Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Add attribute tokens dfw. | Tor Egge | 2023-10-30 | 3 | -0/+14 |
| | |||||
* | - Reduce peak memory usage for unit tests from 5G to 0.5G. | Henning Baldersheim | 2023-10-27 | 1 | -7/+8 |
| | |||||
* | Use the entire docid range as approx hits with non-btree dictionary. | Geir Storli | 2023-10-26 | 1 | -1/+3 |
| | | | | | This fixes a bug introduced in the improved modeling of match strategies to use in numeric range search. | ||||
* | Simplify after review feedback. | Geir Storli | 2023-10-26 | 1 | -9/+7 |
| | |||||
* | Improve modelling of match strategies to use in numeric range search. | Geir Storli | 2023-10-26 | 3 | -113/+98 |
| | | | | | | | | This should improve the performance by choosing the strategy that is most optimal in different scenarios: lookup-based filter matching vs posting lists merging. The modelling is based on results from the range search performance test: https://github.com/vespa-engine/system-test/blob/master/tests/performance/range_search/. | ||||
* | Revert "Improve modelling of match strategies to use in numeric range search." | Tor Egge | 2023-10-19 | 3 | -96/+113 |
| | |||||
* | Simplify after review feedback. | Geir Storli | 2023-10-19 | 1 | -9/+7 |
| | |||||
* | Improve modelling of match strategies to use in numeric range search. | Geir Storli | 2023-10-19 | 3 | -113/+98 |
| | | | | | | | | This should improve the performance by choosing the strategy that is most optimal in different scenarios: lookup-based filter matching vs posting lists merging. The modelling is based on results from the range search performance test: https://github.com/vespa-engine/system-test/blob/master/tests/performance/range_search/. | ||||
* | Merge pull request #29005 from vespa-engine/toregge/remove-obsolete-todo | Henning Baldersheim | 2023-10-18 | 1 | -1/+0 |
|\ | | | | | Remove obsolete TODO. | ||||
| * | Remove obsolete TODO. | Tor Egge | 2023-10-18 | 1 | -1/+0 |
| | | |||||
* | | getBloatCount => getErasedCount | Henning Baldersheim | 2023-10-18 | 3 | -4/+4 |
| | | |||||
* | | - Modifications are guarded by external lock. | Henning Baldersheim | 2023-10-17 | 4 | -26/+31 |
|/ | | | | - Use relaxed atomics to ensure reader visibility. | ||||
* | Merge pull request #28950 from ↵ | Henning Baldersheim | 2023-10-17 | 1 | -0/+56 |
|\ | | | | | | | | | vespa-engine/toregge/add-info-about-evolution-of-disk-index-dictionary-format Add some info about evolution of disk index dictionary format. | ||||
| * | Add some info about evolution of disk index dictionary format. | Tor Egge | 2023-10-16 | 1 | -0/+56 |
| | | |||||
* | | fix some trailing whitespace | Arne Juul | 2023-10-16 | 5 | -10/+10 |
|/ | |||||
* | Move more checks to TokenExtractor. | Tor Egge | 2023-10-12 | 7 | -90/+324 |
| | |||||
* | Avoid reflection by providing explicit construction | Henning Baldersheim | 2023-10-11 | 64 | -265/+125 |
| | |||||
* | - Use a Supplier<Identifable> to construct objects from id. | Henning Baldersheim | 2023-10-11 | 21 | -48/+37 |
| | | | | | - Provide lambda for default constructing frequently used objects to avoid reflection. | ||||
* | Pass empty vector to TokenExtractor::extract. | Tor Egge | 2023-10-10 | 2 | -1/+1 |
| | |||||
* | Consolidate extraction of tokens from annotated string field value. | Tor Egge | 2023-10-10 | 8 | -87/+143 |
| | |||||
* | Merge pull request #28836 from vespa-engine/bratseth/copyright-2 | Jon Bratseth | 2023-10-10 | 2457 | -2472/+2472 |
|\ | | | | | Update copyright | ||||
| * | Correct copyright headers | Jon Bratseth | 2023-10-09 | 14 | -14/+14 |
| | | |||||
| * | Update copyright | Jon Bratseth | 2023-10-09 | 2456 | -2458/+2458 |
| | | |||||
* | | Rename id -> localChunkId | Henning Baldersheim | 2023-10-09 | 2 | -4/+4 |
| | | |||||
* | | Add test for BucketIndexStore and the iterator. | Henning Baldersheim | 2023-10-09 | 3 | -10/+44 |
| | | |||||
* | | Factor out the lid to bucket mapping. | Henning Baldersheim | 2023-10-09 | 2 | -65/+99 |
| | | |||||
