Mark Callaghan
Mark Callaghan
This is a feature request. Can integrated BlobDB respect the block cache size limits whether or not it shares the block cache? I started down this path because I was...
db_bench has flush, compact0 and compact1 commands to request flushing the memtable, compacting from L0->L1 and from L1->L2. It also has waitforcompaction to wait for compaction to finish. MyRocks has...
This is a feature request. I want optional response time constraints in db_bench so that outlier responses get ignored.
When running tools/benchmark.sh for universal compaction with a database much larger than memory I get unexpected out of disk space errors. The interesting details are: * compaction IO stats show...
From tests I run with RocksDB 8.7 and 8.8 what I see with rareq-sz from iostat is many small reads (block sized) from compaction when compaction_readahead_size is larger than /sys/block/$device/queue/max_sectors_kb....
The request is to make it clear in fio output that page cache invalidation is in progress for tests that use buffered IO. It isn't clear to me in the...
I ran the Insert Benchmark on a server with 15 cores, 120G of RAM and 1.5T of SSD. For this server I have been loading 4B rows during the initial...
There are hard and soft limits on pending compaction bytes that can cause write stalls. It would be great to be able to monitor the value. See the header for...
This fixes https://github.com/facebook/mysql-5.6/issues/1336 This adds the my.cnf options: rocksdb_block_cache_numshardbits This option can be set so that RocksDB to fix the number of block cache shards. The default value is -1...
MyRocks 5.6.35 and 8.0.28 adds the unittest directory based on the value of WITH_EMBEDDED_SERVER. It should do this based on the value of WITH_UNIT_TESTS. See how storage/perfschema does this [for...