You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
rocksdb/table/block_fetcher.cc

405 lines
15 KiB

// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under both the GPLv2 (found in the
// COPYING file in the root directory) and Apache 2.0 License
// (found in the LICENSE.Apache file in the root directory).
//
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
#include "table/block_fetcher.h"
Fix stress test failure "Corruption: checksum mismatch" or "Iterator Diverged" with async_io enabled (#10032) Summary: In case of non sequential reads with `async_io`, `FilePRefetchBuffer::TryReadFromCacheAsync` can be called for previous blocks with `offset < bufs_[curr_].offset_` which wasn't handled correctly resulting wrong data being returned from buffer. Since `FilePRefetchBuffer::PrefetchAsync` can be called for any data block, it sets `prev_len_` to 0 indicating `FilePRefetchBuffer::TryReadFromCacheAsync` to go for the prefetching even though offset < bufs_[curr_].offset_ This is because async prefetching is always done in second buffer (to avoid mutex) even though curr_ is empty leading to offset < bufs_[curr_].offset_ in some cases. If prev_len_ is non zero then `TryReadFromCacheAsync` returns false if `offset < bufs_[curr_].offset_ && prev_len != 0` indicating reads are not sequential and previous call wasn't PrefetchAsync. - This PR also simplifies `FilePRefetchBuffer::TryReadFromCacheAsync` as it was getting complicated covering different scenarios based on `async_io` enabled/disabled. If `for_compaction` is set true, it now calls `FilePRefetchBufferTryReadFromCache` following synchronous flow as before. Its decided in BlockFetcher.cc Pull Request resolved: https://github.com/facebook/rocksdb/pull/10032 Test Plan: 1. export CRASH_TEST_EXT_ARGS=" --async_io=1" make crash_test -j completed successfully locally 2. make crash_test -j completed successfully locally 3. Reran CircleCi mini crashtest job 4 - 5 times. 4. Updated prefetch_test for more coverage. Reviewed By: anand1976 Differential Revision: D36579858 Pulled By: akankshamahajan15 fbshipit-source-id: 0c428d62b45e12e082a83acf533a5e37a584bedf
2 years ago
#include <cassert>
#include <cinttypes>
#include <string>
#include "logging/logging.h"
#include "memory/memory_allocator.h"
#include "monitoring/perf_context_imp.h"
Improve / clean up meta block code & integrity (#9163) Summary: * Checksums are now checked on meta blocks unless specifically suppressed or not applicable (e.g. plain table). (Was other way around.) This means a number of cases that were not checking checksums now are, including direct read TableProperties in Version::GetTableProperties (fixed in meta_blocks ReadTableProperties), reading any block from PersistentCache (fixed in BlockFetcher), read TableProperties in SstFileDumper (ldb/sst_dump/BackupEngine) before table reader open, maybe more. * For that to work, I moved the global_seqno+TableProperties checksum logic to the shared table/ code, because that is used by many utilies such as SstFileDumper. * Also for that to work, we have to know when we're dealing with a block that has a checksum (trailer), so added that capability to Footer based on magic number, and from there BlockFetcher. * Knowledge of trailer presence has also fixed a problem where other table formats were reading blocks including bytes for a non-existant trailer--and awkwardly kind-of not using them, e.g. no shared code checking checksums. (BlockFetcher compression type was populated incorrectly.) Now we only read what is needed. * Minimized code duplication and differing/incompatible/awkward abstractions in meta_blocks.{cc,h} (e.g. SeekTo in metaindex block without parsing block handle) * Moved some meta block handling code from table_properties*.* * Moved some code specific to block-based table from shared table/ code to BlockBasedTable class. The checksum stuff means we can't completely separate it, but things that don't need to be in shared table/ code should not be. * Use unique_ptr rather than raw ptr in more places. (Note: you can std::move from unique_ptr to shared_ptr.) Without enhancements to GetPropertiesOfAllTablesTest (see below), net reduction of roughly 100 lines of code. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9163 Test Plan: existing tests and * Enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to verify that checksums are now checked on direct read of table properties by TableCache (new test would fail before this change) * Also enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to test putting table properties under old meta name * Also generally enhanced that same test to actually test what it was supposed to be testing already, by kicking things out of table cache when we don't want them there. Reviewed By: ajkr, mrambacher Differential Revision: D32514757 Pulled By: pdillinger fbshipit-source-id: 507964b9311d186ae8d1131182290cbd97a99fa9
3 years ago
#include "rocksdb/compression_type.h"
#include "rocksdb/env.h"
#include "table/block_based/block.h"
#include "table/block_based/block_based_table_reader.h"
Improve / clean up meta block code & integrity (#9163) Summary: * Checksums are now checked on meta blocks unless specifically suppressed or not applicable (e.g. plain table). (Was other way around.) This means a number of cases that were not checking checksums now are, including direct read TableProperties in Version::GetTableProperties (fixed in meta_blocks ReadTableProperties), reading any block from PersistentCache (fixed in BlockFetcher), read TableProperties in SstFileDumper (ldb/sst_dump/BackupEngine) before table reader open, maybe more. * For that to work, I moved the global_seqno+TableProperties checksum logic to the shared table/ code, because that is used by many utilies such as SstFileDumper. * Also for that to work, we have to know when we're dealing with a block that has a checksum (trailer), so added that capability to Footer based on magic number, and from there BlockFetcher. * Knowledge of trailer presence has also fixed a problem where other table formats were reading blocks including bytes for a non-existant trailer--and awkwardly kind-of not using them, e.g. no shared code checking checksums. (BlockFetcher compression type was populated incorrectly.) Now we only read what is needed. * Minimized code duplication and differing/incompatible/awkward abstractions in meta_blocks.{cc,h} (e.g. SeekTo in metaindex block without parsing block handle) * Moved some meta block handling code from table_properties*.* * Moved some code specific to block-based table from shared table/ code to BlockBasedTable class. The checksum stuff means we can't completely separate it, but things that don't need to be in shared table/ code should not be. * Use unique_ptr rather than raw ptr in more places. (Note: you can std::move from unique_ptr to shared_ptr.) Without enhancements to GetPropertiesOfAllTablesTest (see below), net reduction of roughly 100 lines of code. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9163 Test Plan: existing tests and * Enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to verify that checksums are now checked on direct read of table properties by TableCache (new test would fail before this change) * Also enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to test putting table properties under old meta name * Also generally enhanced that same test to actually test what it was supposed to be testing already, by kicking things out of table cache when we don't want them there. Reviewed By: ajkr, mrambacher Differential Revision: D32514757 Pulled By: pdillinger fbshipit-source-id: 507964b9311d186ae8d1131182290cbd97a99fa9
3 years ago
#include "table/block_based/block_type.h"
#include "table/block_based/reader_common.h"
#include "table/format.h"
#include "table/persistent_cache_helper.h"
#include "util/compression.h"
#include "util/stop_watch.h"
namespace ROCKSDB_NAMESPACE {
