Revision 7103559f49b46b3287973045f741c0679e3e9e44 authored by Sagar Vemuri on 21 June 2018, 18:02:49 UTC, committed by Facebook Github Bot on 21 June 2018, 18:13:08 UTC
Summary: This PR extends the improvements in #3282 to also work when using Direct IO. We see **4.5X performance improvement** in seekrandom benchmark doing long range scans, when using direct reads, on flash. **Description:** This change improves the performance of iterators doing long range scans (e.g. big/full index or table scans in MyRocks) by using readahead and prefetching additional data on each disk IO, and storing in a local buffer. This prefetching is automatically enabled on noticing more than 2 IOs for the same table file during iteration. The readahead size starts with 8KB and is exponentially increased on each additional sequential IO, up to a max of 256 KB. This helps in cutting down the number of IOs needed to complete the range scan. **Implementation Details:** - Used `FilePrefetchBuffer` as the underlying buffer to store the readahead data. `FilePrefetchBuffer` can now take file_reader, readahead_size and max_readahead_size as input to the constructor, and automatically do readahead. - `FilePrefetchBuffer::TryReadFromCache` can now call `FilePrefetchBuffer::Prefetch` if readahead is enabled. - `AlignedBuffer` (which is the underlying store for `FilePrefetchBuffer`) now takes a few additional args in `AlignedBuffer::AllocateNewBuffer` to allow copying data from the old buffer. - Made sure not to re-read partial chunks of data that were already available in the buffer, from device again. - Fixed a couple of cases where `AlignedBuffer::cursize_` was not being properly kept up-to-date. **Constraints:** - Similar to #3282, this gets currently enabled only when ReadOptions.readahead_size = 0 (which is the default value). - Since the prefetched data is stored in a temporary buffer allocated on heap, this could increase the memory usage if you have many iterators doing long range scans simultaneously. - Enabled only for user reads, and disabled for compactions. Compaction reads are controlled by the options `use_direct_io_for_flush_and_compaction` and `compaction_readahead_size`, and the current feature takes precautions not to mess with them. **Benchmarks:** I used the same benchmark as used in #3282. Data fill: ``` TEST_TMPDIR=/data/users/$USER/benchmarks/iter ./db_bench -benchmarks=fillrandom -num=1000000000 -compression_type="none" -level_compaction_dynamic_level_bytes ``` Do a long range scan: Seekrandom with large number of nexts ``` TEST_TMPDIR=/data/users/$USER/benchmarks/iter ./db_bench -benchmarks=seekrandom -use_direct_reads -duration=60 -num=1000000000 -use_existing_db -seek_nexts=10000 -statistics -histogram ``` ``` Before: seekrandom : 37939.906 micros/op 26 ops/sec; 29.2 MB/s (1636 of 1999 found) With this change: seekrandom : 8527.720 micros/op 117 ops/sec; 129.7 MB/s (6530 of 7999 found) ``` ~4.5X perf improvement. Taken on an average of 3 runs. Closes https://github.com/facebook/rocksdb/pull/3884 Differential Revision: D8082143 Pulled By: sagar0 fbshipit-source-id: 4d7a8561cbac03478663713df4d31ad2620253bb
1 parent 524c6e6
File | Mode | Size |
---|---|---|
advisor | ||
dump | ||
rdb | ||
CMakeLists.txt | -rw-r--r-- | 544 bytes |
Dockerfile | -rw-r--r-- | 81 bytes |
auto_sanity_test.sh | -rwxr-xr-x | 2.7 KB |
benchmark.sh | -rwxr-xr-x | 16.4 KB |
benchmark_leveldb.sh | -rwxr-xr-x | 5.1 KB |
blob_dump.cc | -rw-r--r-- | 3.3 KB |
check_format_compatible.sh | -rwxr-xr-x | 4.5 KB |
db_bench.cc | -rw-r--r-- | 813 bytes |
db_bench_tool.cc | -rw-r--r-- | 197.2 KB |
db_bench_tool_test.cc | -rw-r--r-- | 9.5 KB |
db_crashtest.py | -rw-r--r-- | 13.2 KB |
db_repl_stress.cc | -rw-r--r-- | 4.5 KB |
db_sanity_test.cc | -rw-r--r-- | 8.4 KB |
db_stress.cc | -rw-r--r-- | 114.7 KB |
dbench_monitor | -rwxr-xr-x | 2.6 KB |
generate_random_db.sh | -rwxr-xr-x | 734 bytes |
ldb.cc | -rw-r--r-- | 572 bytes |
ldb_cmd.cc | -rw-r--r-- | 96.3 KB |
ldb_cmd_impl.h | -rw-r--r-- | 14.1 KB |
ldb_cmd_test.cc | -rw-r--r-- | 1.8 KB |
ldb_test.py | -rw-r--r-- | 24.4 KB |
ldb_tool.cc | -rw-r--r-- | 4.6 KB |
pflag | -rwxr-xr-x | 4.0 KB |
reduce_levels_test.cc | -rw-r--r-- | 5.2 KB |
regression_test.sh | -rwxr-xr-x | 15.8 KB |
report_lite_binary_size.sh | -rwxr-xr-x | 1.2 KB |
rocksdb_dump_test.sh | -rwxr-xr-x | 364 bytes |
run_flash_bench.sh | -rwxr-xr-x | 13.5 KB |
run_leveldb.sh | -rwxr-xr-x | 6.2 KB |
sample-dump.dmp | -rw-r--r-- | 100 bytes |
sst_dump.cc | -rw-r--r-- | 581 bytes |
sst_dump_test.cc | -rw-r--r-- | 6.4 KB |
sst_dump_tool.cc | -rw-r--r-- | 23.8 KB |
sst_dump_tool_imp.h | -rw-r--r-- | 2.8 KB |
verify_random_db.sh | -rwxr-xr-x | 1.0 KB |
write_stress.cc | -rw-r--r-- | 10.9 KB |
write_stress_runner.py | -rw-r--r-- | 2.3 KB |
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...