======================================================================================= DBT2 REPORT ======================================================================================= MACHINE: orisndriz01 RUN ID: 676 RUN DATE: 2010-10-28T20:02:31.970032 SERVER: drizzled VERSION: staging REVISION: 1890 COMMENT: 1890: Lee Bieber 2010-10-28 [merge] Merge Stewart - fix bug 587772: READ COMMITTED isolation level doesn't work (at least with InnoDB)
================================================================================================ TRENDING OVER LAST 5 runs Connections TPM % Diff from Avg Diff Min Max Avg STD ================================================================================================ 16 130.16 +0.69% 0.89 125.86 133.68 129.27 2.96 32 124.50 -1.82% -2.31 123.51 130.39 126.81 2.78 64 123.71 -3.21% -4.10 121.37 133.88 127.81 4.60 128 124.50 -1.88% -2.38 120.98 130.36 126.88 3.59 256 126.67 -1.77% -2.28 126.25 134.66 128.95 3.01 512 134.27 +3.56% 4.61 126.64 134.27 129.66 2.69 1024 128.79 -0.73% -0.94 127.43 132.31 129.73 2.07 ================================================================================================ TRENDING OVER LAST 20 runs Connections TPM % Diff from Avg Diff Min Max Avg STD ================================================================================================ 16 130.16 +1.65% 2.11 122.35 133.68 128.05 2.91 32 124.50 -2.15% -2.74 121.56 130.39 127.24 2.30 64 123.71 -3.81% -4.90 121.37 133.88 128.61 3.48 128 124.50 -2.80% -3.59 120.98 133.29 128.09 2.86 256 126.67 -0.92% -1.18 121.96 134.66 127.85 3.37 512 134.27 +5.34% 6.81 118.83 134.27 127.46 3.72 1024 128.79 +0.48% 0.62 121.37 132.31 128.17 2.81 ================================================================================================ TRENDING OVER ALL runs Connections TPM % Diff from Avg Diff Min Max Avg STD ================================================================================================ 16 130.16 +2.38% 3.03 114.19 135.64 127.13 3.48 32 124.50 -2.17% -2.76 112.06 137.79 127.26 3.67 64 123.71 -2.66% -3.38 88.73 137.45 127.09 3.88 128 124.50 -2.03% -2.58 21.11 136.56 127.08 6.23 256 126.67 -0.49% -0.62 114.02 136.03 127.29 3.54 512 134.27 +5.59% 7.11 57.46 135.83 127.16 4.79 1024 128.79 +1.16% 1.48 88.73 135.83 127.31 3.84 ========================================================================================================================= FULL REVISION COMMENTARY: 1860.2.19: Stewart Smith 2010-10-27 fix HailDB doEndStatement() since it doesn't need to have commit(false) called twice 1860.2.18: Stewart Smith 2010-10-27 [merge] merge trunk 1860.2.17: Stewart Smith 2010-10-26 add missing engine_states.cc 1860.2.16: Stewart Smith 2010-10-26 add missing cursor_states.cc 1860.2.15: Stewart Smith 2010-10-26 for Innobase, close read view in commit/rollback that happens for statement, not in endStatement() as we may have to roll back the statement 1860.2.14: Stewart Smith 2010-10-26 in Storage Engine API Tester: add cursor_states_to_dot utility 1860.2.13: Stewart Smith 2010-10-25 Storage Engine API Tester create_table test should clean up after itself 1860.2.12: Stewart Smith 2010-10-25 Add test to Storage Engine API Tester for rolling back a INSERT statement that goes awry... or at least that we get some state transitions correct. 1860.2.11: Stewart Smith 2010-10-25 add small noinst program to generate a dot graph of the Storage Engine API states from the transition multimap 1860.2.10: Stewart Smith 2010-10-25 remove the 'expected state' part of SEAPITester. The state transitions do this fine for us 1860.2.9: Stewart Smith 2010-10-25 endStatement() not being called at end of statement, breaking READ COMMITTED isolation level in innobase plugin 1860.2.8: Stewart Smith 2010-10-25 update state transitions in Storage Engine API tester to match reality (well, mostly). The series of calls is rather screwy. also implement Cursor states and enough to run a simple SELECT statement. 1860.2.7: Stewart Smith 2010-10-25 add test for READ COMMITTED isolation level 1860.2.6: Stewart Smith 2010-10-20 PBXT no longer needs to check that doStartTransation isn't being called twice (although we keep an assert, just to be safe 1860.2.5: Stewart Smith 2010-10-20 In TransactionalStorageEngine::startTransaction(), we were calling transaction_services.registerResourceForTransaction() which also called start transaction on the engine... the next call in TransactionalStorageEngine::startTransaction(). Fix it so hat registering resource for the txn only registers it. 1860.2.4: Stewart Smith 2010-10-20 complete test for a very simple empty transaction in storage_engine_api_tester 1860.2.3: Stewart Smith 2010-10-20 In StorageEngine API, transaction can be empty: BEGIN->COMMIT or BEGIN->ROLLBACK. Allow that in storage_engine_api_tester. 1860.2.2: Stewart Smith 2010-10-20 add initial implementation of storage_engine_api_tester. It's a storage engine with a bunch of asserts on state transations to test that the upper layer is calling the StorageEngine API correctly. 1860.2.1: Stewart Smith 2010-10-19 add test for READ_COMMITTED isolation level (currently fails with Innobase plugin at least) _______________________________________________ Mailing list: https://launchpad.net/~drizzle-benchmark Post to : [email protected] Unsubscribe : https://launchpad.net/~drizzle-benchmark More help : https://help.launchpad.net/ListHelp