Improve / clean up meta block code & integrity (#9163) Summary: * Checksums are now checked on meta blocks unless specifically suppressed or not applicable (e.g. plain table). (Was other way around.) This means a number of cases that were not checking checksums now are, including direct read TableProperties in Version::GetTableProperties (fixed in meta_blocks ReadTableProperties), reading any block from PersistentCache (fixed in BlockFetcher), read TableProperties in SstFileDumper (ldb/sst_dump/BackupEngine) before table reader open, maybe more. * For that to work, I moved the global_seqno+TableProperties checksum logic to the shared table/ code, because that is used by many utilies such as SstFileDumper. * Also for that to work, we have to know when we're dealing with a block that has a checksum (trailer), so added that capability to Footer based on magic number, and from there BlockFetcher. * Knowledge of trailer presence has also fixed a problem where other table formats were reading blocks including bytes for a non-existant trailer--and awkwardly kind-of not using them, e.g. no shared code checking checksums. (BlockFetcher compression type was populated incorrectly.) Now we only read what is needed. * Minimized code duplication and differing/incompatible/awkward abstractions in meta_blocks.{cc,h} (e.g. SeekTo in metaindex block without parsing block handle) * Moved some meta block handling code from table_properties*.* * Moved some code specific to block-based table from shared table/ code to BlockBasedTable class. The checksum stuff means we can't completely separate it, but things that don't need to be in shared table/ code should not be. * Use unique_ptr rather than raw ptr in more places. (Note: you can std::move from unique_ptr to shared_ptr.) Without enhancements to GetPropertiesOfAllTablesTest (see below), net reduction of roughly 100 lines of code. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9163 Test Plan: existing tests and * Enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to verify that checksums are now checked on direct read of table properties by TableCache (new test would fail before this change) * Also enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to test putting table properties under old meta name * Also generally enhanced that same test to actually test what it was supposed to be testing already, by kicking things out of table cache when we don't want them there. Reviewed By: ajkr, mrambacher Differential Revision: D32514757 Pulled By: pdillinger fbshipit-source-id: 507964b9311d186ae8d1131182290cbd97a99fa9
3 years ago
inline void BlockFetcher::ProcessTrailerIfPresent() {
if (footer_.GetBlockTrailerSize() > 0) {
assert(footer_.GetBlockTrailerSize() == BlockBasedTable::kBlockTrailerSize);
if (read_options_.verify_checksums) {
io_status_ = status_to_io_status(VerifyBlockChecksum(
footer_.checksum_type(), slice_.data(), block_size_,
file_->file_name(), handle_.offset()));
RecordTick(ioptions_.stats, BLOCK_CHECKSUM_COMPUTE_COUNT);
Improve / clean up meta block code & integrity (#9163) Summary: * Checksums are now checked on meta blocks unless specifically suppressed or not applicable (e.g. plain table). (Was other way around.) This means a number of cases that were not checking checksums now are, including direct read TableProperties in Version::GetTableProperties (fixed in meta_blocks ReadTableProperties), reading any block from PersistentCache (fixed in BlockFetcher), read TableProperties in SstFileDumper (ldb/sst_dump/BackupEngine) before table reader open, maybe more. * For that to work, I moved the global_seqno+TableProperties checksum logic to the shared table/ code, because that is used by many utilies such as SstFileDumper. * Also for that to work, we have to know when we're dealing with a block that has a checksum (trailer), so added that capability to Footer based on magic number, and from there BlockFetcher. * Knowledge of trailer presence has also fixed a problem where other table formats were reading blocks including bytes for a non-existant trailer--and awkwardly kind-of not using them, e.g. no shared code checking checksums. (BlockFetcher compression type was populated incorrectly.) Now we only read what is needed. * Minimized code duplication and differing/incompatible/awkward abstractions in meta_blocks.{cc,h} (e.g. SeekTo in metaindex block without parsing block handle) * Moved some meta block handling code from table_properties*.* * Moved some code specific to block-based table from shared table/ code to BlockBasedTable class. The checksum stuff means we can't completely separate it, but things that don't need to be in shared table/ code should not be. * Use unique_ptr rather than raw ptr in more places. (Note: you can std::move from unique_ptr to shared_ptr.) Without enhancements to GetPropertiesOfAllTablesTest (see below), net reduction of roughly 100 lines of code. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9163 Test Plan: existing tests and * Enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to verify that checksums are now checked on direct read of table properties by TableCache (new test would fail before this change) * Also enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to test putting table properties under old meta name * Also generally enhanced that same test to actually test what it was supposed to be testing already, by kicking things out of table cache when we don't want them there. Reviewed By: ajkr, mrambacher Differential Revision: D32514757 Pulled By: pdillinger fbshipit-source-id: 507964b9311d186ae8d1131182290cbd97a99fa9
3 years ago
}
compression_type_ =
BlockBasedTable::GetBlockCompressionType(slice_.data(), block_size_);
} else {
// E.g. plain table or cuckoo table
compression_type_ = kNoCompression;
}
}
inline bool BlockFetcher::TryGetUncompressBlockFromPersistentCache() {
if (cache_options_.persistent_cache &&
!cache_options_.persistent_cache->IsCompressed()) {
Status status = PersistentCacheHelper::LookupUncompressedPage(
cache_options_, handle_, contents_);
if (status.ok()) {
// uncompressed page is found for the block handle
return true;
} else {
// uncompressed page is not found
if (ioptions_.logger && !status.IsNotFound()) {
assert(!status.ok());
ROCKS_LOG_INFO(ioptions_.logger,
"Error reading from persistent cache. %s",
status.ToString().c_str());
}
}
}
return false;
}
inline bool BlockFetcher::TryGetFromPrefetchBuffer() {
if (prefetch_buffer_ != nullptr) {
IOOptions opts;
IOStatus io_s = file_->PrepareIOOptions(read_options_, opts);
Provide implementation to prefetch data asynchronously in FilePrefetchBuffer (#9674) Summary: In FilePrefetchBuffer if reads are sequential, after prefetching call ReadAsync API to prefetch data asynchronously so that in next prefetching data will be available. Data prefetched asynchronously will be readahead_size/2. It uses two buffers, one for synchronous prefetching and one for asynchronous. In case, the data is overlapping, the data is copied from both buffers to third buffer to make it continuous. This feature is under ReadOptions::async_io and is under experimental. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9674 Test Plan: 1. Add new unit tests 2. Run **db_stress** to make sure nothing crashes. - Normal prefetch without `async_io` ran successfully: ``` export CRASH_TEST_EXT_ARGS=" --async_io=0" make crash_test -j ``` 3. **Run Regressions**. i) Main branch without any change for normal prefetching with async_io disabled: ``` ./db_bench -db=/tmp/prefix_scan_prefetch_main -benchmarks="fillseq" -key_size=32 -value_size=512 -num=5000000 - use_direct_io_for_flush_and_compaction=true -target_file_size_base=16777216 ``` ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` ii) normal prefetching after changes with async_io disable: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_withchange -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 14:11:31 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_withchange] seekrandom : 471347.227 micros/op 2 ops/sec; 348.1 MB/s (255 of 255 found) ``` Reviewed By: anand1976 Differential Revision: D34731543 Pulled By: akankshamahajan15 fbshipit-source-id: 8e23aa93453d5fe3c672b9231ad582f60207937f
3 years ago
if (io_s.ok()) {
bool read_from_prefetch_buffer = false;
Fix stress test failure "Corruption: checksum mismatch" or "Iterator Diverged" with async_io enabled (#10032) Summary: In case of non sequential reads with `async_io`, `FilePRefetchBuffer::TryReadFromCacheAsync` can be called for previous blocks with `offset < bufs_[curr_].offset_` which wasn't handled correctly resulting wrong data being returned from buffer. Since `FilePRefetchBuffer::PrefetchAsync` can be called for any data block, it sets `prev_len_` to 0 indicating `FilePRefetchBuffer::TryReadFromCacheAsync` to go for the prefetching even though offset < bufs_[curr_].offset_ This is because async prefetching is always done in second buffer (to avoid mutex) even though curr_ is empty leading to offset < bufs_[curr_].offset_ in some cases. If prev_len_ is non zero then `TryReadFromCacheAsync` returns false if `offset < bufs_[curr_].offset_ && prev_len != 0` indicating reads are not sequential and previous call wasn't PrefetchAsync. - This PR also simplifies `FilePRefetchBuffer::TryReadFromCacheAsync` as it was getting complicated covering different scenarios based on `async_io` enabled/disabled. If `for_compaction` is set true, it now calls `FilePRefetchBufferTryReadFromCache` following synchronous flow as before. Its decided in BlockFetcher.cc Pull Request resolved: https://github.com/facebook/rocksdb/pull/10032 Test Plan: 1. export CRASH_TEST_EXT_ARGS=" --async_io=1" make crash_test -j completed successfully locally 2. make crash_test -j completed successfully locally 3. Reran CircleCi mini crashtest job 4 - 5 times. 4. Updated prefetch_test for more coverage. Reviewed By: anand1976 Differential Revision: D36579858 Pulled By: akankshamahajan15 fbshipit-source-id: 0c428d62b45e12e082a83acf533a5e37a584bedf
