Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Revert "remove restricted dpbindings" | Harald Musum | 2023-08-21 | 7 | -44/+20 |
| | |||||
* | Merge pull request #27996 from ↵ | Morten Tokle | 2023-08-21 | 7 | -20/+44 |
|\ | | | | | | | | | vespa-engine/mortent/remove-restricted-dpbindings2 remove restricted dpbindings | ||||
| * | Rename method, change scope | Morten Tokle | 2023-08-09 | 5 | -6/+6 |
| | | |||||
| * | Keep original port when restricting bindings to dataplane port | Morten Tokle | 2023-08-09 | 3 | -5/+36 |
| | | |||||
| * | Remove flag restrict-data-plane-bindings | Morten Tokle | 2023-08-09 | 4 | -12/+5 |
| | | |||||
* | | Use zstd as default compression method for logarchive for hosted apps | Harald Musum | 2023-08-17 | 1 | -2/+4 |
| | | |||||
* | | Require redundancy of at least 1 | Henning Baldersheim | 2023-08-17 | 1 | -2/+2 |
| | | |||||
* | | Merge pull request #28054 from ↵ | Arne H Juul | 2023-08-17 | 5 | -0/+39 |
|\ \ | | | | | | | | | | | | | vespa-engine/geirst/target-hits-max-adjustment-factor Control the auto-adjustment of targetHits in ANN using post-filtering. | ||||
| * | | Control the auto-adjustment of targetHits in ANN using post-filtering. | Geir Storli | 2023-08-15 | 5 | -0/+39 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When searching the HNSW index in a post-filtering case, targetHits is auto-adjusted in an effort to still expose targetHits hits to first-phase ranking after post-filtering. The following formula is now used to ensure an upper bound of adjustedTargetHits, avoiding that the search in the HNSW index takes too long. adjustedTargetHits = min(targetHits / estimatedHitRatio, targetHits * targetHitsMaxAdjustmentFactor). The target-hits-max-adjustment-factor can be set in a rank profile and overriden per query. The value is in the range [1.0,inf], with the default being 20.0. When setting this to 1.0, auto-adjustment of targetHits is effectively disabled. | ||||
* | | | Allow infrastructure to override application type | Morten Tokle | 2023-08-15 | 2 | -4/+42 |
|/ / | |||||
* / | Correctly classify references to missing fields | Jon Bratseth | 2023-08-14 | 3 | -12/+9 |
|/ | |||||
* | Always use stricter validation for complex fields with struct field attributes. | Geir Storli | 2023-08-08 | 5 | -32/+23 |
| | |||||
* | Clarify error handling | Henning Baldersheim | 2023-08-03 | 1 | -2/+1 |
| | |||||
* | GC enable-conditional-put-remove-write-repair | Henning Baldersheim | 2023-08-02 | 2 | -15/+3 |
| | |||||
* | Merge pull request #27869 from vespa-engine/hmusum/remove-version-7-schemas | Henning Baldersheim | 2023-08-02 | 25 | -1531/+0 |
|\ | | | | | Don't install XML schema files for vespa 7.x anymore | ||||
| * | Don't install XML schema files for vespa 7.x anymore | Harald Musum | 2023-07-21 | 25 | -1531/+0 |
| | | |||||
* | | Since redundancy is created late in the process we need to access it via a ↵ | Henning Baldersheim | 2023-08-02 | 3 | -14/+20 |
| | | | | | | | | provider instead of providing in the constructor. | ||||
* | | Modernize code | Henning Baldersheim | 2023-08-02 | 2 | -32/+17 |
| | | |||||
* | | Revert "Revert "- initial document count must be number of unique documents ↵ | Henning Baldersheim | 2023-08-02 | 3 | -12/+23 |
| | | | | | | | | per node." MERGEOK" | ||||
* | | Merge pull request #27948 from vespa-engine/balder/cleanup-unused-complexity | Bjørn Christian Seime | 2023-08-02 | 3 | -130/+1 |
|\ \ | | | | | | | Cleanup unused stuff to reduce entanglement | ||||
| * | | Cleanup unused stuff to reduce entanglement | Henning Baldersheim | 2023-08-02 | 3 | -130/+1 |
| | | | |||||
* | | | Merge pull request #27940 from vespa-engine/olaa/infrastructure-metric-sets | Ola Aunrønning | 2023-08-02 | 15 | -1385/+27 |
|\ \ \ | |/ / |/| | | Infrastructure metric sets | ||||
| * | | Move metric sets to metrics module | Ola Aunronning | 2023-08-01 | 15 | -1385/+27 |
| | | | |||||
