* Protect penciller from empty ledger cache updates
which may occur when loading the ledger from the journal, after the ledger has been cleared.
* Score caching and randomisation
The test allkeydelta_journal_multicompact can occasionally fail when a compaction doesn't happen, but then does the next loop. Suspect this is as a result of score caching, randomisation of key grabs for scoring, plus jitter on size boundaries.
Modified test for predictability.
Plus formatting changes
* Avoid small batches
Avoid small batches due to large SQN gaps
* Rationalise tests
Two tests overlaps with the new, much broader, replace_everything/1 test. Ported over any remaining checks of interest and dropped two tests.
* Mas i370 patch d (#383)
* Refactor penciller memory
In high-volume tests on large key-count clusters, so significant variation in the P0031 time has been seen:
TimeBucket PatchA
a.0ms_to_1ms 18554
b.1ms_to_2ms 51778
c.2ms_to_3ms 696
d.3ms_to_5ms 220
e.5ms_to_8ms 59
f.8ms_to_13ms 40
g.13ms_to_21ms 364
h.21ms_to_34ms 277
i.34ms_to_55ms 34
j.55ms_to_89ms 17
k.89ms_to_144ms 21
l.144ms_to_233ms 31
m.233ms_to_377ms 45
n.377ms_to_610ms 52
o.610ms_to_987ms 59
p.987ms_to_1597ms 55
q.1597ms_to_2684ms 54
r.2684ms_to_4281ms 29
s.4281ms_to_6965ms 7
t.6295ms_to_11246ms 1
It is unclear why this varies so much. The time to add to the cache appears to be minimal (but perhaps there is an issue with timing points in the code), whereas the time to add to the index is much more significant and variable. There is also variable time when the memory is rolled (although the actual activity here appears to be minimal.
The refactoring here is two-fold:
- tidy and simplify by keeping LoopState managed within handle_call, and add more helpful dialyzer specs;
- change the update to the index to be a simple extension of a list, rather than any conversion.
This alternative version of the pmem index in unit test is orders of magnitude faster to add - and is the same order of magnitude to check. Anticipation is that it may be more efficient in terms of memory changes.
* Compress SST index
Reduces the size of the leveled_sst index with two changes:
1 - Where there is a common prefix of tuple elements (e.g. Bucket) across the whole leveled_sst file - only the non-common part is indexed, and a function is used to compare.
2 - There is less "indexing" of the index i.e. only 1 in 16 keys are passed into the gb_trees part instead of 1 in 4
* Immediate hibernate
Reasons for delay in hibernate were not clear.
Straight after creation the process will not be in receipt of messages (must wait for the manifest to be updated), so better to hibernate now. This also means the log PC023 provides more accurate information.
* Refactor BIC
This patch avoids the following:
- repeated replacement of the same element in the BIC (via get_kvrange), by checking presence via GET before sing SET
- Stops re-reading of all elements to discover high modified date
Also there appears to have been a bug where a missing HMD for the file is required to add to the cache. However, now the cache may be erased without erasing the HMD. This means that the cache can never be rebuilt
* Use correct size in test results
erts_debug:flat_size/1 returns size in words (i.e. 8 bytes on 64-bit CPU) not bytes
* Don't change summary record
As it is persisted as part of the file write, any change to the summary record cannot be rolled back
* Clerk to prompt L0 write
Simplifies the logic if the clerk request work for the penciller prompts L0 writes as well as Manifest changes.
The advantage now is that if the penciller memory is full, and PUT load stops, the clerk should still be able to prompt persistence. the penciller can therefore make use of dead time this way
* Add push on journal compact
If there has been a backlog, followed by a quiet period - there may be a large ledger cache left unpushed. Journal compaction events are about once per hour, so the performance overhead of a false push should be minimal, with the advantage of clearing any backlog before load starts again.
This is only relevant to riak users with very off/full batch type workloads.
* Extend tests
To more consistently trigger all overload scenarios
* Fix range keys smaller than prefix
Can't make end key an empty binary in this case, as it may be bigger than any keys within the range, but will appear to be smaller.
Unit tests and ct tests added to expose the potential issue
* Tidy-up
- Remove penciller logs which are no longer called
- Get pclerk to only wait MIN_TIMEOUT after doing work, in case there is a backlog
- Remove update_levelzero_cache function as it is unique to handle_call of push_mem, and simple enough to be inline
- Alight testutil slow offer with standard slow offer used
* Tidy-up
Remove pre-otp20 references.
Reinstate the check that the starting pid is still active, this was added to tidy up shutdown.
Resolve failure to run on otp20 due to `-if` sttaement
* Tidy up
Using null rather then {null, Key} is potentially clearer as it is not a concern what they Key is in this case, and removes a comparison step from the leveled_codec:endkey_passed/2 function.
There were issues with coverage in eunit tests as the leveled_pclerk shut down. This prompted a general tidy of leveled_pclerk (remove passing of LoopState into internal functions, and add dialyzer specs.
* Remove R16 relic
* Further testing another issue
The StartKey must always be less than or equal to the prefix when the first N characters are stripped, but this is not true of the EndKey (for the query) which does not have to be between the FirstKey and the LastKey.
If the EndKey query does not match it must be greater than the Prefix (as otherwise it would not have been greater than the FirstKey - so set to null.
* Fix unit test
Unit test had a typo - and result interpretation had a misunderstanding.
* Code and spec tidy
Also look to the cover the situation when the FirstKey is the same as the Prefix with tests.
This is, in theory, not an issue as it is the EndKey for each sublist which is indexed in leveled_tree. However, guard against it mapping to null here, just in case there are dangers lurking (note that tests will still pass without `M > N` guard in place.