2 years ago
if (read_options_.async_io && !for_compaction_) {
Provide implementation to prefetch data asynchronously in FilePrefetchBuffer (#9674) Summary: In FilePrefetchBuffer if reads are sequential, after prefetching call ReadAsync API to prefetch data asynchronously so that in next prefetching data will be available. Data prefetched asynchronously will be readahead_size/2. It uses two buffers, one for synchronous prefetching and one for asynchronous. In case, the data is overlapping, the data is copied from both buffers to third buffer to make it continuous. This feature is under ReadOptions::async_io and is under experimental. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9674 Test Plan: 1. Add new unit tests 2. Run **db_stress** to make sure nothing crashes. - Normal prefetch without `async_io` ran successfully: ``` export CRASH_TEST_EXT_ARGS=" --async_io=0" make crash_test -j ``` 3. **Run Regressions**. i) Main branch without any change for normal prefetching with async_io disabled: ``` ./db_bench -db=/tmp/prefix_scan_prefetch_main -benchmarks="fillseq" -key_size=32 -value_size=512 -num=5000000 - use_direct_io_for_flush_and_compaction=true -target_file_size_base=16777216 ``` ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` ii) normal prefetching after changes with async_io disable: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_withchange -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 14:11:31 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_withchange] seekrandom : 471347.227 micros/op 2 ops/sec; 348.1 MB/s (255 of 255 found) ``` Reviewed By: anand1976 Differential Revision: D34731543 Pulled By: akankshamahajan15 fbshipit-source-id: 8e23aa93453d5fe3c672b9231ad582f60207937f
3 years ago
read_from_prefetch_buffer = prefetch_buffer_->TryReadFromCacheAsync(
Add rate limiter priority to ReadOptions (#9424) Summary: Users can set the priority for file reads associated with their operation by setting `ReadOptions::rate_limiter_priority` to something other than `Env::IO_TOTAL`. Rate limiting `VerifyChecksum()` and `VerifyFileChecksums()` is the motivation for this PR, so it also includes benchmarks and minor bug fixes to get that working. `RandomAccessFileReader::Read()` already had support for rate limiting compaction reads. I changed that rate limiting to be non-specific to compaction, but rather performed according to the passed in `Env::IOPriority`. Now the compaction read rate limiting is supported by setting `rate_limiter_priority = Env::IO_LOW` on its `ReadOptions`. There is no default value for the new `Env::IOPriority` parameter to `RandomAccessFileReader::Read()`. That means this PR goes through all callers (in some cases multiple layers up the call stack) to find a `ReadOptions` to provide the priority. There are TODOs for cases I believe it would be good to let user control the priority some day (e.g., file footer reads), and no TODO in cases I believe it doesn't matter (e.g., trace file reads). The API doc only lists the missing cases where a file read associated with a provided `ReadOptions` cannot be rate limited. For cases like file ingestion checksum calculation, there is no API to provide `ReadOptions` or `Env::IOPriority`, so I didn't count that as missing. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9424 Test Plan: - new unit tests - new benchmarks on ~50MB database with 1MB/s read rate limit and 100ms refill interval; verified with strace reads are chunked (at 0.1MB per chunk) and spaced roughly 100ms apart. - setup command: `./db_bench -benchmarks=fillrandom,compact -db=/tmp/testdb -target_file_size_base=1048576 -disable_auto_compactions=true -file_checksum=true` - benchmarks command: `strace -ttfe pread64 ./db_bench -benchmarks=verifychecksum,verifyfilechecksums -use_existing_db=true -db=/tmp/testdb -rate_limiter_bytes_per_sec=1048576 -rate_limit_bg_reads=1 -rate_limit_user_ops=true -file_checksum=true` - crash test using IO_USER priority on non-validation reads with https://github.com/facebook/rocksdb/issues/9567 reverted: `python3 tools/db_crashtest.py blackbox --max_key=1000000 --write_buffer_size=524288 --target_file_size_base=524288 --level_compaction_dynamic_level_bytes=true --duration=3600 --rate_limit_bg_reads=true --rate_limit_user_ops=true --rate_limiter_bytes_per_sec=10485760 --interval=10` Reviewed By: hx235 Differential Revision: D33747386 Pulled By: ajkr fbshipit-source-id: a2d985e97912fba8c54763798e04f006ccc56e0c
3 years ago
opts, file_, handle_.offset(), block_size_with_trailer_, &slice_,
Fix stress test failure "Corruption: checksum mismatch" or "Iterator Diverged" with async_io enabled (#10032) Summary: In case of non sequential reads with `async_io`, `FilePRefetchBuffer::TryReadFromCacheAsync` can be called for previous blocks with `offset < bufs_[curr_].offset_` which wasn't handled correctly resulting wrong data being returned from buffer. Since `FilePRefetchBuffer::PrefetchAsync` can be called for any data block, it sets `prev_len_` to 0 indicating `FilePRefetchBuffer::TryReadFromCacheAsync` to go for the prefetching even though offset < bufs_[curr_].offset_ This is because async prefetching is always done in second buffer (to avoid mutex) even though curr_ is empty leading to offset < bufs_[curr_].offset_ in some cases. If prev_len_ is non zero then `TryReadFromCacheAsync` returns false if `offset < bufs_[curr_].offset_ && prev_len != 0` indicating reads are not sequential and previous call wasn't PrefetchAsync. - This PR also simplifies `FilePRefetchBuffer::TryReadFromCacheAsync` as it was getting complicated covering different scenarios based on `async_io` enabled/disabled. If `for_compaction` is set true, it now calls `FilePRefetchBufferTryReadFromCache` following synchronous flow as before. Its decided in BlockFetcher.cc Pull Request resolved: https://github.com/facebook/rocksdb/pull/10032 Test Plan: 1. export CRASH_TEST_EXT_ARGS=" --async_io=1" make crash_test -j completed successfully locally 2. make crash_test -j completed successfully locally 3. Reran CircleCi mini crashtest job 4 - 5 times. 4. Updated prefetch_test for more coverage. Reviewed By: anand1976 Differential Revision: D36579858 Pulled By: akankshamahajan15 fbshipit-source-id: 0c428d62b45e12e082a83acf533a5e37a584bedf
2 years ago
&io_s, read_options_.rate_limiter_priority);
Provide implementation to prefetch data asynchronously in FilePrefetchBuffer (#9674) Summary: In FilePrefetchBuffer if reads are sequential, after prefetching call ReadAsync API to prefetch data asynchronously so that in next prefetching data will be available. Data prefetched asynchronously will be readahead_size/2. It uses two buffers, one for synchronous prefetching and one for asynchronous. In case, the data is overlapping, the data is copied from both buffers to third buffer to make it continuous. This feature is under ReadOptions::async_io and is under experimental. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9674 Test Plan: 1. Add new unit tests 2. Run **db_stress** to make sure nothing crashes. - Normal prefetch without `async_io` ran successfully: ``` export CRASH_TEST_EXT_ARGS=" --async_io=0" make crash_test -j ``` 3. **Run Regressions**. i) Main branch without any change for normal prefetching with async_io disabled: ``` ./db_bench -db=/tmp/prefix_scan_prefetch_main -benchmarks="fillseq" -key_size=32 -value_size=512 -num=5000000 - use_direct_io_for_flush_and_compaction=true -target_file_size_base=16777216 ``` ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` ii) normal prefetching after changes with async_io disable: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_withchange -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 14:11:31 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_withchange] seekrandom : 471347.227 micros/op 2 ops/sec; 348.1 MB/s (255 of 255 found) ``` Reviewed By: anand1976 Differential Revision: D34731543 Pulled By: akankshamahajan15 fbshipit-source-id: 8e23aa93453d5fe3c672b9231ad582f60207937f
3 years ago
} else {
read_from_prefetch_buffer = prefetch_buffer_->TryReadFromCache(
opts, file_, handle_.offset(), block_size_with_trailer_, &slice_,
&io_s, read_options_.rate_limiter_priority, for_compaction_);
}
if (read_from_prefetch_buffer) {
ProcessTrailerIfPresent();
if (!io_status_.ok()) {
return true;
}
got_from_prefetch_buffer_ = true;
used_buf_ = const_cast<char*>(slice_.data());
