *testing.T) {
 		a := newListPostings(c.a)
 		b := newListPostings(c.b)
 
-		res, err := expandPostings(Intersect(a, b))
+		res, err := ExpandPostings(Intersect(a, b))
 		testutil.Ok(t, err)
 		testutil.Equals(t, c.res, res)
 	}
@@ -140,7 +140,7 @@ func TestMultiIntersect(t *testing.T) {
 			ps = append(ps, newListPostings(postings))
 		}
 
-		res, err := expandPostings(Intersect(ps...))
+		res, err := ExpandPostings(Intersect(ps...))
 
 		testutil.Ok(t, err)
 		testutil.Equals(t, c.res, res)
@@ -174,7 +174,7 @@ func BenchmarkIntersect(t *testing.B) {
 	t.ResetTimer()
 
 	for i := 0; i < t.N; i++ {
-		if _, err := expandPostings(Intersect(i1, i2, i3, i4)); err != nil {
+		if _, err := ExpandPostings(Intersect(i1, i2, i3, i4)); err != nil {
 			t.Fatal(err)
 		}
 	}
@@ -198,7 +198,7 @@ func TestMultiMerge(t *testing.T) {
 		i2 := newListPostings(c.b)
 		i3 := newListPostings(c.c)
 
-		res, err := expandPostings(Merge(i1, i2, i3))
+		res, err := ExpandPostings(Merge(i1, i2, i3))
 		testutil.Ok(t, err)
 		testutil.Equals(t, c.res, res)
 	}
@@ -230,7 +230,7 @@ func TestMergedPostings(t *testing.T) {
 		a := newListPostings(c.a)
 		b := newListPostings(c.b)
 
-		res, err := expandPostings(newMergedPostings(a, b))
+		res, err := ExpandPostings(newMergedPostings(a, b))
 		testutil.Ok(t, err)
 		testutil.Equals(t, c.res, res)
 	}
@@ -290,7 +290,7 @@ func TestMergedPostingsSeek(t *testing.T) {
 		// After Seek(), At() should be called.
 		if c.success {
 			start := p.At()
-			lst, err := expandPostings(p)
+			lst, err := ExpandPostings(p)
 			testutil.Ok(t, err)
 
 			lst = append([]uint64{start}, lst...)
@@ -347,7 +347,7 @@ func TestRemovedPostings(t *testing.T) {
 		a := newListPostings(c.a)
 		b := newListPostings(c.b)
 
-		res, err := expandPostings(newRemovedPostings(a, b))
+		res, err := ExpandPostings(newRemovedPostings(a, b))
 		testutil.Ok(t, err)
 		testutil.Equals(t, c.res, res)
 	}
@@ -431,7 +431,7 @@ func TestRemovedPostingsSeek(t *testing.T) {
 		// After Seek(), At() should be called.
 		if c.success {
 			start := p.At()
-			lst, err := expandPostings(p)
+			lst, err := ExpandPostings(p)
 			testutil.Ok(t, err)
 
 			lst = append([]uint64{start}, lst...)
@@ -527,7 +527,7 @@ func TestIntersectWithMerge(t *testing.T) {
 	)
 
 	p := Intersect(a, b)
-	res, err := expandPostings(p)
+	res, err := ExpandPostings(p)
 
 	testutil.Ok(t, err)
 	testutil.Equals(t, []uint64{30}, res)
diff --git a/querier.go b/querier.go
index 145258b8e..48d7ba9c2 100644
--- a/querier.go
+++ b/querier.go
@@ -19,7 +19,9 @@ import (
 	"strings"
 
 	"github.com/pkg/errors"
+	"github.com/prometheus/tsdb/chunkenc"
 	"github.com/prometheus/tsdb/chunks"
+	"github.com/prometheus/tsdb/index"
 	"github.com/prometheus/tsdb/labels"
 )
 