* Hibernate on BIC complete
There are three situations when the BIC becomes complete:
- In a file created as part of a merge the BIS is learned in the merge
- After startup, files below L1 learn the block cache through reads that happen to read the block, eventually the while cache will be read, unless...
- Either before/after the cache is complete, it can get whiped by a timeout after a get_sqn request (e.g. as prompted by a journal compaction) ... it will then be re-filled of the back of get/get-range requests.
In all these situations we want to hibernate after the BIC is fill - to reflect the fact that the LoopState should now be relatively stable, so it is a good point to GC and rationalise location of data.
Previously on the the first base was covered. Now all three are covered through the bic_complete message.
* Test all index keys have same term
This works functionally, but is not optimised (the term is replicated in the index)
* Summaries with same index term
If the summary index all have the same index term - only the object keys need to be indexes
* Simplify case statements
We either match the pattern of <<Prefix:N, Suffix>> or the answer should be null
* OK for M == N
If M = N for the first key, it will have a suffix of <<>>. This will match (as expected) a query Start Key of the sam size, and be smaller than any query Start Key that has the same prefix.
If the query Start Key does not match the prefix - it will be null - as it must be smaller than the Prefix (as other wise the query Start Key would be bigger than the Last Key).
The constraint of M > N was introduced before the *_prefix_filter functions were checking the prefix, to avoid issues. Now the prefix is being checked, then M == N is ok.
* Simplify
Correct the test to use a binary field in the range.
To avoid further issue, only apply filter when everything is a binary() type.
* Add test for head_only mode
When leveled is used as a tictacaae key store (in parallel mode), the keys will be head_only entries. Double check they are handled as expected like object keys
* Revert previous change - must support typed buckets
Add assertion to confirm worthwhile optimisation
* Add support for configurable cache multiple (#375)
* Mas i370 patch e (#385)
Improvement to monitoring for efficiency and improved readability of logs and stats.
As part of this, where possible, tried to avoid updating loop state on READ messages in leveled processes (as was the case when tracking stats within each process).
No performance benefits found with change, but improved stats has helped discover other potential gains.
A test thta will cause leveled to crash due to a low cache size being set - but protect against this (as well as the general scenario of the cache being full).
There could be a potential case where a L0 file present (post pending) without work backlog being set. In this case we want to roll the level zero to memory, but don't accept the cache update if the L0 cache is already full.
logs should always at least have a log_level defined which is in the ?LOG_LEVELS - so the final case of
`-is_active_level([] , _, _) -> false.`
is unnecessary (even if a bad log level is passed in).
correct Bucket and Key definition to hit expected eunit test cases in leveled_pmem
Simple change to check for presence of objetc in list before adding it:
```Head check took 124416 microseconds checking list of length 5000
Head check took 130114 microseconds checking list of length 5000
Check for presence of repeated objects
Head check took 1725 microseconds checking list of length 5```
head_only mode cna be run with_lookup - but there is no L0 index created in this case.
So the L0 index wasn't returning a potition list and the L0 cache wasn't being checked.
Code now checks every position in the L0 cache, when a lookup is attempted in head_only mode.
Switch from magic hash to md5 - to hopefully remove the need for some
of the artificial jumps required to get expected fall positive ratios.
Also split the hash into two 16-bit integers. We assume that SegmentID
(from the perspective of AAE merkle/tictac trees) will always be at
least 16 bits. the idea is that hashes should be used in blooms and
indexes such that some advantage can be gained from just knowing the
segmentID - in particular when folding over all the keys in a bucket.
Performance testing has been difficult so far - I think due to “cloud”
mysteries.
Remove the ets index in pmem and use a binary index instead. This may
be slower, but avoids the bulk upload to ets, and means that matches
know of position (so only skiplists with a match need be tried).
Also stops the discrepancy between snapshots and non-snapshots - as
previously the snapshots were always slowed by not having access to the
ETS table.
Move to using the DJ Bernstein Magic Hash consistently, and trying to
make sure we only hash once for each operation (as the hash is more
expensive than phash2).
The improved lookup time for missing keys should allow for the L0 index
to be removed, and hence speed up the completion time for push_mem
operations.
It is expected there will be a second stage of creating a tinybloom as
part of the SFT creation process, and then adding that tinybloom to the
manifest. This will then reduce the message passing required for a GET
not in the cache or higher levels
Hope is that this will cause less garbage collection, and also will be
slightly faster.
Note that snapshots don't now get an index - they get the special index
'snap'. However, the SkipLists have bloom protection, and most
snapshots are iterators not fetchers.
Move legacy CDB code used only in unit tests into test area. Fix column
width in pmem and comment out the unused case statement (in healthy
tests) from the penciller test code
There was some unpredictable performance in tests, that was related to
the amount of time it took the sft gen_server to accept a cast whihc
passed the levelzero_cache.
The response time looked to be broadly proportional to the size of the
cache - so it appeared to be an issue with passing the large object to
the process queue.
To avoid this, the penciller now instructs the SFT gen_server to
callback to the server for each tree in the cache in turn as it is
building the list from the cache. Each of these requests should be
reltaively short, and the processing in-between should space out the
requests so the Pencille ris not blocked from answering queries when
pompting a L0 write.
Plugged the ne wpencille rmemory into the Penciller, and took advantage
of the increased speed to simplify the callbacks involved.
The outcome is much simpler code
Not yet integrated, but there is now unit-tested module for the new way
of managing L0 memory cache in the Penciller.
This mechansim is considerably more efficient than previous efforts and
should allow for further simplification of the code.
