Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Remove the unused [header] and [body] fieldsets to simplify backend ↵ | Henning Baldersheim | 2020-08-04 | 1 | -1/+3 |
| | | | | optimization during get. | ||||
* | cleanup | Arne Juul | 2020-07-17 | 1 | -1/+1 |
| | |||||
* | ChildProc -> ChildProcess | Arne Juul | 2020-07-17 | 1 | -6/+6 |
| | |||||
* | rename SlaveProc -> ChildProc | Arne Juul | 2020-07-17 | 1 | -6/+6 |
| | |||||
* | Detect changed bucket space for document type. | Tor Egge | 2020-07-16 | 2 | -69/+96 |
| | |||||
* | finish rewrite to use GeoLocation | Arne Juul | 2020-07-15 | 1 | -1/+1 |
| | |||||
* | preserve old behavior | Arne Juul | 2020-07-15 | 1 | -1/+2 |
| | |||||
* | Rename context to IDestructorCallbackSP | Henning Baldersheim | 2020-07-08 | 3 | -6/+6 |
| | |||||
* | Merge pull request #13820 from vespa-engine/toregge/consolidate-search-for-gtest | Tor Egge | 2020-07-07 | 11 | -11/+0 |
|\ | | | | | Consolidate search for GTest. | ||||
| * | Consolidate search for GTest. | Tor Egge | 2020-07-07 | 11 | -11/+0 |
| | | |||||
* | | Ensure proper lifetime of the attribute executor. | Henning Baldersheim | 2020-07-07 | 1 | -4/+3 |
| | | |||||
* | | Sync executor before completing destruction to ensure everyone is done. | Henning Baldersheim | 2020-07-07 | 1 | -1/+0 |
| | | |||||
* | | Async handling of gid/lid changes. | Henning Baldersheim | 2020-07-07 | 4 | -21/+27 |
|/ | |||||
* | Use find_package to find gtest library. | Tor Egge | 2020-06-29 | 11 | -12/+23 |
| | |||||
* | Let getExecutorId be a pure interface. | Henning Baldersheim | 2020-06-26 | 1 | -2/+2 |
| | |||||
* | Use ForegroundThreadExecutor in unit tests. | Geir Storli | 2020-06-25 | 5 | -17/+23 |
| | |||||
* | Run prepare step of two-phase puts in the shared executor instead of the ↵ | Geir Storli | 2020-06-25 | 1 | -14/+22 |
| | | | | | | | | attribute field executor. Benchmarking has shown that we get an uneven balance between the underlying threads when using the attribute field executor. This is a sequenced task executor, where each underlying thread only is allowed to handle a subset of the prepare tasks (based on the local document id). Using the shared executor should give an even balance. | ||||
* | Expose the shared executor via attribute manager interface. | Geir Storli | 2020-06-25 | 7 | -44/+55 |
| | |||||
* | Let the executor create an executor id to its liking. | Henning Baldersheim | 2020-06-25 | 2 | -2/+2 |
| | | | | Taske full advantage over strands instead of being limited to threads. | ||||
* | Add a rank feature that will use the distribution key to provide a globally ↵ | Henning Baldersheim | 2020-06-22 | 1 | -1/+6 |
| | | | | | | unique number. unique = (docId << 16 | distributionKey) | ||||
* | Merge pull request #13645 from ↵ | Geir Storli | 2020-06-19 | 1 | -4/+4 |
|\ | | | | | | | | | vespa-engine/geirst/multi-threaded-hnsw-indexing-flag Multi-threaded hnsw indexing flag | ||||
| * | Wire flag to enable multi-threaded indexing for a hnsw index. | Geir Storli | 2020-06-19 | 1 | -4/+4 |
| | | |||||
* | | Wire in nearest neighbor brute force limit. | Tor Egge | 2020-06-19 | 1 | -1/+3 |
|/ | |||||
* | Test more edge cases for two-phase put. | Geir Storli | 2020-06-17 | 1 | -16/+78 |
| | |||||
* | Remove default tensor spec in doctype builder. | Geir Storli | 2020-06-17 | 2 | -6/+7 |
| | |||||
* | Test that attribute writer can handle put in two phases. | Geir Storli | 2020-06-17 | 1 | -9/+86 |
| | |||||
