summaryrefslogtreecommitdiffstats
path: root/include/framework
Commit message (Collapse)AuthorAgeFilesLines
* Beginnings of per-level cost modelingDouglas Rumbaugh2025-09-173-4/+25
| | | | | | | | | | | | The total time required for each shard construction is now measured, and hooks have been added to InternalLevel to support the creation and use of per-level cost models. The appropriate calls to these functions are now made during reconstruction, but the scheduling process doesn't yet call them, nor are the models actually implemented.
* Per record cost estimation progressDouglas B. Rumbaugh2025-08-141-6/+5
|
* Merge branch 'new-concurrency' of ↵Douglas Rumbaugh2025-05-291-0/+1
|\ | | | | | | github.com:dbrumbaugh/dynamic-extension-working into new-concurrency
| * Added selectivity sweep benchmarkDouglas Rumbaugh2025-05-251-1/+1
| |
* | Stall rate benchmarkingDouglas Rumbaugh2025-05-291-1/+3
|/
* Updates for tail latency stall testingDouglas Rumbaugh2025-05-252-2/+86
|
* Stats tweaks and KNN benchmarkDouglas Rumbaugh2025-04-303-1/+12
|
* Improved statistics calculationDouglas Rumbaugh2025-04-253-83/+116
|
* Scheduler statistics tracking updateDouglas Rumbaugh2025-04-255-32/+115
| | | | | | | | | The current scheme is really inefficient in terms of retreival of the results, but keeps the critical path mostly clear. It's probably worth it to do a more organized tracking of the data as it comes in, to avoid an n^2 statistics generation step at the end.
* switched k to a doubleDouglas Rumbaugh2025-04-173-9/+12
|
* DynamicExtension.h: Updated function docstringsDouglas Rumbaugh2025-04-101-23/+26
|
* DynamicExtension.h: Added the insertion rate limiterDouglas Rumbaugh2025-04-101-52/+98
| | | | | | The exact determination of *how* to use this mechanism will be contigent on some of the math I'm working on at the moment.
* BSM bugfixesDouglas Rumbaugh2025-04-081-12/+7
|
* Implemented the legacy policies and fixed a few bugsDouglas Rumbaugh2025-04-076-81/+109
|
* working commit (temporary progress, doesn't build)Douglas Rumbaugh2025-04-057-34/+57
|
* LockManager.h: adjusted unlock procedureDouglas Rumbaugh2025-03-031-1/+2
| | | | | There was a race condition. Now there isn't... or at least there's one less.
* Fixed a few concurrency bugsDouglas Rumbaugh2025-03-039-196/+303
|
* Query Preemption: still has one or two bugs, but mostly worksDouglas Rumbaugh2025-02-244-13/+51
|
* Fixed a few benchmark bugsDouglas Rumbaugh2025-02-171-1/+6
|
* Multi-threaded mixed workload benchmarkDouglas Rumbaugh2025-02-171-10/+1
|
* Fixed legacy tiering policyDouglas Rumbaugh2025-02-173-10/+16
|
* Parallel background reconstructions appear to be working!Douglas Rumbaugh2025-02-145-84/+48
|
* progress towards resolving asynch structure mergesDouglas Rumbaugh2025-02-134-138/+110
|
* more updatesDouglas Rumbaugh2025-02-112-16/+28
|
* Progress: began adding parallel merging and locking of levelsDouglas Rumbaugh2025-02-1114-85/+179
|
* Tiering: adjusted layout policy to properly handle L0Douglas Rumbaugh2025-02-061-1/+1
|
* Background compaction stuffDouglas Rumbaugh2025-02-0610-84/+333
|
* Bug fixesDouglas Rumbaugh2025-02-042-3/+3
|
* More updatesDouglas Rumbaugh2025-01-316-33/+82
|
* progressDouglas Rumbaugh2025-01-2711-215/+196
|
* some progressDouglas Rumbaugh2025-01-243-74/+33
|
* Finished rename of Epoch to Version + imported changes to psudb-commonDouglas Rumbaugh2025-01-224-7/+176
|
* ProgressDouglas Rumbaugh2025-01-1711-345/+254
|
* Additional layout policies + more flexibility in buffer flushingDouglas Rumbaugh2025-01-165-29/+229
|
* More updatesDouglas B. Rumbaugh2024-12-223-6/+6
| | | | | | | BSM is currently broken--I'll need to think a bit about how best to actually implement this in the current framework (if I even want to port it over) because it doesn't use the same flushing mechanism as most other approaches.
