Save work converting between Builder and Labels.
Also expose ProcessBuilder, so callers can supply a Builder.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
This lets relabelling work on a `Builder` rather than converting to and
from `Labels` on every rule.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
The difference is modest, but we've used `slices.Sort` in lots of other
places so why not here.
name old time/op new time/op delta
Builder 1.04µs ± 3% 0.95µs ± 3% -8.27% (p=0.008 n=5+5)
name old alloc/op new alloc/op delta
Builder 312B ± 0% 288B ± 0% -7.69% (p=0.008 n=5+5)
name old allocs/op new allocs/op delta
Builder 2.00 ± 0% 1.00 ± 0% -50.00% (p=0.008 n=5+5)
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
This makes the buffer the correct size for the common case that labels
have only been added. It will be too large for the case that labels are
changed, but the current buffer resize logic in `appendLabelTo` doubles
the buffer, so a small over-estimate is better.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
This change removes restrictions to allow adding exemplars
to all time series. It also contains some improvements in test values
so that it is easier to track what is tested.
The advantage of doing this is having a little less error-prone tests:
"yy" is not really descriptive but "counter-test" can give people
a better idea about what is tested so it is harder to make mistakes.
Closes gh-11982
Signed-off-by: Jonatan Ivanov <jonatan.ivanov@gmail.com>
Previous code was effectively doing BigEndian.Uint64, so call that and save time.
An md5.Sum result is always 16 bytes. The first 8 are not used in the result, just as before.
Signed-off-by: Renning Bruns <ren@renmail.net>
These benchmarks were testing things related to what Prometheus does, but not testing actual Prometheus code.
Moved the label-copying benchmark into the labels package.
This commit adds an alternate implementation for `labels.Labels`, behind
a build tag `stringlabels`.
Instead of storing label names and values as individual strings, they
are all concatenated into one string in this format:
[len][name0][len][value0][len][name1][len][value1]...
The lengths are varint encoded so usually a single byte.
The previous `[]string` had 24 bytes of overhead for the slice and 16
for each label name and value; this one has 16 bytes overhead plus 1
for each name and value.
In `ScratchBuilder.Overwrite` and `Labels.Hash` we use an unsafe
conversion from string to byte slice. `Overwrite` is explicitly unsafe,
but for `Hash` this is a pure performance hack.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
Parsing errors in the Prometheus HTTP format parser are very hard to
investigate since they only approximately indicate what is going wrong
in the parser and don't provide any information about the incorrect
input. As such it is very hard to tell what is wrong in the format
exposed by the application.
Signed-off-by: Damien Grisonnet <dgrisonn@redhat.com>
This commit adds a new 'keep_firing_for' field to Prometheus alerting
rules. The 'resolve_delay' field specifies the minimum amount of time
that an alert should remain firing, even if the expression does not
return any results.
This feature was discussed at a previous dev summit, and it was
determined that a feature like this would be useful in order to allow
the expression time to stabilize and prevent confusing resolved messages
from being propagated through Alertmanager.
This approach is simpler than having two PromQL queries, as was
sometimes discussed, and it should be easy to implement.
This commit does not include tests for the 'resolve_delay' field. This
is intentional, as the purpose of this commit is to gather comments on
the proposed design of the 'resolve_delay' field before implementing
tests. Once the design of the 'resolve_delay' field has been finalized,
a follow-up commit will be submitted with tests."
See https://github.com/prometheus/prometheus/issues/11570
Signed-off-by: Julien Pivotto <roidelapluie@o11y.eu>
If a (float or integer) histogram is a gauge histogram, set the
CounterResetHint accordingly. (The default value is fine for the
normal counter histograms.)
Signed-off-by: beorn7 <beorn@grafana.com>
With this commit, the parser stops to see a gauge histogram (whether
native or conventional) as an unexpected metric type. It ingests it
normally, it even sets the `GaugeHistogram` type in the metadata (as
it has already done for a conventional gauge histogram scraped using
OpenMetrics), but it otherwise treats it as a normal counter-like
histogram.
Once #11783 is merged, though, it should be very easy to utilize the
type information.
