summaryrefslogtreecommitdiffstats
path: root/include
Commit message (Collapse)AuthorAgeFilesLines
* Properly implemented support for iteratively decomposable problemsDouglas Rumbaugh2024-04-199-64/+123
|
* Fixed some benchmarking bugsDouglas Rumbaugh2024-04-191-1/+1
|
* Removed debug print statementsDouglas Rumbaugh2024-04-192-4/+0
|
* PGM Shard: Fully disabled bloom filterDouglas Rumbaugh2024-04-191-1/+0
|
* More trie baseline testsDouglas Rumbaugh2024-04-191-0/+199
|
* Benchmark updatesDouglas Rumbaugh2024-04-199-28/+22
|
* Updated FSTrie to use const char * instead of std::stringDouglas Rumbaugh2024-04-152-13/+22
| | | | Note: this requires the caller to manage the memory of the strings
* Missed file from last commitDouglas Rumbaugh2024-04-151-1/+1
|
* stuffDouglas Rumbaugh2024-04-111-1/+1
|
* Updated FSTrie benchmark and some minor fixesDouglas B. Rumbaugh2024-03-262-2/+4
|
* Updates for build on OpenBSDDouglas B. Rumbaugh2024-03-254-3/+9
| | | | | | | | | Necessary updates to get the codebase building under OpenBSD 7.5 with clang. This is a minimal set of changes to get building to work, which includes disabling several things that aren't directly compatable. More work will be necessary to get full functionality. In particular, Triespline, PGM, and the reference M-tree do not currently build on OpenBSD with clang due to GNU dependencies or other gcc specific features.
* PointLookup: added a point lookup query for unique indexes, and some testsDouglas Rumbaugh2024-03-222-0/+123
|
* Record.h: Fixed wrapped record conceptDouglas Rumbaugh2024-03-221-3/+3
|
* PGM.h: fixed an out of bounds array access on point lookup misses.Douglas Rumbaugh2024-03-221-3/+21
|
* MutableBuffer: added visibility flag to records and refactored timestampDouglas Rumbaugh2024-03-221-1/+5
|
* FSTrie testing and debuggingDouglas Rumbaugh2024-03-221-101/+20
|
* Record.h: Removed manual constructor and adjusted wrapped header fieldsDouglas Rumbaugh2024-03-221-10/+26
|
* Disabled lookahead for paper revisionDouglas Rumbaugh2024-03-221-1/+1
|
* Updates/progress towards succinct trie supportDouglas Rumbaugh2024-03-203-15/+41
|
* FST Shard w/ testsDouglas Rumbaugh2024-03-191-0/+266
| | | | | | | | | Needs some debugging--some methods currently fail within the library itself. The build system doesn't currently build the FST library. To compile, you'll first need to manually build it, and then place the libFST.so file in your LIBRARY_PATH and LD_LIBRARY_PATH.
* VPTree: precalculate distances to make construction more efficientDouglas B. Rumbaugh2024-02-211-19/+30
|
* ExtensionStructure: first basic test of lookahead task stealingDouglas Rumbaugh2024-02-122-1/+14
|
* ExtensionStructure: Added simulated reconstruction lookaheadDouglas Rumbaugh2024-02-121-28/+88
| | | | | The reconstruction task procedure can now simulate future reconstructions to a specified depth.
* MutableBuffer: Allow hwm to equal lwmDouglas Rumbaugh2024-02-121-1/+1
| | | | | The high watermark and low watermark can now be equal, to allow for blocking reconstruction without requiring odd buffer sizes.
* Refactored Reconstruction TasksDouglas Rumbaugh2024-02-125-20/+77
| | | | | Added a ReconVector type to make it easier to do load balancing by shifting tasks around, and clean up a few interfaces.
* Added structure state vector w/ scratch version for reconstructionDouglas Rumbaugh2024-02-121-55/+89
| | | | | | | | | | | | | This approach should allow us to "simulate" a reconstruction to monitor the future state of the structure. The idea being that we can then add pre-emptive reconstructions to load balance and further smooth the tail latency curve. If a given reconstruction is significantly smaller than the next one will be, we can move some of the next one's work preemptively into the current one. The next phase is to do the simulation within the scratch_vector and then do a second pass examining the state of that reconstruction. In principle, we could look arbitrarily far ahead using this technique.
* Removed centralized version structureDouglas Rumbaugh2024-02-092-35/+8
|
* Framework: Fixed a bug where tagged deletes didn't release the epochDouglas Rumbaugh2024-02-091-2/+17
|
* Comment updates/fixesDouglas Rumbaugh2024-02-0923-103/+223
|
* Updated VPTree to new shard/query interfacesDouglas Rumbaugh2024-02-084-235/+228
|
* Added compiler fence to block reorderingDouglas Rumbaugh2024-02-081-0/+6
| | | | I'm reasonably certain that this is a compiler bug...
* Fixed a slight synchronization bug in Epoch retirement "properly"Douglas Rumbaugh2024-02-081-10/+27
|
* Cleaned up shard files (except VPTree)Douglas Rumbaugh2024-02-077-529/+357
| | | | | | Cleaned up shard implementations, fixed a few bugs, and set up some tests. There's still some work to be done in creating tests for the weighted sampling operations for the alias and aug btree shards.
* Fully realized shard concept interfaceDouglas Rumbaugh2024-02-0714-40/+33
|
* Fully implemented Query concept and adjusted queries to use itDouglas Rumbaugh2024-02-0711-52/+44
|
* BufferView: Adjusted BV to avoid repeated modulus operationsDouglas Rumbaugh2024-02-052-15/+48
|
* Set up tombstone deletes properlyDouglas Rumbaugh2024-02-054-26/+32
|
* temporary hack to get workingDouglas Rumbaugh2024-01-311-6/+12
|
* Adjusted epoch transition methodologyDouglas Rumbaugh2024-01-313-152/+120
|
* More locking!Douglas Rumbaugh2024-01-311-0/+2
|
* Updated insert query throughput to use IRS queriesDouglas Rumbaugh2024-01-311-7/+11
|
* Possibly fixed epoch retirement sync errorDouglas Rumbaugh2024-01-311-1/+5
|
* Insert throughput benchmarkDouglas Rumbaugh2024-01-311-1/+1
|
* Query/Insert intermix benchmarksDouglas Rumbaugh2024-01-311-3/+4
|
* Temporary thread affinity for reconstructionDouglas Rumbaugh2024-01-312-1/+33
|
* TrieSpline + testsDouglas Rumbaugh2024-01-301-61/+90
|
* Benchmarking updatesDouglas Rumbaugh2024-01-301-6/+23
|
* DynamicExtension: Fixed reconstruction trigger data raceDouglas Rumbaugh2024-01-241-4/+7
| | | | | Tweak the reconstruction trigger code to ensure that multiple reconstructions won't be triggered at the same time.
* Adjusted recon_benchmark and properly shutdown FIFOSchedulerDouglas Rumbaugh2024-01-221-0/+1
|
* FIFOScheduler: added automake wakeupDouglas Rumbaugh2024-01-221-1/+12
| | | | | | | | | | | | Sometimes, when the max thread count is exceeded, it is possible for the scheduler to lock up. This is because the scheduler is only run when a new job is put into the queue, and so it is possible for a job to be blocked by thread limitations and be left sitting in the queue. If the main program is waiting on this job to finish before scheduling a new one, then the system deadlocks. I added a second background thread to the scheduler that wakes the scheduler up every 20us to resolve this and prevent these deadlocks.