* Continued developmentDouglas B. Rumbaugh2024-12-2210-121/+162
|
* Began overhauling reconstruction mechanismDouglas B. Rumbaugh2024-12-2210-707/+434
|
* Adjusted query result interfacesHEADmasterDouglas B. Rumbaugh2024-12-063-115/+92
| | | | | | | Now, the vector<> is part of the user-defined type, not required by the framework. This should allow for more flexibility in either using alternative containers, or for more sensible implementations of queries with single value results (like range count).
* Warning fixesDouglas B. Rumbaugh2024-12-062-5/+6
|
* Interface update (#5)Douglas B. Rumbaugh2024-12-0615-2369/+2440
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Query Interface Adjustments/Refactoring Began the process of adjusting the query interface (and also the shard interface, to a lesser degree) to better accommodate the user. In particular the following changes have been made, 1. The number of necessary template arguments for the query type has been drastically reduced, while also removing the void pointers and manual delete functions from the interface. This was accomplished by requiring many of the sub-types associated with a query (parameters, etc.) to be nested inside the main query class, and by forcing the SHARD type to expose its associated record type. 2. User-defined query return types are now supported. Queries no longer are required to return strictly sets of records. Instead, the query now has LocalResultType and ResultType template parameters (which can be defaulted using a typedef in the Query type itself), allowing much more flexibility. Note that, at least for the short term, the LocalResultType must still expose the same is_deleted/is_tombstone interface as a Wrapped<R> used to, as this is currently needed for delete filtering. A better approach to this is, hopefully, forthcoming. 3. Updated the ISAMTree.h shard and rangequery.h query to use the new interfaces, and adjusted the associated unit tests as well. 4. Dropped the unnecessary "get_data()" function from the ShardInterface concept. 5. Dropped the need to specify a record type in the ShardInterface concept. This is now handled using a required Shard::RECORD member of the Shard class itself, which should expose the name of the record type. * Updates to framework to support new Query/Shard interfaces Pretty extensive adjustments to the framework, particularly to the templates themselves, along with some type-renaming work, to support the new query and shard interfaces. Adjusted the external query interface to take an rvalue reference, rather than a pointer, to the query parameters. * Removed framework-level delete filtering This was causing some issues with the new query interface, and should probably be reworked anyway, so I'm temporarily (TM) removing the feature. * Updated benchmarks + remaining code for new interface
* MTree structure sizeDouglas Rumbaugh2024-05-101-1/+1
|
* Fixed arithmetic bugDouglas Rumbaugh2024-05-091-1/+1
|
* TSParmsweep: enabled forcing a full buffer scanDouglas Rumbaugh2024-05-081-1/+5
|
* Added a Bentley-Saxe layout policyDouglas Rumbaugh2024-05-014-8/+95
|
* Properly implemented support for iteratively decomposable problemsDouglas Rumbaugh2024-04-192-20/+26
|
* Removed debug print statementsDouglas Rumbaugh2024-04-192-4/+0
|
* Benchmark updatesDouglas Rumbaugh2024-04-192-1/+5
|
* Updated FSTrie to use const char * instead of std::stringDouglas Rumbaugh2024-04-151-0/+17
| | | | Note: this requires the caller to manage the memory of the strings
* Missed file from last commitDouglas Rumbaugh2024-04-151-1/+1
|
* Updates for build on OpenBSDDouglas B. Rumbaugh2024-03-252-1/+7
| | | | | | | | | 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.