* | Implement initial support for two-phase puts in attribute writer. | Geir Storli | 2020-06-17 | 1 | -0/+34 |
| | | | | This is only turned on for tensor attributes with a hnsw index that allows multi-threaded indexing. | ||||
* | Rename variable '_m' -> '_mgr'. | Geir Storli | 2020-06-16 | 1 | -16/+16 |
| | |||||
* | Simplify attribute writer tests by using mock attribute manager. | Geir Storli | 2020-06-16 | 2 | -25/+25 |
| | |||||
* | Revert "Two phase put in attribute writer" | Harald Musum | 2020-06-15 | 4 | -162/+50 |
| | |||||
* | Remove default tensor spec in doctype builder. | Geir Storli | 2020-06-15 | 2 | -6/+7 |
| | |||||
* | Test that attribute writer can handle put in two phases. | Geir Storli | 2020-06-15 | 1 | -9/+86 |
| | |||||
* | Implement initial support for two-phase puts in attribute writer. | Geir Storli | 2020-06-15 | 1 | -0/+34 |
| | | | | This is only turned on for tensor attributes with a hnsw index that allows multi-threaded indexing. | ||||
* | Rename variable '_m' -> '_mgr'. | Geir Storli | 2020-06-15 | 1 | -16/+16 |
| | |||||
* | Simplify attribute writer tests by using mock attribute manager. | Geir Storli | 2020-06-15 | 2 | -25/+25 |
| | |||||
* | Merge pull request #13542 from ↵ | Geir Storli | 2020-06-10 | 3 | -323/+302 |
|\ | | | | | | | | | vespa-engine/geirst/rewrite-attribute-tests-to-gtest Rewrite attribute tests to gtest | ||||
| * | Simplify. | Geir Storli | 2020-06-10 | 1 | -3/+1 |
| | | |||||
| * | Rewrite attribute (writer) tests to use gtest. | Geir Storli | 2020-06-10 | 2 | -316/+302 |
| | | |||||
| * | Stop using script to run the attribute test. | Geir Storli | 2020-06-10 | 2 | -7/+2 |
| | | | | | | | | DirectoryHandler is already used to handle creation and deletion of test folder. | ||||
* | | Adjust prepare restart strategy for differentiated replay operation costs. | Tor Egge | 2020-06-10 | 1 | -20/+44 |
|/ | |||||
* | use vector of UP as MultiSearch::Children | Arne Juul | 2020-06-04 | 1 | -14/+17 |
| | | | | | | * add helper class for constructing MultiSearch::Children (mostly for unit tests) * rewrite as needed to adapt | ||||
* | Propagate provider tombstone metadata to internal SPI GetResult | Tor Brede Vekterli | 2020-05-26 | 1 | -0/+1 |
| | |||||
* | Prepare to support matched-elements-only for arrays and weighted sets of ↵ | Geir Storli | 2020-05-20 | 1 | -10/+10 |
| | | | | | | | primitive types. This renames StructFieldMapper to MatchingElementsFields with added support for basic fields. Also some terminology is aligned. | ||||
* | - Handle more than 64k hits in the element vector. | Henning Baldersheim | 2020-05-18 | 1 | -6/+6 |
| | | | | | - Avoid computing all vectors in full separately and instead do an incremental inline merge with. - Also avoid requiring the searchiterator aspect on the wrappers. | ||||
* | Run initializer task with highest transient memory usage first. | Tor Egge | 2020-05-18 | 1 | -2/+28 |
| | |||||
* | Add unit test for AttributeUsageSamplerFunctor. | Tor Egge | 2020-05-13 | 2 | -0/+142 |
| | |||||
* | Calculate transient memory usage for load of attribute vector. | Tor Egge | 2020-05-12 | 1 | -0/+6 |
| | |||||
* | Verify that transient memory is zero when attribute to load doesn't have | Tor Egge | 2020-05-08 | 1 | -2/+10 |
| | | | | a posting list. | ||||
* | Add get_transient_memory_usage() method to AttibuteInitializer. | Tor Egge | 2020-05-08 | 1 | -0/+30 |
| | |||||
* | Add async update and followup on PR comments. | Henning Baldersheim | 2020-05-05 | 1 | -16/+16 |
| |