}
Provide implementation to prefetch data asynchronously in FilePrefetchBuffer (#9674) Summary: In FilePrefetchBuffer if reads are sequential, after prefetching call ReadAsync API to prefetch data asynchronously so that in next prefetching data will be available. Data prefetched asynchronously will be readahead_size/2. It uses two buffers, one for synchronous prefetching and one for asynchronous. In case, the data is overlapping, the data is copied from both buffers to third buffer to make it continuous. This feature is under ReadOptions::async_io and is under experimental. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9674 Test Plan: 1. Add new unit tests 2. Run **db_stress** to make sure nothing crashes. - Normal prefetch without `async_io` ran successfully: ``` export CRASH_TEST_EXT_ARGS=" --async_io=0" make crash_test -j ``` 3. **Run Regressions**. i) Main branch without any change for normal prefetching with async_io disabled: ``` ./db_bench -db=/tmp/prefix_scan_prefetch_main -benchmarks="fillseq" -key_size=32 -value_size=512 -num=5000000 - use_direct_io_for_flush_and_compaction=true -target_file_size_base=16777216 ``` ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` ii) normal prefetching after changes with async_io disable: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_withchange -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 14:11:31 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_withchange] seekrandom : 471347.227 micros/op 2 ops/sec; 348.1 MB/s (255 of 255 found) ``` Reviewed By: anand1976 Differential Revision: D34731543 Pulled By: akankshamahajan15 fbshipit-source-id: 8e23aa93453d5fe3c672b9231ad582f60207937f
3 years ago
}
if (!io_s.ok()) {
io_status_ = io_s;
return true;
}
}
return got_from_prefetch_buffer_;
}
inline bool BlockFetcher::TryGetCompressedBlockFromPersistentCache() {
if (cache_options_.persistent_cache &&
cache_options_.persistent_cache->IsCompressed()) {
// lookup uncompressed cache mode p-cache
std::unique_ptr<char[]> raw_data;
io_status_ = status_to_io_status(PersistentCacheHelper::LookupRawPage(
cache_options_, handle_, &raw_data, block_size_with_trailer_));
if (io_status_.ok()) {
heap_buf_ = CacheAllocationPtr(raw_data.release());
used_buf_ = heap_buf_.get();
slice_ = Slice(heap_buf_.get(), block_size_);
Improve / clean up meta block code & integrity (#9163) Summary: * Checksums are now checked on meta blocks unless specifically suppressed or not applicable (e.g. plain table). (Was other way around.) This means a number of cases that were not checking checksums now are, including direct read TableProperties in Version::GetTableProperties (fixed in meta_blocks ReadTableProperties), reading any block from PersistentCache (fixed in BlockFetcher), read TableProperties in SstFileDumper (ldb/sst_dump/BackupEngine) before table reader open, maybe more. * For that to work, I moved the global_seqno+TableProperties checksum logic to the shared table/ code, because that is used by many utilies such as SstFileDumper. * Also for that to work, we have to know when we're dealing with a block that has a checksum (trailer), so added that capability to Footer based on magic number, and from there BlockFetcher. * Knowledge of trailer presence has also fixed a problem where other table formats were reading blocks including bytes for a non-existant trailer--and awkwardly kind-of not using them, e.g. no shared code checking checksums. (BlockFetcher compression type was populated incorrectly.) Now we only read what is needed. * Minimized code duplication and differing/incompatible/awkward abstractions in meta_blocks.{cc,h} (e.g. SeekTo in metaindex block without parsing block handle) * Moved some meta block handling code from table_properties*.* * Moved some code specific to block-based table from shared table/ code to BlockBasedTable class. The checksum stuff means we can't completely separate it, but things that don't need to be in shared table/ code should not be. * Use unique_ptr rather than raw ptr in more places. (Note: you can std::move from unique_ptr to shared_ptr.) Without enhancements to GetPropertiesOfAllTablesTest (see below), net reduction of roughly 100 lines of code. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9163 Test Plan: existing tests and * Enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to verify that checksums are now checked on direct read of table properties by TableCache (new test would fail before this change) * Also enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to test putting table properties under old meta name * Also generally enhanced that same test to actually test what it was supposed to be testing already, by kicking things out of table cache when we don't want them there. Reviewed By: ajkr, mrambacher Differential Revision: D32514757 Pulled By: pdillinger fbshipit-source-id: 507964b9311d186ae8d1131182290cbd97a99fa9
3 years ago
ProcessTrailerIfPresent();
return true;
} else if (!io_status_.IsNotFound() && ioptions_.logger) {
assert(!io_status_.ok());
ROCKS_LOG_INFO(ioptions_.logger,
"Error reading from persistent cache. %s",
io_status_.ToString().c_str());
}
}
return false;
}
inline void BlockFetcher::PrepareBufferForBlockFromFile() {
// cache miss read from device
if ((do_uncompress_ || ioptions_.allow_mmap_reads) &&
block_size_with_trailer_ < kDefaultStackBufferSize) {
// If we've got a small enough hunk of data, read it in to the
// trivially allocated stack buffer instead of needing a full malloc()
//
// `GetBlockContents()` cannot return this data as its lifetime is tied to
// this `BlockFetcher`'s lifetime. That is fine because this is only used
// in cases where we do not expect the `GetBlockContents()` result to be the
// same buffer we are assigning here. If we guess incorrectly, there will be
// a heap allocation and memcpy in `GetBlockContents()` to obtain the final
// result. Considering we are eliding a heap allocation here by using the
// stack buffer, the cost of guessing incorrectly here is one extra memcpy.
//
// When `do_uncompress_` is true, we expect the uncompression step will
// allocate heap memory for the final result. However this expectation will
// be wrong if the block turns out to already be uncompressed, which we
// won't know for sure until after reading it.
//
// When `ioptions_.allow_mmap_reads` is true, we do not expect the file
// reader to use the scratch buffer at all, but instead return a pointer
// into the mapped memory. This expectation will be wrong when using a
// file reader that does not implement mmap reads properly.
used_buf_ = &stack_buf_[0];
} else if (maybe_compressed_ && !do_uncompress_) {
compressed_buf_ = AllocateBlock(block_size_with_trailer_,
memory_allocator_compressed_);
used_buf_ = compressed_buf_.get();
} else {
heap_buf_ =
AllocateBlock(block_size_with_trailer_, memory_allocator_);
used_buf_ = heap_buf_.get();
}
}
inline void BlockFetcher::InsertCompressedBlockToPersistentCacheIfNeeded() {
if (io_status_.ok() && read_options_.fill_cache &&
cache_options_.persistent_cache &&
cache_options_.persistent_cache->IsCompressed()) {
// insert to raw cache
PersistentCacheHelper::InsertRawPage(cache_options_, handle_, used_buf_,
block_size_with_trailer_);
}
}
inline void BlockFetcher::InsertUncompressedBlockToPersistentCacheIfNeeded() {
if (io_status_.ok() && !got_from_prefetch_buffer_ &&
read_options_.fill_cache && cache_options_.persistent_cache &&
!cache_options_.persistent_cache->IsCompressed()) {
// insert to uncompressed cache
PersistentCacheHelper::InsertUncompressedPage(cache_options_, handle_,
*contents_);
}
}
inline void BlockFetcher::CopyBufferToHeapBuf() {
assert(used_buf_ != heap_buf_.get());
heap_buf_ = AllocateBlock(block_size_with_trailer_, memory_allocator_);
memcpy(heap_buf_.get(), used_buf_, block_size_with_trailer_);
#ifndef NDEBUG
num_heap_buf_memcpy_++;
#endif
}
inline void BlockFetcher::CopyBufferToCompressedBuf() {
assert(used_buf_ != compressed_buf_.get());
compressed_buf_ = AllocateBlock(block_size_with_trailer_,
memory_allocator_compressed_);
memcpy(compressed_buf_.get(), used_buf_, block_size_with_trailer_);
#ifndef NDEBUG
num_compressed_buf_memcpy_++;
#endif
}
// Entering this method means the block is not compressed or do not need to be
// uncompressed. The block can be in one of the following buffers:
// 1. prefetch buffer if prefetch is enabled and the block is prefetched before
// 2. stack_buf_ if block size is smaller than the stack_buf_ size and block
// is not compressed
// 3. heap_buf_ if the block is not compressed
// 4. compressed_buf_ if the block is compressed
// 5. direct_io_buf_ if direct IO is enabled
// After this method, if the block is compressed, it should be in
// compressed_buf_, otherwise should be in heap_buf_.