Signed-off-by: beorn7 <beorn@grafana.com>
So far, the parser hasn't validated that the type is valid in the
`Next()` call. Later, in the `Series()` call, however, it assumes that
we will only see valid types and therefore panics with `encountered
unexpected metric type, this is a bug`.
This commit fixes said bug by adding validation to the `Next()` call.
Signed-off-by: beorn7 <beorn@grafana.com>
In some cases, the Prometheus HTTP format parser was not returning the
right token in the error output which made debugging impossible.
Signed-off-by: Damien Grisonnet <dgrisonn@redhat.com>
Extends Appender.AppendHistogram function to accept the FloatHistogram. TSDB supports appending, querying, WAL replay, for this new type of histogram.
Signed-off-by: Marc Tudurí <marctc@protonmail.com>
Signed-off-by: Ganesh Vernekar <ganeshvern@gmail.com>
Co-authored-by: Ganesh Vernekar <ganeshvern@gmail.com>
Instead of passing in a `ScratchBuilder` and `Labels`, just pass the
builder and the caller can extract labels from it. In many cases the
caller didn't use the Labels value anyway.
Now in `Labels.ScratchBuilder` we need a slightly different API: one
to assign what will be the result, instead of overwriting some other
`Labels`. This is safer and easier to reason about.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
Without changing the definition of `labels.Labels`, add methods which
enable code using it to work without knowledge of the internals.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
For performance reasons we may use a different implementation of Hash()
in future, so note this so callers can be warned.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
The bucket receiving math.MaxFloat64 observations now has
math.MaxFloat64 as upper bound, while the bucket after it (the last
possible bucket) has +Inf.
This also adds a test for getBound and moves the getBound code to
generic.go (where it should have been in the first place).
Signed-off-by: beorn7 <beorn@grafana.com>
* histogram: Simplify iterators
We don't really need currLower and currUpper and can calculate it when
needed (as already done for the floatBucketIterator). The calculation
is cheap, while keeping those extra variables around costs RAM
(potentially a lot with many iterators).
* histogram: Convert Bucket/FloatBucket to one generic type
* histogram: Move some bucket iterator code into generic base iterator
* histogram: Remove cumulative iterator for FloatHistogram
We added it in the past for completeness (Histogram has one), but it
has never been used. Plus, even the cumulative iterator for Histogram
is only there for test reasons.
We can always add it back, and then maybe even using generics.
Signed-off-by: beorn7 <beorn@grafana.com>
Where the code was multiplying bytes by number of operations, this
resulted in absurdly high throughput numbers.
Also, in `BenchmarkParse()`, don't run the `expfmt` case twice.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
And use the new method to call to compact Histograms during
parsing. This happens for both `Histogram` and `FloatHistogram`. In
this way, if targets decide to optimize the exposition size by merging
spans with empty buckets in between, we still get a normalized
results. It will also normalize away any valid but weird
representations like empty spans, spans with offset zero, and empty
buckets at the start or end of a span.
The implementation seemed easy at first as it just turns the
`compactBuckets` helper into a generic function (which now got its own
file). However, the integer Histograms have delta buckets instead of
absolute buckets, which had to be treated specially in the generic
`compactBuckets` function. To make sure it works, I have added plenty
of explicit tests for `Histogram` in addition to the `FloatHistogram`
tests.
I have also updated the doc comment for the `Compact` method.
Based on the insights now expressed in the doc comment, compacting
with a maxEmptyBuckets > 0 is rarely useful. Therefore, this commit
also sets the value to 0 in the two cases we were using 3 so far. We
might still want to reconsider, so I don't want to remove the
maxEmptyBuckets parameter right now.
Signed-off-by: beorn7 <beorn@grafana.com>
Replacing code which assumes the internal structure of `Labels`.
