Commit graph

14 commits

Author SHA1 Message Date
Kuba Podgórski
9a02004a4f Move vprint to separate pakage 2021-03-29 14:29:19 +02:00
Travis
d192c1f24f
Merge branch 'master' into disco 2021-02-05 15:58:36 -06:00
Alan Bernstein
e397d35ed5 Include roaring field and key details in usage endpoint 2021-01-29 17:48:24 -06:00
Travis
19f91782e7
remove all instances of txsrc 2021-01-21 21:59:51 -06:00
Travis
13984353e4
remove instances of os.Getenv("PILOSA_TXSRC") 2021-01-21 21:18:46 -06:00
Travis
f292d6061a
replace pilosa.DefaultTxsrc with storage.DefaultBackend 2021-01-20 22:06:13 -06:00
Jason E. Aten
035073555a pilosa: only open views with data
- Previously, on timequantum schemas, we would
create and open a view for the cartesian
product of every possible view and shard.

- This caused us to be very slow on re-open,
and to use lots of memory for views that
held nothing.

- This change makes startup faster, memory
use much lower, and should speed migration.
2020-12-19 00:03:31 +00:00
Seebs
30d5f891c0 create naive ApplyFilter 2020-12-16 13:16:46 -06:00
Jason E. Aten
b52a814b3c short_txkey elides index and shard from the txkey
- use short_txkey for rbf
 - short_txkey breaks a bunch of bolt_test.go, so leave it on (long) txkey for now.
 - remove SliceOfShards method from Tx interface
2020-12-11 22:24:56 +00:00
Jason E. Aten
9c7bc603af rbf: reuse cursors with sync.Pool or arena
- the sync.Pool default uses little memory under CI.

- arena approach provides ability to control the maximum memory
  used by rbf Cursors.

- cursor caching is adjustable with --rbf-cursor-cache
  currently 0 by default (meaning use sync.Pool), and
  larger than 0 meaning use an arena of this size.
  With the arena, 20 or less is needed to pass CI.

- rbf test suite runs ~ 4x faster

- kitchen sink ingest test runs 16% faster.

- report TotalAlloc in CALLSTATs

fixes #1105
2020-11-19 17:02:49 +00:00
Jason Aten
748f6a61bc fine tune Tx placement, isolate Tx backends more.
- Tx creation is lazier so we don't create xtra shards.

 - Then the dir scan for blue-green state checking finds only the right shards.
2020-09-14 18:03:25 -05:00
Jason Aten
2eb097c14d blue_green migration. holdbkg.go holder goroutine.
- blue_green for doing migration. Called before Holder.Open finishes.
- holdbkg.go added for index lookup. Less wedging between a deadlock and a race.
- fix fault under read-only map under lmdb at
  TestExecutor_Execute_Row_Range/RowIDColumnID by doing cow in roaring.
- roaring -tags gofuzz builds again
- roaringparanoia build tag added to make test targets in Makefile
- add rbf.NewDBWithAllocZero for out-of-bounds memory checks
- .circleci/config.yml test-shardwidth-22 with large run container, kept OOM-ing we suspect.

Fixes #819
2020-09-11 14:24:15 -05:00
Ben Johnson
150c8a5b06 database per shard, HolderConfig, rbf bit-wise import speedups.
- introduce Query Context (Qcx) for managing database-per-shard.
- replaces the MultiTx, so mtx.go is retired and removed.
- introduces the HolderConfig struct and all Holders now have
  a path from birth.
- rbf speedups on bitwise writes
- badgerdb is removed due to unresolvable write conflicts.

fixes #703 #676
2020-09-04 13:00:33 -05:00
Jason Aten
123ce41840 add lmdb, Tx call stats, and prep for db/shard.
- lmdb as a backend (lmdb.go)
   (lmdb is the fastest known transactional storage backend)
 - per Tx call statics report enabled with PILOSA_CALLSTAT=true (stattx.go)
 - framework for per-shard db (dbshard.go)
 - txfactory handles any pair under blue-green testing (txfactory.go)
 - enable CGO in Dockerfiles for lmdb
2020-08-17 18:26:58 -05:00