inline void BlockFetcher::GetBlockContents() {
if (slice_.data() != used_buf_) {
// the slice content is not the buffer provided
*contents_ = BlockContents(Slice(slice_.data(), block_size_));
} else {
// page can be either uncompressed or compressed, the buffer either stack
// or heap provided. Refer to https://github.com/facebook/rocksdb/pull/4096
if (got_from_prefetch_buffer_ || used_buf_ == &stack_buf_[0]) {
CopyBufferToHeapBuf();
} else if (used_buf_ == compressed_buf_.get()) {
if (compression_type_ == kNoCompression &&
memory_allocator_ != memory_allocator_compressed_) {
CopyBufferToHeapBuf();
} else {
heap_buf_ = std::move(compressed_buf_);
}
} else if (direct_io_buf_.get() != nullptr) {
if (compression_type_ == kNoCompression) {
CopyBufferToHeapBuf();
} else {
CopyBufferToCompressedBuf();
heap_buf_ = std::move(compressed_buf_);
}
}
*contents_ = BlockContents(std::move(heap_buf_), block_size_);
}
#ifndef NDEBUG
contents_->is_raw_block = true;
#endif
}
IOStatus BlockFetcher::ReadBlockContents() {
if (TryGetUncompressBlockFromPersistentCache()) {
compression_type_ = kNoCompression;
#ifndef NDEBUG
contents_->is_raw_block = true;
#endif // NDEBUG
return IOStatus::OK();
}
if (TryGetFromPrefetchBuffer()) {
if (!io_status_.ok()) {
return io_status_;
}
} else if (!TryGetCompressedBlockFromPersistentCache()) {
IOOptions opts;
io_status_ = file_->PrepareIOOptions(read_options_, opts);
// Actual file read
if (io_status_.ok()) {
if (file_->use_direct_io()) {
PERF_TIMER_GUARD(block_read_time);
Add rate limiter priority to ReadOptions (#9424) Summary: Users can set the priority for file reads associated with their operation by setting `ReadOptions::rate_limiter_priority` to something other than `Env::IO_TOTAL`. Rate limiting `VerifyChecksum()` and `VerifyFileChecksums()` is the motivation for this PR, so it also includes benchmarks and minor bug fixes to get that working. `RandomAccessFileReader::Read()` already had support for rate limiting compaction reads. I changed that rate limiting to be non-specific to compaction, but rather performed according to the passed in `Env::IOPriority`. Now the compaction read rate limiting is supported by setting `rate_limiter_priority = Env::IO_LOW` on its `ReadOptions`. There is no default value for the new `Env::IOPriority` parameter to `RandomAccessFileReader::Read()`. That means this PR goes through all callers (in some cases multiple layers up the call stack) to find a `ReadOptions` to provide the priority. There are TODOs for cases I believe it would be good to let user control the priority some day (e.g., file footer reads), and no TODO in cases I believe it doesn't matter (e.g., trace file reads). The API doc only lists the missing cases where a file read associated with a provided `ReadOptions` cannot be rate limited. For cases like file ingestion checksum calculation, there is no API to provide `ReadOptions` or `Env::IOPriority`, so I didn't count that as missing. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9424 Test Plan: - new unit tests - new benchmarks on ~50MB database with 1MB/s read rate limit and 100ms refill interval; verified with strace reads are chunked (at 0.1MB per chunk) and spaced roughly 100ms apart. - setup command: `./db_bench -benchmarks=fillrandom,compact -db=/tmp/testdb -target_file_size_base=1048576 -disable_auto_compactions=true -file_checksum=true` - benchmarks command: `strace -ttfe pread64 ./db_bench -benchmarks=verifychecksum,verifyfilechecksums -use_existing_db=true -db=/tmp/testdb -rate_limiter_bytes_per_sec=1048576 -rate_limit_bg_reads=1 -rate_limit_user_ops=true -file_checksum=true` - crash test using IO_USER priority on non-validation reads with https://github.com/facebook/rocksdb/issues/9567 reverted: `python3 tools/db_crashtest.py blackbox --max_key=1000000 --write_buffer_size=524288 --target_file_size_base=524288 --level_compaction_dynamic_level_bytes=true --duration=3600 --rate_limit_bg_reads=true --rate_limit_user_ops=true --rate_limiter_bytes_per_sec=10485760 --interval=10` Reviewed By: hx235 Differential Revision: D33747386 Pulled By: ajkr fbshipit-source-id: a2d985e97912fba8c54763798e04f006ccc56e0c
3 years ago
io_status_ = file_->Read(
opts, handle_.offset(), block_size_with_trailer_, &slice_, nullptr,
&direct_io_buf_, read_options_.rate_limiter_priority);
PERF_COUNTER_ADD(block_read_count, 1);
used_buf_ = const_cast<char*>(slice_.data());
} else {
PrepareBufferForBlockFromFile();
PERF_TIMER_GUARD(block_read_time);
Add rate limiter priority to ReadOptions (#9424) Summary: Users can set the priority for file reads associated with their operation by setting `ReadOptions::rate_limiter_priority` to something other than `Env::IO_TOTAL`. Rate limiting `VerifyChecksum()` and `VerifyFileChecksums()` is the motivation for this PR, so it also includes benchmarks and minor bug fixes to get that working. `RandomAccessFileReader::Read()` already had support for rate limiting compaction reads. I changed that rate limiting to be non-specific to compaction, but rather performed according to the passed in `Env::IOPriority`. Now the compaction read rate limiting is supported by setting `rate_limiter_priority = Env::IO_LOW` on its `ReadOptions`. There is no default value for the new `Env::IOPriority` parameter to `RandomAccessFileReader::Read()`. That means this PR goes through all callers (in some cases multiple layers up the call stack) to find a `ReadOptions` to provide the priority. There are TODOs for cases I believe it would be good to let user control the priority some day (e.g., file footer reads), and no TODO in cases I believe it doesn't matter (e.g., trace file reads). The API doc only lists the missing cases where a file read associated with a provided `ReadOptions` cannot be rate limited. For cases like file ingestion checksum calculation, there is no API to provide `ReadOptions` or `Env::IOPriority`, so I didn't count that as missing. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9424 Test Plan: - new unit tests - new benchmarks on ~50MB database with 1MB/s read rate limit and 100ms refill interval; verified with strace reads are chunked (at 0.1MB per chunk) and spaced roughly 100ms apart. - setup command: `./db_bench -benchmarks=fillrandom,compact -db=/tmp/testdb -target_file_size_base=1048576 -disable_auto_compactions=true -file_checksum=true` - benchmarks command: `strace -ttfe pread64 ./db_bench -benchmarks=verifychecksum,verifyfilechecksums -use_existing_db=true -db=/tmp/testdb -rate_limiter_bytes_per_sec=1048576 -rate_limit_bg_reads=1 -rate_limit_user_ops=true -file_checksum=true` - crash test using IO_USER priority on non-validation reads with https://github.com/facebook/rocksdb/issues/9567 reverted: `python3 tools/db_crashtest.py blackbox --max_key=1000000 --write_buffer_size=524288 --target_file_size_base=524288 --level_compaction_dynamic_level_bytes=true --duration=3600 --rate_limit_bg_reads=true --rate_limit_user_ops=true --rate_limiter_bytes_per_sec=10485760 --interval=10` Reviewed By: hx235 Differential Revision: D33747386 Pulled By: ajkr fbshipit-source-id: a2d985e97912fba8c54763798e04f006ccc56e0c
3 years ago
io_status_ = file_->Read(opts, handle_.offset(),
block_size_with_trailer_, &slice_, used_buf_,
nullptr, read_options_.rate_limiter_priority);
PERF_COUNTER_ADD(block_read_count, 1);
#ifndef NDEBUG
if (slice_.data() == &stack_buf_[0]) {
num_stack_buf_memcpy_++;
} else if (slice_.data() == heap_buf_.get()) {
num_heap_buf_memcpy_++;
} else if (slice_.data() == compressed_buf_.get()) {
num_compressed_buf_memcpy_++;
}
#endif
}
}
// TODO: introduce dedicated perf counter for range tombstones
switch (block_type_) {
case BlockType::kFilter:
case BlockType::kFilterPartitionIndex:
case BlockType::kDeprecatedFilter:
PERF_COUNTER_ADD(filter_block_read_count, 1);
break;
case BlockType::kCompressionDictionary:
PERF_COUNTER_ADD(compression_dict_block_read_count, 1);
break;
case BlockType::kIndex:
PERF_COUNTER_ADD(index_block_read_count, 1);
break;
// Nothing to do here as we don't have counters for the other types.