Add a convenience function `EmptyLabels()` which is more efficient than
calling `New()`.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* Update go to 1.19, set min version to 1.18
Signed-off-by: Julien Pivotto <roidelapluie@o11y.eu>
* Update golangci-lint
Signed-off-by: Julien Pivotto <roidelapluie@o11y.eu>
Signed-off-by: Julien Pivotto <roidelapluie@o11y.eu>
* model/relabel: Add benchmark
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* model/relabel: re-use Builder across relabels
Saves memory allocations.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* labels.Builder: allow re-use of result slice
This reduces memory allocations where the caller has a suitable slice available.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* model/relabel: re-use source values slice
To reduce memory allocations.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* Unwind one change causing test failures
Restore original behaviour in PopulateLabels, where we must not overwrite the input set.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* relabel: simplify values optimisation
Use a stack-based array for up to 16 source labels, which will be the
vast majority of cases.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* lint
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* Append metadata to the WAL
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Remove extra whitespace; Reword some docstrings and comments
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Use RLock() for hasNewMetadata check
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Use single byte for metric type in RefMetadata
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Update proposed WAL format for single-byte type metadata
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Implementa MetadataAppender interface for the Agent
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Address first round of review comments
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Amend description of metadata in wal.md
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Correct key used to retrieve metadata from cache
When we're setting metadata entries in the scrapeCace, we're using the
p.Help(), p.Unit(), p.Type() helpers, which retrieve the series name and
use it as the cache key. When checking for cache entries though, we used
p.Series() as the key, which included the metric name _with_ its labels.
That meant that we were never actually hitting the cache. We're fixing
this by utiling the __name__ internal label for correctly getting the
cache entries after they've been set by setHelp(), setType() or
setUnit().
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Put feature behind a feature flag
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Fix AppendMetadata docstring
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Reorder WAL format document
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Change error message of AppendMetadata; Fix access of s.meta in AppendMetadata
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Reuse temporary buffer in Metadata encoder
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Only keep latest metadata for each refID during checkpointing
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Fix test that's referencing decoding metadata
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Avoid creating metadata block if no new metadata are present
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Add tests for corrupt metadata block and relevant record type
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Fix CR comments
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Extract logic about changing metadata in an anonymous function
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Implement new proposed WAL format and amend relevant tests
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Use 'const' for metadata field names
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Apply metadata to head memSeries in Commit, not in AppendMetadata
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Add docstring and rename extracted helper in scrape.go
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Add tests for tsdb-related cases
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Fix linter issues vol1
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Fix linter issues vol2
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Fix Windows test by closing WAL reader files
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Use switch instead of two if statements in metadata decoding
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Fix review comments around TestMetadata* tests
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Add code for replaying WAL; test correctness of in-memory data after a replay
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Remove scrape-loop related code from PR
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Address first round of comments
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Simplify tests by sorting slices before comparison
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Fix test to use separate transactions
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Empty out buffer and record slices after encoding latest metadata
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Fix linting issue
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Update calculation for DroppedMetadata metric
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Rename MetadataAppender interface and AppendMetadata method to MetadataUpdater/UpdateMetadata
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Reuse buffer when encoding latest metadata for each series
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Fix review comments; Check all returned error values using two helpers
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Simplify use of helpers
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Satisfy linter
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
I'd like to unwrap errors returned from rulefmt but both Error and WrappedError types are missing Unwrap() method.
Signed-off-by: Łukasz Mierzwa <l.mierzwa@gmail.com>
* Drop extra string held in relabel.Regexp struct
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Use slice operations instead of TrimPrefix/TrimSuffix; Override String() method
Signed-off-by: Paschalis Tsilias <paschalist0@gmail.com>
* Labels: create signature with/without labels
Instead of creating a new Labels slice then converting to signature,
go directly to the signature and save time.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* Labels: refactor Builder tests
Have one test with a range of cases, and have them check the final
output rather than checking the internal structure of the Builder.
Also add a couple of cases where the value is "", which should be
interpreted as 'delete'.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* Labels: add 'Keep' function to Builder
This lets us replace `Labels.WithLabels` with the more general `Builder`.
In `engine.resultMetric()` we can call `Keep()` instead of checking
and calling `Del()`.
Avoid calling `Sort()` in `Builder.Labels()` if we didn't add anything,
so that `Keep()` has the same performance as `WithLabels()`.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
We know the max size of our map so we can create it with that information and avoid extra allocations
Signed-off-by: Łukasz Mierzwa <l.mierzwa@gmail.com>
If the zero threshold overlaps with the highest negative bucket and/or
the lowest positive bucket, its upper or lower boundary, respectively,
has to be adjusted. In valid histograms, only ever the highest
negative bucket and/or the lowest positive bucket may overlap with the
zero bucket. This is assumed in this code to simplify the checks.
