summaryrefslogtreecommitdiffstats
path: root/searchcore
Commit message (Collapse)AuthorAgeFilesLines
* Revert "Count number of bits to create a better estimate of number of hits ↵Harald Musum2022-05-231-2/+1
| | | | it…"
* - Stick to using GrowStragey everywhere.Henning Baldersheim2022-05-231-7/+2
| | | | - Store the full config in the rcu vector.
* Merge pull request #22716 from ↵Henning Baldersheim2022-05-231-1/+2
|\ | | | | | | | | vespa-engine/balder/let-whitelist-blueprint-be-less-conservative Count number of bits to create a better estimate of number of hits it…
| * Count number of bits to create a better estimate of number of hits it will ↵Henning Baldersheim2022-05-231-1/+2
| | | | | | | | produce.
* | - Avoid using shared_ptr for simpe AllocConfig and ThreadingServiceConfig.Henning Baldersheim2022-05-2331-151/+141
|/ | | | - Reduce exposure of DocumentDBConfig and DocumentDBConfigManager classes.
* Fold searchlib_uca into searchlibHenning Baldersheim2022-05-218-8/+0
|
* Fold fastlib into vespalib and gc some unused code.Henning Baldersheim2022-05-211-1/+0
| | | | Also move some code only used by juniper up into juniper test module.
* Fold staging_vespalib into vespalibHenning Baldersheim2022-05-202-3/+3
|
* Fold juniper into searchsummary library.Henning Baldersheim2022-05-191-1/+0
|
* GC unused vespa-dump-feed binaryHenning Baldersheim2022-05-194-230/+0
|
* take read guards while accessing bitvectorsHåvard Pettersen2022-05-192-1/+6
|
* Move state_server, metrivs and some all executors from staging_vespalib too ↵Henning Baldersheim2022-05-1821-52/+45
| | | | vespalib.
* Remove feeding concurrency setting per document db that is no longer used.Geir Storli2022-05-181-4/+0
|
* Move tracking of num field writer threads from ThreadingServiceConfig to ↵Geir Storli2022-05-1816-123/+101
| | | | | | SharedThreadingServiceConfig. This is a follow-up for when the shared field writer executor is used across all document dbs.
* Merge pull request #22633 from ↵Henning Baldersheim2022-05-1836-124/+93
|\ | | | | | | | | vespa-engine/balder/move-persistence-spi-types-under-vespa Balder/move persistence spi types under vespa
| * - Move persitence/spi/types.h under to persitence/spi/types.hHenning Baldersheim2022-05-1836-124/+93
| | | | | | | | - Cut dependency to persistencetypes for searchlib.
* | Move ScheduledExecutor to searchcore and drop fnet dependency in ↵Henning Baldersheim2022-05-1811-9/+202
|/ | | | staging_vespalib in order to prepare collapsing stagingg vespalib into vespalib
* Remove unused code after the shared field writer is used for all document dbs.Geir Storli2022-05-1614-193/+74
|
* Collapse searchcorespi into searchcoreHenning Baldersheim2022-05-1587-10/+5905
|
* Merge pull request #22588 from ↵Henning Baldersheim2022-05-131-0/+32
|\ | | | | | | | | vespa-engine/havardpe/default-query-feature-tensor-value support default tensor values for query feature
| * support default tensor values for query featureHåvard Pettersen2022-05-131-0/+32
| |
* | Add explorer for the hardware information on the content node machine.Geir Storli2022-05-135-1/+65
| |
* | Use vespalib::ResourceLimits in proton::HwInfoSampler.Tor Egge2022-05-131-6/+8
|/
* Merge pull request #22569 from ↵Tor Egge2022-05-123-29/+102
|\ | | | | | | | | vespa-engine/geirst/global-filter-and-searchable-copies Auto-adjust global filter tuning parameters to handle searchable-copi…
| * Auto-adjust global filter tuning parameters to handle searchable-copies > 1.Geir Storli2022-05-123-29/+102
| |
* | Merge pull request #22571 from ↵Henning Baldersheim2022-05-124-3/+43
|\ \ | | | | | | | | | | | | vespa-engine/vekterli/improve-handling-of-preempted-bucket-move Don't attempt to execute document moves from a cancelled bucket mover [run-systemtest]
| * | Don't attempt to actually execute document moves from a cancelled bucket moverTor Brede Vekterli2022-05-124-3/+43
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This prevents the following race condition where the bucket mover logic fails to notify the content layer that the bucket sub DB status has changed for a particular bucket: 1. Bucket state is changed over SPI, a mover is created and registered and a BucketTask is scheduled onto the persistence queues to actually do the document reads and finalize the move. 2. Before the bucket task is executed, bucket state is changed again over the SPI. A new mover is created, the old one is cancelled (tagging mover as not consistent) and another BucketTask is scheduled onto the persistence queues. Note: the old task still remains. 3. Old bucket task is executed and performs the actual document moving despite being cancelled. No notification is done towards the content layer since the mover was tagged as not being consistent. 4. New bucket task is executed and tries to move the same document set as the old mover. Since the documents are no longer present in the source document DB, the moves fail. This tags the mover as inconsistent and no notification is done. Bucket is automatically rechecked, but since all docs are already moved away there is nothing more to do and no subsequent mover is created. This means the "should notify?" edge is not triggered and the content layer remains blissfully unaware of any sub DB changes. This commit simply changes cancellation to actually inhibit document moves from taking place. This lets the preempting mover successfully complete its moves, thus triggering the notify-edge as expected.