@@ -202,18 +204,17 @@ func (q *blockQuerier) Close() error {
 // PostingsForMatchers assembles a single postings iterator against the index reader
 // based on the given matchers. It returns a list of label names that must be manually
 // checked to not exist in series the postings list points to.
-func PostingsForMatchers(index IndexReader, ms ...labels.Matcher) (Postings, error) {
-	var (
-		its []Postings
-	)
+func PostingsForMatchers(ix IndexReader, ms ...labels.Matcher) (index.Postings, error) {
+	var its []index.Postings
+
 	for _, m := range ms {
-		it, err := postingsForMatcher(index, m)
+		it, err := postingsForMatcher(ix, m)
 		if err != nil {
 			return nil, err
 		}
 		its = append(its, it)
 	}
-	return index.SortedPostings(Intersect(its...)), nil
+	return ix.SortedPostings(index.Intersect(its...)), nil
 }
 
 // tuplesByPrefix uses binary search to find prefix matches within ts.
@@ -247,24 +248,24 @@ func tuplesByPrefix(m *labels.PrefixMatcher, ts StringTuples) ([]string, error)
 	return matches, nil
 }
 
-func postingsForMatcher(index IndexReader, m labels.Matcher) (Postings, error) {
+func postingsForMatcher(ix IndexReader, m labels.Matcher) (index.Postings, error) {
 	// If the matcher selects an empty value, it selects all the series which dont
 	// have the label name set too. See: https://github.com/prometheus/prometheus/issues/3575
 	// and https://github.com/prometheus/prometheus/pull/3578#issuecomment-351653555
 	if m.Matches("") {
-		return postingsForUnsetLabelMatcher(index, m)
+		return postingsForUnsetLabelMatcher(ix, m)
 	}
 
 	// Fast-path for equal matching.
 	if em, ok := m.(*labels.EqualMatcher); ok {
-		it, err := index.Postings(em.Name(), em.Value())
+		it, err := ix.Postings(em.Name(), em.Value())
 		if err != nil {
 			return nil, err
 		}
 		return it, nil
 	}
 
-	tpls, err := index.LabelValues(m.Name())
+	tpls, err := ix.LabelValues(m.Name())
 	if err != nil {
 		return nil, err
 	}
@@ -289,24 +290,24 @@ func postingsForMatcher(index IndexReader, m labels.Matcher) (Postings, error) {
 	}
 
 	if len(res) == 0 {
-		return EmptyPostings(), nil
+		return index.EmptyPostings(), nil
 	}
 
-	var rit []Postings
+	var rit []index.Postings
 
 	for _, v := range res {
-		it, err := index.Postings(m.Name(), v)
+		it, err := ix.Postings(m.Name(), v)
 		if err != nil {
 			return nil, err
 		}
 		rit = append(rit, it)
 	}
 
-	return Merge(rit...), nil
+	return index.Merge(rit...), nil
 }
 
-func postingsForUnsetLabelMatcher(index IndexReader, m labels.Matcher) (Postings, error) {
-	tpls, err := index.LabelValues(m.Name())
+func postingsForUnsetLabelMatcher(ix IndexReader, m labels.Matcher) (index.Postings, error) {
+	tpls, err := ix.LabelValues(m.Name())
 	if err != nil {
 		return nil, err
 	}
@@ -323,23 +324,22 @@ func postingsForUnsetLabelMatcher(index IndexReader, m labels.Matcher) (Postings
 		}
 	}
 
-	var rit []Postings
+	var rit []index.Postings
 	for _, v := range res {
-		it, err := index.Postings(m.Name(), v)
+		it, err := ix.Postings(m.Name(), v)
 		if err != nil {
 			return nil, err
 		}
 
 		rit = append(rit, it)
 	}
-	mrit := Merge(rit...)
 