Signed-off-by: beorn7 <beorn@grafana.com>
* refactor: move from io/ioutil to io and os packages
* use fs.DirEntry instead of os.FileInfo after os.ReadDir
Signed-off-by: MOREL Matthieu <matthieu.morel@cnp.fr>
* labels.Equal benchmark for equal, not equal, and differing lengths
Signed-off-by: Nick Pillitteri <nick.pillitteri@grafana.com>
* Compare equality of label.Label structs directly
Compare the structs using `==` instead of the name and value
of each label. This is functionally equivalent and about ~10%
faster in my testing.
Signed-off-by: Nick Pillitteri <nick.pillitteri@grafana.com>
* Use longer more realistic names and values in benchmark
Signed-off-by: Nick Pillitteri <nick.pillitteri@grafana.com>
* MergeFloatBucketIterator for []FloatBucketIterator
Signed-off-by: Ganesh Vernekar <ganeshvern@gmail.com>
* histogram_quantile for histograms
Signed-off-by: Ganesh Vernekar <ganeshvern@gmail.com>
* Fix histogram_quantile
Signed-off-by: Ganesh Vernekar <ganeshvern@gmail.com>
* Unit test and enhancements
Signed-off-by: Ganesh Vernekar <ganeshvern@gmail.com>
* Iterators to iterate buckets in reverse and all buckets together including zero bucket
Signed-off-by: Ganesh Vernekar <ganeshvern@gmail.com>
* Consider all buckets for histogram_quantile and fix the implementation
Signed-off-by: Ganesh Vernekar <ganeshvern@gmail.com>
* Remove unneeded code
Signed-off-by: Ganesh Vernekar <ganeshvern@gmail.com>
* Fix lint
Signed-off-by: Ganesh Vernekar <ganeshvern@gmail.com>
- Pick At... method via return value of Next/Seek.
- Do not clobber returned buckets.
- Add partial FloatHistogram suppert.
Note that the promql package is now _only_ dealing with
FloatHistograms, following the idea that PromQL only knows float
values.
As a byproduct, I have removed the histogramSeries metric. In my
understanding, series can have both float and histogram samples, so
that metric doesn't make sense anymore.
As another byproduct, I have converged the sampleBuf and the
histogramSampleBuf in memSeries into one. The sample type stored in
the sampleBuf has been extended to also contain histograms even before
this commit.
Signed-off-by: beorn7 <beorn@grafana.com>
Including a few adjustments for normal Histogram, too, e.g. use
pointer receiver to avoid the large copy on method calls.
Signed-off-by: beorn7 <beorn@grafana.com>
This creates a new `model` directory and moves all data-model related
packages over there:
exemplar labels relabel rulefmt textparse timestamp value
All the others are more or less utilities and have been moved to `util`:
gate logging modetimevfs pool runtime
Signed-off-by: beorn7 <beorn@grafana.com>
A lot of this code was hacked together, literally during a
hackathon. This commit intends not to change the code substantially,
but just make the code obey the usual style practices.
A (possibly incomplete) list of areas:
* Generally address linter warnings.
* The `pgk` directory is deprecated as per dev-summit. No new packages should
be added to it. I moved the new `pkg/histogram` package to `model`
anticipating what's proposed in #9478.
* Make the naming of the Sparse Histogram more consistent. Including
abbreviations, there were just too many names for it: SparseHistogram,
Histogram, Histo, hist, his, shs, h. The idea is to call it "Histogram" in
general. Only add "Sparse" if it is needed to avoid confusion with
conventional Histograms (which is rare because the TSDB really has no notion
of conventional Histograms). Use abbreviations only in local scope, and then
really abbreviate (not just removing three out of seven letters like in
"Histo"). This is in the spirit of
https://github.com/golang/go/wiki/CodeReviewComments#variable-names
* Several other minor name changes.
* A lot of formatting of doc comments. For one, following
https://github.com/golang/go/wiki/CodeReviewComments#comment-sentences
, but also layout question, anticipating how things will look like
when rendered by `godoc` (even where `godoc` doesn't render them
right now because they are for unexported types or not a doc comment
at all but just a normal code comment - consistency is queen!).
* Re-enabled `TestQueryLog` and `TestEndopints` (they pass now,
leaving them disabled was presumably an oversight).