* | | Explicit destructor. | Henning Baldersheim | 2023-10-09 | 2 | -5/+7 |
| | | |||||
* | | - Use a single store for mapping lid to its data that are split into ↵ | Henning Baldersheim | 2023-10-06 | 5 | -78/+149 |
|/ | | | | | | partitions and chunks. - This enable memory to be released after compaction is done. | ||||
* | - Avoid holding a bucketizer guard. Just get it everytime you need it. | Henning Baldersheim | 2023-10-05 | 2 | -25/+3 |
| | | | | | - Max hold time is often above 2-3 seconds. This makes it very likely that a sudden buildup might add l ot of memory to onhold. | ||||
* | Use ConstBufferRef and add some noexcept | Henning Baldersheim | 2023-10-05 | 16 | -76/+79 |
| | |||||
* | Merge pull request #28801 from ↵ | Henning Baldersheim | 2023-10-05 | 2 | -0/+6 |
|\ | | | | | | | | | vespa-engine/balder/disable-cache-for-removed-subdb Disable cache for removed only docsubdb. | ||||
| * | Add test for disabling of cache in removed db | Henning Baldersheim | 2023-10-05 | 2 | -0/+5 |
| | | |||||
| * | Disable cache for removed only docsubdb. | Henning Baldersheim | 2023-10-05 | 1 | -0/+1 |
| | | |||||
* | | Merge pull request #28800 from ↵ | Henning Baldersheim | 2023-10-05 | 2 | -6/+6 |
|\ \ | |/ |/| | | | | | vespa-engine/balder/reduce-max-number-of-lids-2-8m - Reduce max lids per file and max file size to 4M and 256M during un… | ||||
| * | - Reduce max lids per file and max file size to 4M and 256M during unit testing. | Henning Baldersheim | 2023-10-05 | 2 | -6/+6 |
| | | | | | | | | - Reduce max lids from 40M to 8M as default configuration. | ||||
* | | Merge branch 'master' into balder/refactor-for-clarity | Henning Baldersheim | 2023-10-05 | 4 | -20/+35 |
|\ \ | |||||
| * | | - Instead of keeping a map of bucketId => lids, just append everything to a ↵ | Henning Baldersheim | 2023-10-04 | 4 | -21/+36 |
| |/ | | | | | | | | | | | | | vector and sort when complete. - This significantly improves memory usage during compaction. Instead of many heap allocations - You now get fewer mmapped allocations that are dropped when done. | ||||
* / | - Number of partitions is fixed compile time => use std::array. | Henning Baldersheim | 2023-10-05 | 4 | -22/+25 |
|/ | | | | - Use unique_ptr on outer object instead of unique_ptr on multiple non-movable inner objects. | ||||
* | GC unused include | Henning Baldersheim | 2023-10-04 | 1 | -2/+0 |
| | |||||
* | Process idx file in streaming fashion instead of first reading all and then ↵ | Henning Baldersheim | 2023-10-04 | 2 | -73/+48 |
| | | | | process. | ||||
* | GC unused and non computed return value. | Henning Baldersheim | 2023-10-04 | 4 | -46/+53 |
| | | | | Refactor to prepare for streaming read. | ||||
* | Use large allocator and control size of TmpChunkMeta. | Henning Baldersheim | 2023-10-04 | 1 | -1/+2 |
| | |||||
* | Merge pull request #28776 from ↵ | Tor Egge | 2023-10-03 | 1 | -3/+4 |
|\ | | | | | | | | | vespa-engine/toregge/avoid-unaligned-read-while-decoding-serialized-query-stack-dump Avoid unaligned read while decoding serialized query stack dump. | ||||
| * | Avoid unaligned read while decoding serialized query stack dump. | Tor Egge | 2023-10-03 | 1 | -3/+4 |
| | | |||||
* | | Merge pull request #28773 from ↵ | Henning Baldersheim | 2023-10-03 | 3 | -5/+6 |
|\ \ | |/ |/| | | | | | vespa-engine/geirst/dfa-table-as-default-fuzzy-matching-algorithm Use DfaTable as default fuzzy matching algorithm for maxEditDistance … | ||||
| * | Use DfaTable as default fuzzy matching algorithm for maxEditDistance <= 2. | Geir Storli | 2023-10-03 | 3 | -5/+6 |
| | | |||||
* | | Prevent eternal loop if bit vectors are shorter than docid limit | Henning Baldersheim | 2023-10-03 | 3 | -8/+8 |
| | | |||||
* | | Add disabled test to prove eternal loop. | Henning Baldersheim | 2023-10-03 | 1 | -4/+35 |
| | | |||||
* | | Add test counting seeks | Henning Baldersheim | 2023-10-03 | 1 | -0/+16 |
| | | |||||
* | | Refactor test | Henning Baldersheim | 2023-10-03 | 1 | -127/+90 |
|/ |