-	allPostings, err := index.Postings(allPostingsKey.Name, allPostingsKey.Value)
+	allPostings, err := ix.Postings("", "")
 	if err != nil {
 		return nil, err
 	}
 
-	return newRemovedPostings(allPostings, mrit), nil
+	return index.Without(allPostings, index.Merge(rit...)), nil
 }
 
 func mergeStrings(a, b []string) []string {
@@ -458,19 +458,19 @@ func (s *mergedSeriesSet) Next() bool {
 // actual series itself.
 type ChunkSeriesSet interface {
 	Next() bool
-	At() (labels.Labels, []ChunkMeta, Intervals)
+	At() (labels.Labels, []chunks.Meta, Intervals)
 	Err() error
 }
 
 // baseChunkSeries loads the label set and chunk references for a postings
 // list from an index. It filters out series that have labels set that should be unset.
 type baseChunkSeries struct {
-	p          Postings
+	p          index.Postings
 	index      IndexReader
 	tombstones TombstoneReader
 
 	lset      labels.Labels
-	chks      []ChunkMeta
+	chks      []chunks.Meta
 	intervals Intervals
 	err       error
 }
@@ -492,7 +492,7 @@ func LookupChunkSeries(ir IndexReader, tr TombstoneReader, ms ...labels.Matcher)
 	}, nil
 }
 
-func (s *baseChunkSeries) At() (labels.Labels, []ChunkMeta, Intervals) {
+func (s *baseChunkSeries) At() (labels.Labels, []chunks.Meta, Intervals) {
 	return s.lset, s.chks, s.intervals
 }
 
@@ -500,14 +500,14 @@ func (s *baseChunkSeries) Err() error { return s.err }
 
 func (s *baseChunkSeries) Next() bool {
 	var (
-		lset   labels.Labels
-		chunks []ChunkMeta
-		err    error
+		lset     labels.Labels
+		chkMetas []chunks.Meta
+		err      error
 	)
 
 	for s.p.Next() {
 		ref := s.p.At()
-		if err := s.index.Series(ref, &lset, &chunks); err != nil {
+		if err := s.index.Series(ref, &lset, &chkMetas); err != nil {
 			// Postings may be stale. Skip if no underlying series exists.
 			if errors.Cause(err) == ErrNotFound {
 				continue
@@ -517,7 +517,7 @@ func (s *baseChunkSeries) Next() bool {
 		}
 
 		s.lset = lset
-		s.chks = chunks
+		s.chks = chkMetas
 		s.intervals, err = s.tombstones.Get(s.p.At())
 		if err != nil {
 			s.err = errors.Wrap(err, "get tombstones")
@@ -526,7 +526,7 @@ func (s *baseChunkSeries) Next() bool {
 
 		if len(s.intervals) > 0 {
 			// Only those chunks that are not entirely deleted.
-			chks := make([]ChunkMeta, 0, len(s.chks))
+			chks := make([]chunks.Meta, 0, len(s.chks))
 			for _, chk := range s.chks {
 				if !(Interval{chk.MinTime, chk.MaxTime}.isSubrange(s.intervals)) {
 					chks = append(chks, chk)
@@ -553,12 +553,12 @@ type populatedChunkSeries struct {
 	mint, maxt int64
 
 	err       error
-	chks      []ChunkMeta
+	chks      []chunks.Meta
 	lset      labels.Labels
 	intervals Intervals
 }
 
-func (s *populatedChunkSeries) At() (labels.Labels, []ChunkMeta, Intervals) {
+func (s *populatedChunkSeries) At() (labels.Labels, []chunks.Meta, Intervals) {
 	return s.lset, s.chks, s.intervals
 }
 
