aboutsummaryrefslogtreecommitdiffstats
path: root/searchlib/src/vespa/searchlib/tensor/dense_tensor_attribute.cpp
Commit message (Expand)AuthorAgeFilesLines
* - Optimize distance calculation for tensors with single dense subspace.Henning Baldersheim2024-04-161-2/+2
* Update copyrightJon Bratseth2023-10-091-1/+1
* Relax validation of nearest neighbor searcher (java).Tor Egge2023-01-161-19/+0
* Setup hnsw index for mixed tensor types.Tor Egge2022-11-231-8/+1
* Move prepare_set_tensor() and complete_set_tensor() to TensorAttribute.Tor Egge2022-11-101-39/+1
* Add search::tensor::SubspaceType.Tor Egge2022-11-041-5/+1
* Pass VectorBundle to NearestNeighborIndex::prepare_add_document member function.Tor Egge2022-11-041-1/+6
* Add search::tensor::DocVectorAccess::get_vectors() member function.Tor Egge2022-11-031-0/+7
* Pass subspace to DocVectorAccess::get_vector member function.Tor Egge2022-11-021-2/+2
* Use member functions in TensorAttribute.Tor Egge2022-11-011-111/+0
* Consolidate tensor attribute loaders.Tor Egge2022-11-011-267/+7
* Consolidate tensor attribute savers.Tor Egge2022-10-311-25/+3
* Move tensor attribute constants to separate header file.Tor Egge2022-10-281-3/+1
* Rename removeOldGenerations() -> reclaim_memory().Geir Storli2022-10-121-3/+3
* Rename onGenerationChange() -> before_inc_generation() and change parameter t...Geir Storli2022-10-121-5/+3
* Rename clearHoldLists() -> reclaim_all_memory().Geir Storli2022-10-121-1/+1
* Rename trimHoldLists() -> reclaim_memory().Geir Storli2022-10-121-1/+1
* Rename transferHoldLists() -> assign_generation().Geir Storli2022-10-121-1/+1
* Implement GenerationHolder in terms of the generic generation hold list.Geir Storli2022-10-101-1/+1
* Share code for loading and saving tensor attribute betweenTor Egge2022-10-071-5/+2
* Use vespalib::datastore::CompactionContext when compacting tensor attributes.Tor Egge2022-10-061-7/+0
* Revert "Revert "Don't make changes to the hnsw index when the inserted tensor...Geir Storli2022-06-221-4/+22
* Revert "Don't make changes to the hnsw index when the inserted tensor is unch...Geir Storli2022-06-131-22/+4
* Don't make changes to the nearest neighbor index (hnsw) when the inserted ten...Geir Storli2022-06-101-4/+22
* Include searchcommon/attribut/config.h yourself and prepare for hiding large ...Henning Baldersheim2022-05-251-0/+1
* Use AtomicEntryRef in tensor attribute.Tor Egge2022-03-211-7/+6
* Move ownership of memory allocator to AttributeVector.Tor Egge2022-02-111-12/+1
* Tag cpu category for tasks run in the proton shared executor.Geir Storli2022-01-281-2/+5
* Calculate need for compaction when updating stats.Tor Egge2021-12-091-1/+1
* Set AttributeHeader _fileName member when extracting tagsTor Egge2021-12-061-1/+1
* Update 2017 copyright notices.gjoranv2021-10-071-1/+1
* Change loading of nearest neighbor index to use direct I/O instead of mmapping.Geir Storli2021-09-151-35/+50
* Refactor to represent files with a generic header in a new class.Geir Storli2021-09-141-2/+2
* Reduce memory spike during loading of hnsw index by committing at regular int...Geir Storli2021-09-101-4/+17
* Report address space usage for components in tensor attributes.Geir Storli2021-08-201-0/+9
* Compact HNSW index when ratio of dead bytes / address space is too highTor Egge2021-08-181-0/+21
* Avoid starting a separate thread for completing index insert.Henning Baldersheim2021-08-131-34/+73
* Notify when _pending reaches zero.Henning Baldersheim2021-08-131-3/+6
* Refactor for readability and maintenance.Henning Baldersheim2021-08-131-29/+87
* Use the executor for the part that can be parallell when rebuilding index on ...Henning Baldersheim2021-08-131-7/+31
* Add an executor to the AttributeVector::load/onLoad interface so attributes c...Henning Baldersheim2021-08-121-1/+2
* swappable -> pagedHenning Baldersheim2021-08-121-1/+1
* A swappable attribute will use a file backed memory allocator.Henning Baldersheim2021-08-121-1/+12
* Only use tensor store memory usage when considering compaction.Tor Egge2021-06-011-0/+10
* Update searchlib/src/vespa/searchlib/tensor/dense_tensor_attribute.cpp Tor Egge2021-05-271-1/+1
* Use modulo instead of bitwise and.Tor Egge2021-05-271-1/+1
* Load of tensor attribute with nearest neighbor index enabled but noTor Egge2021-05-271-0/+3
* Remove experimental use of mmap file allocator for dense tensor data.Geir Storli2021-03-261-12/+1
* Rename get_memory_allocator to make_memory_allocator in mmap file allocatorTor Egge2021-02-151-1/+1
* Use mmap file allocator in dense tensor attribute if huge flag isTor Egge2021-02-151-1/+12