default:
break;
}
PERF_COUNTER_ADD(block_read_byte, block_size_with_trailer_);
if (!io_status_.ok()) {
return io_status_;
}
if (slice_.size() != block_size_with_trailer_) {
return IOStatus::Corruption(
"truncated block read from " + file_->file_name() + " offset " +
std::to_string(handle_.offset()) + ", expected " +
std::to_string(block_size_with_trailer_) + " bytes, got " +
std::to_string(slice_.size()));
}
Improve / clean up meta block code & integrity (#9163) Summary: * Checksums are now checked on meta blocks unless specifically suppressed or not applicable (e.g. plain table). (Was other way around.) This means a number of cases that were not checking checksums now are, including direct read TableProperties in Version::GetTableProperties (fixed in meta_blocks ReadTableProperties), reading any block from PersistentCache (fixed in BlockFetcher), read TableProperties in SstFileDumper (ldb/sst_dump/BackupEngine) before table reader open, maybe more. * For that to work, I moved the global_seqno+TableProperties checksum logic to the shared table/ code, because that is used by many utilies such as SstFileDumper. * Also for that to work, we have to know when we're dealing with a block that has a checksum (trailer), so added that capability to Footer based on magic number, and from there BlockFetcher. * Knowledge of trailer presence has also fixed a problem where other table formats were reading blocks including bytes for a non-existant trailer--and awkwardly kind-of not using them, e.g. no shared code checking checksums. (BlockFetcher compression type was populated incorrectly.) Now we only read what is needed. * Minimized code duplication and differing/incompatible/awkward abstractions in meta_blocks.{cc,h} (e.g. SeekTo in metaindex block without parsing block handle) * Moved some meta block handling code from table_properties*.* * Moved some code specific to block-based table from shared table/ code to BlockBasedTable class. The checksum stuff means we can't completely separate it, but things that don't need to be in shared table/ code should not be. * Use unique_ptr rather than raw ptr in more places. (Note: you can std::move from unique_ptr to shared_ptr.) Without enhancements to GetPropertiesOfAllTablesTest (see below), net reduction of roughly 100 lines of code. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9163 Test Plan: existing tests and * Enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to verify that checksums are now checked on direct read of table properties by TableCache (new test would fail before this change) * Also enhanced DBTablePropertiesTest.GetPropertiesOfAllTablesTest to test putting table properties under old meta name * Also generally enhanced that same test to actually test what it was supposed to be testing already, by kicking things out of table cache when we don't want them there. Reviewed By: ajkr, mrambacher Differential Revision: D32514757 Pulled By: pdillinger fbshipit-source-id: 507964b9311d186ae8d1131182290cbd97a99fa9
3 years ago
ProcessTrailerIfPresent();
if (io_status_.ok()) {
InsertCompressedBlockToPersistentCacheIfNeeded();
} else {
return io_status_;
}
}
if (do_uncompress_ && compression_type_ != kNoCompression) {
PERF_TIMER_GUARD(block_decompress_time);
// compressed page, uncompress, update cache
UncompressionContext context(compression_type_);
UncompressionInfo info(context, uncompression_dict_, compression_type_);
io_status_ = status_to_io_status(UncompressBlockContents(
info, slice_.data(), block_size_, contents_, footer_.format_version(),
ioptions_, memory_allocator_));
#ifndef NDEBUG
num_heap_buf_memcpy_++;
#endif
compression_type_ = kNoCompression;
} else {
GetBlockContents();
}
InsertUncompressedBlockToPersistentCacheIfNeeded();
return io_status_;
}
Seek parallelization (#9994) Summary: The RocksDB iterator is a hierarchy of iterators. MergingIterator maintains a heap of LevelIterators, one for each L0 file and for each non-zero level. The Seek() operation naturally lends itself to parallelization, as it involves positioning every LevelIterator on the correct data block in the correct SST file. It lookups a level for a target key, to find the first key that's >= the target key. This typically involves reading one data block that is likely to contain the target key, and scan forward to find the first valid key. The forward scan may read more data blocks. In order to find the right data block, the iterator may read some metadata blocks (required for opening a file and searching the index). This flow can be parallelized. Design: Seek will be called two times under async_io option. First seek will send asynchronous request to prefetch the data blocks at each level and second seek will follow the normal flow and in FilePrefetchBuffer::TryReadFromCacheAsync it will wait for the Poll() to get the results and add the iterator to min_heap. - Status::TryAgain is passed down from FilePrefetchBuffer::PrefetchAsync to block_iter_.Status indicating asynchronous request has been submitted. - If for some reason asynchronous request returns error in submitting the request, it will fallback to sequential reading of blocks in one pass. - If the data already exists in prefetch_buffer, it will return the data without prefetching further and it will be treated as single pass of seek. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9994 Test Plan: - **Run Regressions.** ``` ./db_bench -db=/tmp/prefix_scan_prefetch_main -benchmarks="fillseq" -key_size=32 -value_size=512 -num=5000000 -use_direct_io_for_flush_and_compaction=true -target_file_size_base=16777216 ``` i) Previous release 7.0 run for normal prefetching with async_io disabled: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` ii) normal prefetching after changes with async_io disable: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Set seed to 1652922591315307 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:09:51 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483080.466 micros/op 2 ops/sec 120.287 seconds 249 operations; 340.8 MB/s (249 of 249 found) ``` iii) db_bench with async_io enabled completed succesfully ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 -async_io=1 -adaptive_readahead=1 Set seed to 1652924062021732 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:34:22 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 553913.576 micros/op 1 ops/sec 120.199 seconds 217 operations; 293.6 MB/s (217 of 217 found) ``` - db_stress with async_io disabled completed succesfully ``` export CRASH_TEST_EXT_ARGS=" --async_io=0" make crash_test -j ``` I**n Progress**: db_stress with async_io is failing and working on debugging/fixing it. Reviewed By: anand1976 Differential Revision: D36459323 Pulled By: akankshamahajan15 fbshipit-source-id: abb1cd944abe712bae3986ae5b16704b3338917c
2 years ago
IOStatus BlockFetcher::ReadAsyncBlockContents() {
if (TryGetUncompressBlockFromPersistentCache()) {
compression_type_ = kNoCompression;
#ifndef NDEBUG
contents_->is_raw_block = true;
#endif // NDEBUG
return IOStatus::OK();
} else if (!TryGetCompressedBlockFromPersistentCache()) {
Fix stress test failure "Corruption: checksum mismatch" or "Iterator Diverged" with async_io enabled (#10032) Summary: In case of non sequential reads with `async_io`, `FilePRefetchBuffer::TryReadFromCacheAsync` can be called for previous blocks with `offset < bufs_[curr_].offset_` which wasn't handled correctly resulting wrong data being returned from buffer. Since `FilePRefetchBuffer::PrefetchAsync` can be called for any data block, it sets `prev_len_` to 0 indicating `FilePRefetchBuffer::TryReadFromCacheAsync` to go for the prefetching even though offset < bufs_[curr_].offset_ This is because async prefetching is always done in second buffer (to avoid mutex) even though curr_ is empty leading to offset < bufs_[curr_].offset_ in some cases. If prev_len_ is non zero then `TryReadFromCacheAsync` returns false if `offset < bufs_[curr_].offset_ && prev_len != 0` indicating reads are not sequential and previous call wasn't PrefetchAsync. - This PR also simplifies `FilePRefetchBuffer::TryReadFromCacheAsync` as it was getting complicated covering different scenarios based on `async_io` enabled/disabled. If `for_compaction` is set true, it now calls `FilePRefetchBufferTryReadFromCache` following synchronous flow as before. Its decided in BlockFetcher.cc Pull Request resolved: https://github.com/facebook/rocksdb/pull/10032 Test Plan: 1. export CRASH_TEST_EXT_ARGS=" --async_io=1" make crash_test -j completed successfully locally 2. make crash_test -j completed successfully locally 3. Reran CircleCi mini crashtest job 4 - 5 times. 4. Updated prefetch_test for more coverage. Reviewed By: anand1976 Differential Revision: D36579858 Pulled By: akankshamahajan15 fbshipit-source-id: 0c428d62b45e12e082a83acf533a5e37a584bedf