@@ -651,7 +651,7 @@ func (s *blockSeriesSet) Err() error { return s.err }
 // time series data.
 type chunkSeries struct {
 	labels labels.Labels
-	chunks []ChunkMeta // in-order chunk refs
+	chunks []chunks.Meta // in-order chunk refs
 
 	mint, maxt int64
 
@@ -754,17 +754,17 @@ func (it *chainedSeriesIterator) Err() error {
 // chunkSeriesIterator implements a series iterator on top
 // of a list of time-sorted, non-overlapping chunks.
 type chunkSeriesIterator struct {
-	chunks []ChunkMeta
+	chunks []chunks.Meta
 
 	i   int
-	cur chunks.Iterator
+	cur chunkenc.Iterator
 
 	maxt, mint int64
 
 	intervals Intervals
 }
 
-func newChunkSeriesIterator(cs []ChunkMeta, dranges Intervals, mint, maxt int64) *chunkSeriesIterator {
+func newChunkSeriesIterator(cs []chunks.Meta, dranges Intervals, mint, maxt int64) *chunkSeriesIterator {
 	it := cs[0].Chunk.Iterator()
 
 	if len(dranges) > 0 {
@@ -853,6 +853,46 @@ func (it *chunkSeriesIterator) Err() error {
 	return it.cur.Err()
 }
 
+// deletedIterator wraps an Iterator and makes sure any deleted metrics are not
+// returned.
+type deletedIterator struct {
+	it chunkenc.Iterator
+
+	intervals Intervals
+}
+
+func (it *deletedIterator) At() (int64, float64) {
+	return it.it.At()
+}
+
+func (it *deletedIterator) Next() bool {
+Outer:
+	for it.it.Next() {
+		ts, _ := it.it.At()
+
+		for _, tr := range it.intervals {
+			if tr.inBounds(ts) {
+				continue Outer
+			}
+
+			if ts > tr.Maxt {
+				it.intervals = it.intervals[1:]
+				continue
+			}
+
+			return true
+		}
+
+		return true
+	}
+
+	return false
+}
+
+func (it *deletedIterator) Err() error {
+	return it.it.Err()
+}
+
 type mockSeriesSet struct {
 	next   func() bool
 	series func() Series
diff --git a/querier_test.go b/querier_test.go
index fee93d9af..2eb34f450 100644
--- a/querier_test.go
+++ b/querier_test.go
@@ -20,7 +20,10 @@ import (
 	"sort"
 	"testing"
 
+	"github.com/pkg/errors"
+	"github.com/prometheus/tsdb/chunkenc"
 	"github.com/prometheus/tsdb/chunks"
+	"github.com/prometheus/tsdb/index"
 	"github.com/prometheus/tsdb/labels"
 	"github.com/prometheus/tsdb/testutil"
 )
@@ -228,25 +231,25 @@ func createIdxChkReaders(tc []struct {
 		return labels.Compare(labels.FromMap(tc[i].lset), labels.FromMap(tc[i].lset)) < 0
 	})
 
-	postings := newMemPostings()
-	chkReader := mockChunkReader(make(map[uint64]chunks.Chunk))
+	postings := index.NewMemPostings()
+	chkReader := mockChunkReader(make(map[uint64]chunkenc.Chunk))
 	lblIdx := make(map[string]stringset)
 	mi := newMockIndex()
 
 	for i, s := range tc {
 		i = i + 1 // 0 is not a valid posting.
-		metas := make([]ChunkMeta, 0, len(s.chunks))
+		metas := make([]chunks.Meta, 0, len(s.chunks))
 		for _, chk := range s.chunks {
 			// Collisions can be there, but for tests, its fine.
 			ref := rand.Uint64()
 
-			metas = append(metas, ChunkMeta{
+			metas = append(metas, chunks.Meta{
 				MinTime: chk[0].t,
 				MaxTime: chk[len(chk)-1].t,
 				Ref:     ref,
 			})
 
-			chunk := chunks.NewXORChunk()
+			chunk := chunkenc.NewXORChunk()
 			app, _ := chunk.Appender()
 			for _, smpl := range chk {
 				app.Append(smpl.t, smpl.v)
@@ -257,7 +260,7 @@ func createIdxChkReaders(tc []struct {
 		ls := labels.FromMap(s.lset)
 		mi.AddSeries(uint64(i), ls, metas...)
 
