aboutsummaryrefslogtreecommitdiffstats
path: root/configdefinitions
Commit message (Collapse)AuthorAgeFilesLines
* Remove athenz-provider-service.defValerij Fredriksen2024-05-102-39/+0
|
* maxtokenlength units are characters.Tor Egge2024-05-061-1/+1
|
* Add max token length to ilscripts config.Tor Egge2024-05-061-0/+2
|
* Merge pull request #31011 from ↵Marius Arhaug2024-04-301-2/+0
|\ | | | | | | | | vespa-engine/marius/update-significance-model-fields Update significance model field and logic from architect meeting
| * Update significance model field and logic from architect meetingMariusArhaug2024-04-241-2/+0
| |
* | add prepend supportJo Kristian Bergum2024-04-251-0/+2
|/
* add vespa-otelcol-startArne Juul2024-04-122-1/+6
|
* Otel on logserver WIPOla Aunronning2024-04-121-0/+5
|
* Support pipelining (batching) of mutating ops to same bucketTor Brede Vekterli2024-04-091-1/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Bucket operations require either exclusive (single writer) or shared (multiple readers) access. Prior to this commit, this means that many enqueued feed operations to the same bucket introduce pipeline stalls due to each operation having to wait for all prior operations to the bucket to complete entirely (including fsync of WAL append). This is a likely scenario when feeding a document set that was previously acquired through visiting, as such documents will inherently be output in bucket-order. With this commit, a configurable number of feed operations (put, remove and update) bound for the exact same bucket may be sent asynchronously to the persistence provider in the context of the _same_ write lock. This mirrors how merge operations work for puts and removes. Batching is fairly conservative, and will _not_ batch across further messages when any of the following holds: * A non-feed operation is encountered * More than one mutating operation is encountered for the same document ID * No more persistence throttler tokens can be acquired * Max batch size has been reached Updating the bucket DB, assigning bucket info and sending replies is deferred until _all_ batched operations complete. Max batch size is (re-)configurable live and defaults to a batch size of 1, which shall have the exact same semantics as the legacy behavior. Additionally, clock sampling for persistence threads have been abstracted away to allow for mocking in tests (no need for sleep!).
* Add config for significance modelsMariusArhaug2024-04-033-0/+14
|
* bump maxtermoccurrences 1000 => 10000Tor Egge2024-02-201-1/+1
|
* - Remove multibit_split form config, as it is always off, but leave it for ↵Henning Baldersheim2024-02-051-6/+0
| | | | | | tests. - Reduce penetration of generated StorFilestorConfig.
* Merge pull request #30165 from vespa-engine/balder/gc-unused-distribution-configHenning Baldersheim2024-02-051-20/+0
|\ | | | | Balder/gc unused distribution config
| * GC unused distributor_auto_ownership_transfer_on_whole_group_downHenning Baldersheim2024-02-031-8/+0
| |
| * GC unused disk_distribution config.Henning Baldersheim2024-02-031-13/+1
| |
* | common_merge_chain_optimalization_minimum_size hardcoded at 64Henning Baldersheim2024-02-031-6/+0
| |
* | throttle_individual_merge_feed_ops has long been enabled, cleaning upHenning Baldersheim2024-02-031-5/+2
|/
* GC completely unused parameters from the days of VDSHenning Baldersheim2024-01-301-17/+0
|
* GC unused async_operation_dynamic_throttling_window_increment and ↵Henning Baldersheim2024-01-301-21/+0
| | | | async_operation_throttler_type
* GC leftovers from use_per_document_throttled_delete_bucketHenning Baldersheim2024-01-301-8/+0
|
* GC control of use-per-document-delete and max-merge-memory from config ↵Henning Baldersheim2024-01-231-1/+1
| | | | production side in java.
* bump maxtermoccurrences 100 => 1000Henning Baldersheim2024-01-151-1/+1
|
* handle multilingual models betterJo Kristian Bergum2024-01-061-0/+3
|
* Add a splade embedder implementationJo Kristian Bergum2023-12-152-0/+30
|
* Add and wire live config for selecting `DeleteBucket` behaviorTor Brede Vekterli2023-11-101-0/+8
| | | | By default the legacy behavior is used.
* add config for normalizersArne Juul2023-10-111-0/+11
|
* Update copyrightJon Bratseth2023-10-0967-70/+73
|
* - Reduce max lids per file and max file size to 4M and 256M during unit testing.Henning Baldersheim2023-10-051-1/+1
| | | | - Reduce max lids from 40M to 8M as default configuration.
* Install config definitionBjørn Christian Seime2023-09-211-0/+1
|
* Add ColBERT embedderJo Kristian Bergum2023-09-211-0/+36
|
* Add token endpoints to proxy configMorten Tokle2023-09-081-0/+3
|
* Add numProxiesAllowedDown fields to orchestrator defHåkon Hallingstad2023-07-311-0/+8
|
* Add port for token connector to nginx configBjørn Christian Seime2023-07-191-1/+2
|
* Split token authz into dedicated filter `CloudTokenDataPlaneFilter`Bjørn Christian Seime2023-07-193-5/+11
|
* Add expiration concept to data plane tokensBjørn Christian Seime2023-07-121-0/+1
|
* Add parameters for tokens to config definitionBjørn Christian Seime2023-06-141-0/+4
|
* Install config definitionBjørn Christian Seime2023-06-141-1/+1
|
* DataplaneProxyConfig does not contain endpointsOla Aunronning2023-06-131-6/+0
|
* Prefer truncation configuration from tokenizer modelBjørn Christian Seime2023-06-121-3/+10
| | | | | | | Only override truncation if not specified or max length exceeds max tokens accepted by model. Use JNI wrapper directly to determine existing truncation configuration (JSON format is not really documented). Simply configuration for pure tokenizer embedder. Disable DJL usage telemetry.
* Merge pull request #27349 from vespa-engine/bjorncs/disable-paddingBjørn Christian Seime2023-06-081-2/+3
|\ | | | | Bjorncs/disable padding
| * Disable padding and make it configurableBjørn Christian Seime2023-06-081-2/+3
| |
* | Merge branch 'master' into olaa/dataplane-proxy-configOla Aunrønning2023-06-088-1/+104
|\|
| * Fix typoBjørn Christian Seime2023-06-071-1/+1
| |
| * Ensure config definitions are installed on configserverBjørn Christian Seime2023-06-073-0/+3
| |
| * Merge pull request #27297 from vespa-engine/bjorncs/bert-embedder-services-xmlBjørn Christian Seime2023-06-062-0/+34
| |\ | | | | | | Bjorncs/bert embedder services xml
| | * Make pooling strategy configurable for Huggingface embedderBjørn Christian Seime2023-06-051-0/+2
| | |
| | * Move config definition to `configdefinitions`Bjørn Christian Seime2023-06-051-0/+32
| | |
| * | Add necessary options to use failOnWarningsgjoranv2023-06-051-0/+1
| |/
| * Introduce services.xml syntax for configuring HuggingFace embeddersBjørn Christian Seime2023-06-024-0/+60
| |
| * Remove use of stateGatherCount config, simplify and deprecate config fieldHarald Musum2023-05-301-0/+1
| |