2 years ago
assert(prefetch_buffer_ != nullptr);
if (!for_compaction_) {
Seek parallelization (#9994) Summary: The RocksDB iterator is a hierarchy of iterators. MergingIterator maintains a heap of LevelIterators, one for each L0 file and for each non-zero level. The Seek() operation naturally lends itself to parallelization, as it involves positioning every LevelIterator on the correct data block in the correct SST file. It lookups a level for a target key, to find the first key that's >= the target key. This typically involves reading one data block that is likely to contain the target key, and scan forward to find the first valid key. The forward scan may read more data blocks. In order to find the right data block, the iterator may read some metadata blocks (required for opening a file and searching the index). This flow can be parallelized. Design: Seek will be called two times under async_io option. First seek will send asynchronous request to prefetch the data blocks at each level and second seek will follow the normal flow and in FilePrefetchBuffer::TryReadFromCacheAsync it will wait for the Poll() to get the results and add the iterator to min_heap. - Status::TryAgain is passed down from FilePrefetchBuffer::PrefetchAsync to block_iter_.Status indicating asynchronous request has been submitted. - If for some reason asynchronous request returns error in submitting the request, it will fallback to sequential reading of blocks in one pass. - If the data already exists in prefetch_buffer, it will return the data without prefetching further and it will be treated as single pass of seek. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9994 Test Plan: - **Run Regressions.** ``` ./db_bench -db=/tmp/prefix_scan_prefetch_main -benchmarks="fillseq" -key_size=32 -value_size=512 -num=5000000 -use_direct_io_for_flush_and_compaction=true -target_file_size_base=16777216 ``` i) Previous release 7.0 run for normal prefetching with async_io disabled: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` ii) normal prefetching after changes with async_io disable: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Set seed to 1652922591315307 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:09:51 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483080.466 micros/op 2 ops/sec 120.287 seconds 249 operations; 340.8 MB/s (249 of 249 found) ``` iii) db_bench with async_io enabled completed succesfully ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 -async_io=1 -adaptive_readahead=1 Set seed to 1652924062021732 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:34:22 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 553913.576 micros/op 1 ops/sec 120.199 seconds 217 operations; 293.6 MB/s (217 of 217 found) ``` - db_stress with async_io disabled completed succesfully ``` export CRASH_TEST_EXT_ARGS=" --async_io=0" make crash_test -j ``` I**n Progress**: db_stress with async_io is failing and working on debugging/fixing it. Reviewed By: anand1976 Differential Revision: D36459323 Pulled By: akankshamahajan15 fbshipit-source-id: abb1cd944abe712bae3986ae5b16704b3338917c
2 years ago
IOOptions opts;
IOStatus io_s = file_->PrepareIOOptions(read_options_, opts);
Fix stress test failure "Corruption: checksum mismatch" or "Iterator Diverged" with async_io enabled (#10032) Summary: In case of non sequential reads with `async_io`, `FilePRefetchBuffer::TryReadFromCacheAsync` can be called for previous blocks with `offset < bufs_[curr_].offset_` which wasn't handled correctly resulting wrong data being returned from buffer. Since `FilePRefetchBuffer::PrefetchAsync` can be called for any data block, it sets `prev_len_` to 0 indicating `FilePRefetchBuffer::TryReadFromCacheAsync` to go for the prefetching even though offset < bufs_[curr_].offset_ This is because async prefetching is always done in second buffer (to avoid mutex) even though curr_ is empty leading to offset < bufs_[curr_].offset_ in some cases. If prev_len_ is non zero then `TryReadFromCacheAsync` returns false if `offset < bufs_[curr_].offset_ && prev_len != 0` indicating reads are not sequential and previous call wasn't PrefetchAsync. - This PR also simplifies `FilePRefetchBuffer::TryReadFromCacheAsync` as it was getting complicated covering different scenarios based on `async_io` enabled/disabled. If `for_compaction` is set true, it now calls `FilePRefetchBufferTryReadFromCache` following synchronous flow as before. Its decided in BlockFetcher.cc Pull Request resolved: https://github.com/facebook/rocksdb/pull/10032 Test Plan: 1. export CRASH_TEST_EXT_ARGS=" --async_io=1" make crash_test -j completed successfully locally 2. make crash_test -j completed successfully locally 3. Reran CircleCi mini crashtest job 4 - 5 times. 4. Updated prefetch_test for more coverage. Reviewed By: anand1976 Differential Revision: D36579858 Pulled By: akankshamahajan15 fbshipit-source-id: 0c428d62b45e12e082a83acf533a5e37a584bedf
2 years ago
if (!io_s.ok()) {
return io_s;
}
io_s = status_to_io_status(prefetch_buffer_->PrefetchAsync(
opts, file_, handle_.offset(), block_size_with_trailer_,
read_options_.rate_limiter_priority, &slice_));
if (io_s.IsTryAgain()) {
return io_s;
}
Seek parallelization (#9994) Summary: The RocksDB iterator is a hierarchy of iterators. MergingIterator maintains a heap of LevelIterators, one for each L0 file and for each non-zero level. The Seek() operation naturally lends itself to parallelization, as it involves positioning every LevelIterator on the correct data block in the correct SST file. It lookups a level for a target key, to find the first key that's >= the target key. This typically involves reading one data block that is likely to contain the target key, and scan forward to find the first valid key. The forward scan may read more data blocks. In order to find the right data block, the iterator may read some metadata blocks (required for opening a file and searching the index). This flow can be parallelized. Design: Seek will be called two times under async_io option. First seek will send asynchronous request to prefetch the data blocks at each level and second seek will follow the normal flow and in FilePrefetchBuffer::TryReadFromCacheAsync it will wait for the Poll() to get the results and add the iterator to min_heap. - Status::TryAgain is passed down from FilePrefetchBuffer::PrefetchAsync to block_iter_.Status indicating asynchronous request has been submitted. - If for some reason asynchronous request returns error in submitting the request, it will fallback to sequential reading of blocks in one pass. - If the data already exists in prefetch_buffer, it will return the data without prefetching further and it will be treated as single pass of seek. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9994 Test Plan: - **Run Regressions.** ``` ./db_bench -db=/tmp/prefix_scan_prefetch_main -benchmarks="fillseq" -key_size=32 -value_size=512 -num=5000000 -use_direct_io_for_flush_and_compaction=true -target_file_size_base=16777216 ``` i) Previous release 7.0 run for normal prefetching with async_io disabled: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` ii) normal prefetching after changes with async_io disable: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Set seed to 1652922591315307 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:09:51 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483080.466 micros/op 2 ops/sec 120.287 seconds 249 operations; 340.8 MB/s (249 of 249 found) ``` iii) db_bench with async_io enabled completed succesfully ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 -async_io=1 -adaptive_readahead=1 Set seed to 1652924062021732 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:34:22 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 553913.576 micros/op 1 ops/sec 120.199 seconds 217 operations; 293.6 MB/s (217 of 217 found) ``` - db_stress with async_io disabled completed succesfully ``` export CRASH_TEST_EXT_ARGS=" --async_io=0" make crash_test -j ``` I**n Progress**: db_stress with async_io is failing and working on debugging/fixing it. Reviewed By: anand1976 Differential Revision: D36459323 Pulled By: akankshamahajan15 fbshipit-source-id: abb1cd944abe712bae3986ae5b16704b3338917c
2 years ago
if (io_s.ok()) {
// Data Block is already in prefetch.