-		postings.add(uint64(i), ls)
+		postings.Add(uint64(i), ls)
 
 		for _, l := range ls {
 			vs, present := lblIdx[l.Name]
@@ -273,9 +276,9 @@ func createIdxChkReaders(tc []struct {
 		mi.WriteLabelIndex([]string{l}, vs.slice())
 	}
 
-	for l := range postings.m {
-		mi.WritePostings(l.Name, l.Value, postings.get(l.Name, l.Value))
-	}
+	postings.Iter(func(l labels.Label, p index.Postings) error {
+		return mi.WritePostings(l.Name, l.Value, p)
+	})
 
 	return mi, chkReader
 }
@@ -660,7 +663,7 @@ Outer:
 func TestBaseChunkSeries(t *testing.T) {
 	type refdSeries struct {
 		lset   labels.Labels
-		chunks []ChunkMeta
+		chunks []chunks.Meta
 
 		ref uint64
 	}
@@ -676,7 +679,7 @@ func TestBaseChunkSeries(t *testing.T) {
 			series: []refdSeries{
 				{
 					lset: labels.New([]labels.Label{{"a", "a"}}...),
-					chunks: []ChunkMeta{
+					chunks: []chunks.Meta{
 						{Ref: 29}, {Ref: 45}, {Ref: 245}, {Ref: 123}, {Ref: 4232}, {Ref: 5344},
 						{Ref: 121},
 					},
@@ -684,19 +687,19 @@ func TestBaseChunkSeries(t *testing.T) {
 				},
 				{
 					lset: labels.New([]labels.Label{{"a", "a"}, {"b", "b"}}...),
-					chunks: []ChunkMeta{
+					chunks: []chunks.Meta{
 						{Ref: 82}, {Ref: 23}, {Ref: 234}, {Ref: 65}, {Ref: 26},
 					},
 					ref: 10,
 				},
 				{
 					lset:   labels.New([]labels.Label{{"b", "c"}}...),
-					chunks: []ChunkMeta{{Ref: 8282}},
+					chunks: []chunks.Meta{{Ref: 8282}},
 					ref:    1,
 				},
 				{
 					lset: labels.New([]labels.Label{{"b", "b"}}...),
-					chunks: []ChunkMeta{
+					chunks: []chunks.Meta{
 						{Ref: 829}, {Ref: 239}, {Ref: 2349}, {Ref: 659}, {Ref: 269},
 					},
 					ref: 108,
@@ -709,14 +712,14 @@ func TestBaseChunkSeries(t *testing.T) {
 			series: []refdSeries{
 				{
 					lset: labels.New([]labels.Label{{"a", "a"}, {"b", "b"}}...),
-					chunks: []ChunkMeta{
+					chunks: []chunks.Meta{
 						{Ref: 82}, {Ref: 23}, {Ref: 234}, {Ref: 65}, {Ref: 26},
 					},
 					ref: 10,
 				},
 				{
 					lset:   labels.New([]labels.Label{{"b", "c"}}...),
-					chunks: []ChunkMeta{{Ref: 8282}},
+					chunks: []chunks.Meta{{Ref: 8282}},
 					ref:    3,
 				},
 			},
@@ -732,7 +735,7 @@ func TestBaseChunkSeries(t *testing.T) {
 		}
 
 		bcs := &baseChunkSeries{
-			p:          newListPostings(tc.postings),
+			p:          index.NewListPostings(tc.postings),
 			index:      mi,
 			tombstones: EmptyTombstoneReader(),
 		}
@@ -763,20 +766,20 @@ type itSeries struct {
 func (s itSeries) Iterator() SeriesIterator { return s.si }
 func (s itSeries) Labels() labels.Labels    { return labels.Labels{} }
 