* | Start the transport in teh constructor.Henning Baldersheim2022-05-121-4/+5
| |
* | Wrap the FNET_Transport to make it exception safe.Henning Baldersheim2022-05-121-4/+16
|/
* make new bitvector when growingHåvard Pettersen2022-05-117-44/+44
|
* Auto-adjust targetHits for approximate nearest neighbor search when using ↵Geir Storli2022-05-102-3/+10
| | | | | | | | post-filtering. The goal is to expose 'targetHits' hits to first-phase ranking. Before searching the HNSW index, targetHits is adjusted based on the estimated hit ratio of the query to compensate for the hits that will be removed in post-filtering.
* Merge pull request #22516 from ↵Henning Baldersheim2022-05-091-1/+3
|\ | | | | | | | | vespa-engine/havardpe/move-functions-to-appropriate-classes move functions to more appropriate classes
| * move functions to more appropriate classesHåvard Pettersen2022-05-091-1/+3
| | | | | | | | | | | | preparing to make GrowableBitVector an atomic switch between bitvectors rather than a bitvector itself (to avoid overwriting its own state while being visible to other threads).
* | Use a single overall timeout for file resolving of 1 hour instead of 3 ↵Henning Baldersheim2022-05-091-4/+2
|/ | | | separate each 10m.
* Move shared_ptr in BucketMoveJob::StartMove constructorTor Brede Vekterli2022-05-061-4/+4
|
* Re-sample max threads in loop. Wake up waiters after applying new config.Tor Egge2022-05-061-2/+3
|
* Use atomic _maxThreads, _coverage and _minHits member variables inTor Egge2022-05-062-10/+15
| | | | proton::matching::QueryLimiter.
* Merge pull request #22463 from ↵Henning Baldersheim2022-05-061-8/+9
|\ | | | | | | | | vespa-engine/toregge/make-current-in-tlsreplayprogress-atomic Make proton::TlsReplayProgress::_current atomic.
| * Make proton::TlsReplayProgress::_current atomic.Tor Egge2022-05-051-8/+9
| |
* | Use atomic proton::MatchEngine::_nodeMaintenance.Tor Egge2022-05-062-3/+3
|/
* Merge pull request #22454 from vespa-engine/bjorncs/annotationsBjørn Christian Seime2022-05-051-0/+6
|\ | | | | Bjorncs/annotations [run-systemtest]
| * Don't embed annotations in osgi bundlesBjørn Christian Seime2022-05-041-0/+6
| |
* | Merge pull request #22423 from ↵Henning Baldersheim2022-05-052-2/+15
|\ \ | | | | | | | | | | | | vespa-engine/toregge/wait-for-load-done-before-getting-numdocs Wait for load being completed before getting number of documents.
| * | Wait for load being completed before getting number of documents.Tor Egge2022-05-032-2/+15
| |/
* / Make proton::DocumentMetaStore::_shrinkLidSpaceBlockers atomic.Tor Egge2022-05-042-7/+11
|/
* use atomic operations to propagate bits and bitvector sizeHåvard Pettersen2022-05-032-1/+2
| | | | minor refactoring due to adding load/store word functions
* Store document db state in atomic variables.Tor Egge2022-05-022-32/+36
|
* Fix race conditions in vespalib::Cache.Tor Egge2022-04-296-12/+12
|
* Use atomic for _lowest and _highest in proton::LidStateVector.Tor Egge2022-04-282-45/+40
|
* Merge pull request #22318 from ↵Håvard Pettersen2022-04-271-24/+33
|\ | | | | | | | | vespa-engine/toregge/use-atomic-variables-for-raw-document-meta-data Use atomic variables in proton::RawDocumentMetaData.