* Bucket iterator for histogram.Histogram is now created with a
method.
* HistogramChunk.iterator now allows iterator recycling. (I think
@dieterbe only commented it out because he was confused by the
question in the comment.)
* HistogramAppender.Append panics now because we decided to treat
staleness marker differently.
Signed-off-by: beorn7 <beorn@grafana.com>
This removes the dependancy on C leveldb and snappy.
It also takes care of fewer dependencies as they would
anyway not work on any non-Debian, non-Brew system.
Change-Id: Ia70dce1ba8a816a003587927e0b3a3f8ad2fd28c
If the metrics exported by a process already contain any of a target's
base labels (such as "job" or "instance", but also any manually assigned
target-group label), don't overwrite that label, but instead add a new
label consisting of the original label name prepended with "exporter_".
This is to accomodate intermediate exporter jobs, which might indicate
e.g. the jobs and instances for which they are exporting data.
This commit updates the documentation, Makefiles, formatting, and
code semantics to support the 1.1. runtime, which includes ...
1. ``make advice``,
2. ``make format``, and
3. ``go fix`` on various targets.
This commit extracts the model.Values truncation behavior into the actual
tiered storage, which uses it and behaves in a peculiar way—notably the
retention of previous elements if the chunk were to ever go empty. This is
done to enable interpolation between sparse sample values in the evaluation
cycle. Nothing necessarily new here—just an extraction.
Now, the model.Values TruncateBefore functionality would do what a user
would expect without any surprises, which is required for the
DeletionProcessor, which may decide to split a large chunk in two if it
determines that the chunk contains the cut-off time.
This commit reduces the general compile time dependencies to omit
the Protocol Buffer compiler and the Go Protocol Buffer generator
tool. The build steps to furnish them still remain, but they can
optionally be called if data.proto or config.proto are under work.
This commit employs explicit memory freeing for the in-memory storage
arenas. Secondarily, we take advantage of smaller channel buffer sizes
in the test.
This commit introduces to Prometheus a batch database sample curator,
which corroborates the high watermarks for sample series against the
curation watermark table to see whether a curator of a given type
needs to be run.
The curator is an abstract executor, which runs various curation
strategies across the database. It remarks the progress for each
type of curation processor that runs for a given sample series.
A curation procesor is responsible for effectuating the underlying
batch changes that are request. In this commit, we introduce the
CompactionProcessor, which takes several bits of runtime metadata and
combine sparse sample entries in the database together to form larger
groups. For instance, for a given series it would be possible to
have the curator effectuate the following grouping:
- Samples Older than Two Weeks: Grouped into Bunches of 10000
- Samples Older than One Week: Grouped into Bunches of 1000
- Samples Older than One Day: Grouped into Bunches of 100
- Samples Older than One Hour: Grouped into Bunches of 10
The benefits hereof of such a compaction are 1. a smaller search
space in the database keyspace, 2. better employment of compression
for repetious values, and 3. reduced seek times.
Primary changes:
* Strictly typed unmarshalling of metric values
* Schema types are contained by the processor (no "type entity002")
Minor changes:
* Added ProcessorFunc type for expressing processors as simple
functions.
* Added non-destructive `Merge` method to `model.LabelSet`
The Protocol Buffer compiler supports generating a machine-readable
descriptor file encoded as a provided Protocol Buffer message type,
which can be used to decode messages that have been encoded with it
after-the-fact. The generated descriptor also bundles in dependent
message types.
We can use this to perform forensics on old Prometheus clients, if
necessary.
Go's time.Time represents time as UTC in its fundamental data type.
That said, when using ``time.Unix(...)``, it sets the zone for the
time representation to the local. Unfortunately with diagnosis and
our tests, it is a PITA to jump between various zones, even though
the serialized version remains the same.
To keep things easy, all places where times are generated or read
are converted into UTC. These conversions are cheap, for
``Time.In`` merely changes a pointer reference in the struct,
nothing more. This enables me to diagnose test failures with fixture
data very easily.
The curator work can be done easier if dto.SampleKey is no longer
directly accessed but rather has a higher level type around it that
captures a certain modicum of business logic. This doesn't look
terribly interesting today, but it will get more so.