-func chunkFromSamples(s []sample) ChunkMeta {
+func chunkFromSamples(s []sample) chunks.Meta {
 	mint, maxt := int64(0), int64(0)
 
 	if len(s) > 0 {
 		mint, maxt = s[0].t, s[len(s)-1].t
 	}
 
-	c := chunks.NewXORChunk()
+	c := chunkenc.NewXORChunk()
 	ca, _ := c.Appender()
 
 	for _, s := range s {
 		ca.Append(s.t, s.v)
 	}
-	return ChunkMeta{
+	return chunks.Meta{
 		MinTime: mint,
 		MaxTime: maxt,
 		Chunk:   c,
@@ -941,7 +944,7 @@ func TestSeriesIterator(t *testing.T) {
 
 	t.Run("Chunk", func(t *testing.T) {
 		for _, tc := range itcases {
-			chkMetas := []ChunkMeta{
+			chkMetas := []chunks.Meta{
 				chunkFromSamples(tc.a),
 				chunkFromSamples(tc.b),
 				chunkFromSamples(tc.c),
@@ -1012,7 +1015,7 @@ func TestSeriesIterator(t *testing.T) {
 			seekcases2 := append(seekcases, extra...)
 
 			for _, tc := range seekcases2 {
-				chkMetas := []ChunkMeta{
+				chkMetas := []chunks.Meta{
 					chunkFromSamples(tc.a),
 					chunkFromSamples(tc.b),
 					chunkFromSamples(tc.c),
@@ -1099,7 +1102,7 @@ func TestSeriesIterator(t *testing.T) {
 
 // Regression for: https://github.com/prometheus/tsdb/pull/97
 func TestChunkSeriesIterator_DoubleSeek(t *testing.T) {
-	chkMetas := []ChunkMeta{
+	chkMetas := []chunks.Meta{
 		chunkFromSamples([]sample{}),
 		chunkFromSamples([]sample{{1, 1}, {2, 2}, {3, 3}}),
 		chunkFromSamples([]sample{{4, 4}, {5, 5}}),
@@ -1116,7 +1119,7 @@ func TestChunkSeriesIterator_DoubleSeek(t *testing.T) {
 // Regression when seeked chunks were still found via binary search and we always
 // skipped to the end when seeking a value in the current chunk.
 func TestChunkSeriesIterator_SeekInCurrentChunk(t *testing.T) {
-	metas := []ChunkMeta{
+	metas := []chunks.Meta{
 		chunkFromSamples([]sample{}),
 		chunkFromSamples([]sample{{1, 2}, {3, 4}, {5, 6}, {7, 8}}),
 		chunkFromSamples([]sample{}),
@@ -1138,7 +1141,7 @@ func TestChunkSeriesIterator_SeekInCurrentChunk(t *testing.T) {
 // Regression when calling Next() with a time bounded to fit within two samples.
 // Seek gets called and advances beyond the max time, which was just accepted as a valid sample.
 func TestChunkSeriesIterator_NextWithMinTime(t *testing.T) {
-	metas := []ChunkMeta{
+	metas := []chunks.Meta{
 		chunkFromSamples([]sample{{1, 6}, {5, 6}, {7, 8}}),
 	}
 
@@ -1148,7 +1151,7 @@ func TestChunkSeriesIterator_NextWithMinTime(t *testing.T) {
 
 func TestPopulatedCSReturnsValidChunkSlice(t *testing.T) {
 	lbls := []labels.Labels{labels.New(labels.Label{"a", "b"})}
-	chunkMetas := [][]ChunkMeta{
+	chunkMetas := [][]chunks.Meta{
 		{
 			{MinTime: 1, MaxTime: 2, Ref: 1},
 			{MinTime: 3, MaxTime: 4, Ref: 2},
@@ -1157,10 +1160,10 @@ func TestPopulatedCSReturnsValidChunkSlice(t *testing.T) {
 	}
 
 	cr := mockChunkReader(
-		map[uint64]chunks.Chunk{
-			1: chunks.NewXORChunk(),
-			2: chunks.NewXORChunk(),
-			3: chunks.NewXORChunk(),
+		map[uint64]chunkenc.Chunk{
+			1: chunkenc.NewXORChunk(),
+			2: chunkenc.NewXORChunk(),
+			3: chunkenc.NewXORChunk(),
 		},
 	)
 