* | | | Revert "- initial document count must be number of unique documents per node." | Arnstein Ressem | 2023-08-01 | 3 | -23/+12 |
|/ / | |||||
* | | Merge pull request #27935 from ↵ | Yngve Aasheim | 2023-08-01 | 3 | -52/+183 |
|\ \ | | | | | | | | | | | | | vespa-engine/yngveaasheim/prepare-metricset-config-server Add separate infrastructure metric set | ||||
| * | | Include remaining metrics used by alerts. Fix two comments from Ola. | yngveaasheim | 2023-07-31 | 2 | -3/+11 |
| | | | |||||
| * | | Add metric used for alerting | yngveaasheim | 2023-07-31 | 1 | -0/+2 |
| | | | |||||
| * | | Separate out infrastructure metrics, include in predefined metrics sets | yngveaasheim | 2023-07-31 | 3 | -112/+171 |
| | | | |||||
| * | | Document and add some more infrastructure metrics | yngveaasheim | 2023-07-31 | 1 | -0/+1 |
| | | | |||||
| * | | Remove duplicate entries | yngveaasheim | 2023-07-31 | 1 | -5/+0 |
| | | | |||||
| * | | Separate controller specific metrics into separate enum and method | yngveaasheim | 2023-07-31 | 1 | -23/+24 |
| | | | |||||
| * | | Add enums for infrastructure and add to vespametricsset as needed for ↵ | yngveaasheim | 2023-07-31 | 1 | -5/+70 |
| | | | | | | | | | | | | infrastructure services. | ||||
* | | | - initial document count must be number of unique documents per node. | Henning Baldersheim | 2023-07-31 | 3 | -12/+23 |
|/ / | | | | | | | The reason is that it is the backend that decides how it will use this number for presizing internally. | ||||
* | | Remove active attribute | Martin Polden | 2023-07-27 | 1 | -13/+4 |
| | | |||||
* | | Remove global-service-id | Martin Polden | 2023-07-27 | 2 | -4/+0 |
| | | |||||
* | | Revert "Remove global-service-id" | Martin Polden | 2023-07-27 | 3 | -4/+17 |
| | | |||||
* | | Remove active attribute | Martin Polden | 2023-07-26 | 1 | -13/+4 |
| | | |||||
* | | Remove global-service-id | Martin Polden | 2023-07-26 | 2 | -4/+0 |
| | | |||||
* | | Merge pull request #27888 from vespa-engine/mpolden/remove-unused-attrib | Martin Polden | 2023-07-25 | 1 | -2/+0 |
|\ \ | | | | | | | Remove unused gpu-device attribute | ||||
| * | | Remove unused gpu-device attribute | Martin Polden | 2023-07-25 | 1 | -2/+0 |
| | | | |||||
* | | | Use -vocab as suffix to allow only provide transformer model-id | Jo Kristian Bergum | 2023-07-25 | 1 | -2/+2 |
|/ / | |||||
* / | Revert "Enable TLSv1.3 for hosted endpoints" | Bjørn Christian Seime | 2023-07-21 | 1 | -1/+2 |
|/ | | | | This reverts commit ad7707a29b02872e2ab45fbbf22205fbee34ab97. | ||||
* | Merge pull request #27857 from vespa-engine/bjorncs/tls13 | Bjørn Christian Seime | 2023-07-21 | 1 | -2/+1 |
|\ | | | | | Bjorncs/tls13 | ||||
| * | Enable TLSv1.3 for hosted endpoints | Bjørn Christian Seime | 2023-07-20 | 1 | -2/+1 |
| | | |||||
* | | Merge pull request #27854 from vespa-engine/bratseth/scale-down-less | Martin Polden | 2023-07-21 | 1 | -4/+4 |
|\ \ | |/ |/| | Don't scale down if we are likely to scale back up | ||||
| * | Don't scale down if we are likely to scale back up | Jon Bratseth | 2023-07-20 | 1 | -4/+4 |
| | | | | | | | | | | | | | | | | | | When we decide to scale number of nodes down we'll see a lower ideal load because we need to be able to handle one node going down. This may lead us to be closer to ideal (at current peak load) than otherwise anticipated, such that we are quite likely to soon scale back up. This checks for that and avoids scaling down dimensions where this is the case. | ||||
* | | Only include relevant client definitions for filters | Bjørn Christian Seime | 2023-07-20 | 2 | -0/+2 |
| | | |||||
* | | Merge pull request #27843 from vespa-engine/jobergum/add-new-embedding-models | Jo Kristian Bergum | 2023-07-20 | 1 | -0/+9 |
|\ \ | | | | | | | Add new models and int8 versions of selected models | ||||
| * | | Add new models and int8 versions of selected models | Jo Kristian Bergum | 2023-07-20 | 1 | -0/+9 |
| | | | |||||
* | | | Add port for token connector to nginx config | Bjørn Christian Seime | 2023-07-19 | 2 | -4/+8 |
| | | |