@@ -1180,7 +1183,7 @@ func TestPopulatedCSReturnsValidChunkSlice(t *testing.T) {
 	testutil.Assert(t, p.Next() == false, "")
 
 	// Test the case where 1 chunk could cause an unpopulated chunk to be returned.
-	chunkMetas = [][]ChunkMeta{
+	chunkMetas = [][]chunks.Meta{
 		{
 			{MinTime: 1, MaxTime: 2, Ref: 1},
 		},
@@ -1200,7 +1203,7 @@ func TestPopulatedCSReturnsValidChunkSlice(t *testing.T) {
 
 type mockChunkSeriesSet struct {
 	l  []labels.Labels
-	cm [][]ChunkMeta
+	cm [][]chunks.Meta
 
 	i int
 }
@@ -1213,7 +1216,7 @@ func (m *mockChunkSeriesSet) Next() bool {
 	return m.i < len(m.l)
 }
 
-func (m *mockChunkSeriesSet) At() (labels.Labels, []ChunkMeta, Intervals) {
+func (m *mockChunkSeriesSet) At() (labels.Labels, []chunks.Meta, Intervals) {
 	return m.l[m.i], m.cm[m.i], nil
 }
 
@@ -1279,3 +1282,198 @@ func BenchmarkMergedSeriesSet(b *testing.B) {
 		}
 	}
 }
+
+type mockChunkReader map[uint64]chunkenc.Chunk
+
+func (cr mockChunkReader) Chunk(id uint64) (chunkenc.Chunk, error) {
+	chk, ok := cr[id]
+	if ok {
+		return chk, nil
+	}
+
+	return nil, errors.New("Chunk with ref not found")
+}
+
+func (cr mockChunkReader) Close() error {
+	return nil
+}
+
+func TestDeletedIterator(t *testing.T) {
+	chk := chunks.NewXORChunk()
+	app, err := chk.Appender()
+	testutil.Ok(t, err)
+	// Insert random stuff from (0, 1000).
+	act := make([]sample, 1000)
+	for i := 0; i < 1000; i++ {
+		act[i].t = int64(i)
+		act[i].v = rand.Float64()
+		app.Append(act[i].t, act[i].v)
+	}
+
+	cases := []struct {
+		r Intervals
+	}{
+		{r: Intervals{{1, 20}}},
+		{r: Intervals{{1, 10}, {12, 20}, {21, 23}, {25, 30}}},
+		{r: Intervals{{1, 10}, {12, 20}, {20, 30}}},
+		{r: Intervals{{1, 10}, {12, 23}, {25, 30}}},
+		{r: Intervals{{1, 23}, {12, 20}, {25, 30}}},
+		{r: Intervals{{1, 23}, {12, 20}, {25, 3000}}},
+		{r: Intervals{{0, 2000}}},
+		{r: Intervals{{500, 2000}}},
+		{r: Intervals{{0, 200}}},
+		{r: Intervals{{1000, 20000}}},
+	}
+
+	for _, c := range cases {
+		i := int64(-1)
+		it := &deletedIterator{it: chk.Iterator(), intervals: c.r[:]}
+		ranges := c.r[:]
+		for it.Next() {
+			i++
+			for _, tr := range ranges {
+				if tr.inBounds(i) {
+					i = tr.Maxt + 1
+					ranges = ranges[1:]
+				}
+			}
+
+			testutil.Assert(t, i < 1000 == true, "")
+
+			ts, v := it.At()
+			testutil.Equals(t, act[i].t, ts)
+			testutil.Equals(t, act[i].v, v)
+		}
+		// There has been an extra call to Next().
+		i++
+		for _, tr := range ranges {
+			if tr.inBounds(i) {
+				i = tr.Maxt + 1
+				ranges = ranges[1:]
+			}
+		}
+
+		testutil.Assert(t, i < 1000 == false, "")
+		testutil.Ok(t, it.Err())
+	}
+}
+
+type series struct {
+	l      labels.Labels
+	chunks []chunks.Meta
+}
+
+type mockIndex struct {
+	series     map[uint64]series
+	labelIndex map[string][]string
+	postings   map[labels.Label][]uint64
+	symbols    map[string]struct{}
+}
+
+func newMockIndex() mockIndex {
+	ix := mockIndex{
+		series:     make(map[uint64]series),
+		labelIndex: make(map[string][]string),
+		postings:   make(map[labels.Label][]uint64),
+		symbols:    make(map[string]struct{}),
+	}
+	return ix
+}
+
+func (m mockIndex) Symbols() (map[string]struct{}, error) {
+	return m.symbols, nil
+}
+
+func (m mockIndex) AddSeries(ref uint64, l labels.Labels, chunks ...chunks.Meta) error {
+	if _, ok := m.series[ref]; ok {
+		return errors.Errorf("series with reference %d already added", ref)
+	}
+	for _, lbl := range l {
+		m.symbols[lbl.Name] = struct{}{}
+		m.symbols[lbl.Value] = struct{}{}
+	}
+
+	s := series{l: l}
+	// Actual chunk data is not stored in the index.
+	for _, c := range chunks {
+		c.Chunk = nil
+		s.chunks = append(s.chunks, c)
+	}
+	m.series[ref] = s
+
+	return nil
+}
+
+func (m mockIndex) WriteLabelIndex(names []string, values []string) error {
+	// TODO support composite indexes
+	if len(names) != 1 {
+		return errors.New("composite indexes not supported yet")
+	}
+	sort.Strings(values)
+	m.labelIndex[names[0]] = values
+	return nil
+}
+
+func (m mockIndex) WritePostings(name, value string, it index.Postings) error {
+	l := labels.Label{Name: name, Value: value}
+	if _, ok := m.postings[l]; ok {
+		return errors.Errorf("postings for %s already added", l)
+	}
+	ep, err := index.ExpandPostings(it)
+	if err != nil {
+		return err
+	}
+	m.postings[l] = ep
+	return nil
+}
+
+func (m mockIndex) Close() error {
+	return nil
+}
+
+func (m mockIndex) LabelValues(names ...string) (index.StringTuples, error) {
+	// TODO support composite indexes
+	if len(names) != 1 {
+		return nil, errors.New("composite indexes not supported yet")
+	}
+
+	return index.NewStringTuples(m.labelIndex[names[0]], 1)
+}
+
+func (m mockIndex) Postings(name, value string) (index.Postings, error) {
+	l := labels.Label{Name: name, Value: value}
+	return index.NewListPostings(m.postings[l]), nil
+}
+
+func (m mockIndex) SortedPostings(p index.Postings) index.Postings {
+	ep, err := index.ExpandPostings(p)
+	if err != nil {
+		return index.ErrPostings(errors.Wrap(err, "expand postings"))
+	}
+
+	sort.Slice(ep, func(i, j int) bool {
+		return labels.Compare(m.series[ep[i]].l, m.series[ep[j]].l) < 0
+	})
+	return index.NewListPostings(ep)
+}
+
+func (m mockIndex) Series(ref uint64, lset *labels.Labels, chks *[]chunks.Meta) error {
+	s, ok := m.series[ref]
+	if !ok {
+		return ErrNotFound
+	}
+	*lset = append((*lset)[:0], s.l...)
+	*chks = append((*chks)[:0], s.chunks...)
+
+	return nil
+}
+
+func (m mockIndex) LabelIndices() ([][]string, error) {
+	res := make([][]string, 0, len(m.labelIndex))
+
+	for k := range m.labelIndex {
+		res = append(res, []string{k})
+	}
+
+	return res, nil
+}