got_from_prefetch_buffer_ = true;
ProcessTrailerIfPresent();
if (!io_status_.ok()) {
return io_status_;
}
used_buf_ = const_cast<char*>(slice_.data());
if (do_uncompress_ && compression_type_ != kNoCompression) {
PERF_TIMER_GUARD(block_decompress_time);
// compressed page, uncompress, update cache
UncompressionContext context(compression_type_);
UncompressionInfo info(context, uncompression_dict_,
compression_type_);
io_status_ = status_to_io_status(UncompressBlockContents(
info, slice_.data(), block_size_, contents_,
footer_.format_version(), ioptions_, memory_allocator_));
#ifndef NDEBUG
num_heap_buf_memcpy_++;
#endif
compression_type_ = kNoCompression;
} else {
GetBlockContents();
}
InsertUncompressedBlockToPersistentCacheIfNeeded();
Fix stress test failure "Corruption: checksum mismatch" or "Iterator Diverged" with async_io enabled (#10032) Summary: In case of non sequential reads with `async_io`, `FilePRefetchBuffer::TryReadFromCacheAsync` can be called for previous blocks with `offset < bufs_[curr_].offset_` which wasn't handled correctly resulting wrong data being returned from buffer. Since `FilePRefetchBuffer::PrefetchAsync` can be called for any data block, it sets `prev_len_` to 0 indicating `FilePRefetchBuffer::TryReadFromCacheAsync` to go for the prefetching even though offset < bufs_[curr_].offset_ This is because async prefetching is always done in second buffer (to avoid mutex) even though curr_ is empty leading to offset < bufs_[curr_].offset_ in some cases. If prev_len_ is non zero then `TryReadFromCacheAsync` returns false if `offset < bufs_[curr_].offset_ && prev_len != 0` indicating reads are not sequential and previous call wasn't PrefetchAsync. - This PR also simplifies `FilePRefetchBuffer::TryReadFromCacheAsync` as it was getting complicated covering different scenarios based on `async_io` enabled/disabled. If `for_compaction` is set true, it now calls `FilePRefetchBufferTryReadFromCache` following synchronous flow as before. Its decided in BlockFetcher.cc Pull Request resolved: https://github.com/facebook/rocksdb/pull/10032 Test Plan: 1. export CRASH_TEST_EXT_ARGS=" --async_io=1" make crash_test -j completed successfully locally 2. make crash_test -j completed successfully locally 3. Reran CircleCi mini crashtest job 4 - 5 times. 4. Updated prefetch_test for more coverage. Reviewed By: anand1976 Differential Revision: D36579858 Pulled By: akankshamahajan15 fbshipit-source-id: 0c428d62b45e12e082a83acf533a5e37a584bedf
2 years ago
return io_status_;
Seek parallelization (#9994) Summary: The RocksDB iterator is a hierarchy of iterators. MergingIterator maintains a heap of LevelIterators, one for each L0 file and for each non-zero level. The Seek() operation naturally lends itself to parallelization, as it involves positioning every LevelIterator on the correct data block in the correct SST file. It lookups a level for a target key, to find the first key that's >= the target key. This typically involves reading one data block that is likely to contain the target key, and scan forward to find the first valid key. The forward scan may read more data blocks. In order to find the right data block, the iterator may read some metadata blocks (required for opening a file and searching the index). This flow can be parallelized. Design: Seek will be called two times under async_io option. First seek will send asynchronous request to prefetch the data blocks at each level and second seek will follow the normal flow and in FilePrefetchBuffer::TryReadFromCacheAsync it will wait for the Poll() to get the results and add the iterator to min_heap. - Status::TryAgain is passed down from FilePrefetchBuffer::PrefetchAsync to block_iter_.Status indicating asynchronous request has been submitted. - If for some reason asynchronous request returns error in submitting the request, it will fallback to sequential reading of blocks in one pass. - If the data already exists in prefetch_buffer, it will return the data without prefetching further and it will be treated as single pass of seek. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9994 Test Plan: - **Run Regressions.** ``` ./db_bench -db=/tmp/prefix_scan_prefetch_main -benchmarks="fillseq" -key_size=32 -value_size=512 -num=5000000 -use_direct_io_for_flush_and_compaction=true -target_file_size_base=16777216 ``` i) Previous release 7.0 run for normal prefetching with async_io disabled: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` ii) normal prefetching after changes with async_io disable: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Set seed to 1652922591315307 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:09:51 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483080.466 micros/op 2 ops/sec 120.287 seconds 249 operations; 340.8 MB/s (249 of 249 found) ``` iii) db_bench with async_io enabled completed succesfully ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 -async_io=1 -adaptive_readahead=1 Set seed to 1652924062021732 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:34:22 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 553913.576 micros/op 1 ops/sec 120.199 seconds 217 operations; 293.6 MB/s (217 of 217 found) ``` - db_stress with async_io disabled completed succesfully ``` export CRASH_TEST_EXT_ARGS=" --async_io=0" make crash_test -j ``` I**n Progress**: db_stress with async_io is failing and working on debugging/fixing it. Reviewed By: anand1976 Differential Revision: D36459323 Pulled By: akankshamahajan15 fbshipit-source-id: abb1cd944abe712bae3986ae5b16704b3338917c
2 years ago
}
}
Fix stress test failure "Corruption: checksum mismatch" or "Iterator Diverged" with async_io enabled (#10032) Summary: In case of non sequential reads with `async_io`, `FilePRefetchBuffer::TryReadFromCacheAsync` can be called for previous blocks with `offset < bufs_[curr_].offset_` which wasn't handled correctly resulting wrong data being returned from buffer. Since `FilePRefetchBuffer::PrefetchAsync` can be called for any data block, it sets `prev_len_` to 0 indicating `FilePRefetchBuffer::TryReadFromCacheAsync` to go for the prefetching even though offset < bufs_[curr_].offset_ This is because async prefetching is always done in second buffer (to avoid mutex) even though curr_ is empty leading to offset < bufs_[curr_].offset_ in some cases. If prev_len_ is non zero then `TryReadFromCacheAsync` returns false if `offset < bufs_[curr_].offset_ && prev_len != 0` indicating reads are not sequential and previous call wasn't PrefetchAsync. - This PR also simplifies `FilePRefetchBuffer::TryReadFromCacheAsync` as it was getting complicated covering different scenarios based on `async_io` enabled/disabled. If `for_compaction` is set true, it now calls `FilePRefetchBufferTryReadFromCache` following synchronous flow as before. Its decided in BlockFetcher.cc Pull Request resolved: https://github.com/facebook/rocksdb/pull/10032 Test Plan: 1. export CRASH_TEST_EXT_ARGS=" --async_io=1" make crash_test -j completed successfully locally 2. make crash_test -j completed successfully locally 3. Reran CircleCi mini crashtest job 4 - 5 times. 4. Updated prefetch_test for more coverage. Reviewed By: anand1976 Differential Revision: D36579858 Pulled By: akankshamahajan15 fbshipit-source-id: 0c428d62b45e12e082a83acf533a5e37a584bedf
2 years ago
// Fallback to sequential reading of data blocks in case of io_s returns
// error or for_compaction_is true.
return ReadBlockContents();
Seek parallelization (#9994) Summary: The RocksDB iterator is a hierarchy of iterators. MergingIterator maintains a heap of LevelIterators, one for each L0 file and for each non-zero level. The Seek() operation naturally lends itself to parallelization, as it involves positioning every LevelIterator on the correct data block in the correct SST file. It lookups a level for a target key, to find the first key that's >= the target key. This typically involves reading one data block that is likely to contain the target key, and scan forward to find the first valid key. The forward scan may read more data blocks. In order to find the right data block, the iterator may read some metadata blocks (required for opening a file and searching the index). This flow can be parallelized. Design: Seek will be called two times under async_io option. First seek will send asynchronous request to prefetch the data blocks at each level and second seek will follow the normal flow and in FilePrefetchBuffer::TryReadFromCacheAsync it will wait for the Poll() to get the results and add the iterator to min_heap. - Status::TryAgain is passed down from FilePrefetchBuffer::PrefetchAsync to block_iter_.Status indicating asynchronous request has been submitted. - If for some reason asynchronous request returns error in submitting the request, it will fallback to sequential reading of blocks in one pass. - If the data already exists in prefetch_buffer, it will return the data without prefetching further and it will be treated as single pass of seek. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9994 Test Plan: - **Run Regressions.** ``` ./db_bench -db=/tmp/prefix_scan_prefetch_main -benchmarks="fillseq" -key_size=32 -value_size=512 -num=5000000 -use_direct_io_for_flush_and_compaction=true -target_file_size_base=16777216 ``` i) Previous release 7.0 run for normal prefetching with async_io disabled: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.0 Date: Thu Mar 17 13:11:34 2022 CPU: 24 * Intel Core Processor (Broadwell) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483618.390 micros/op 2 ops/sec; 338.9 MB/s (249 of 249 found) ``` ii) normal prefetching after changes with async_io disable: ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 Set seed to 1652922591315307 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:09:51 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 483080.466 micros/op 2 ops/sec 120.287 seconds 249 operations; 340.8 MB/s (249 of 249 found) ``` iii) db_bench with async_io enabled completed succesfully ``` ./db_bench -use_existing_db=true -db=/tmp/prefix_scan_prefetch_main -benchmarks="seekrandom" -key_size=32 -value_size=512 -num=5000000 -use_direct_reads=true -seek_nexts=327680 -duration=120 -ops_between_duration_checks=1 -async_io=1 -adaptive_readahead=1 Set seed to 1652924062021732 because --seed was 0 Initializing RocksDB Options from the specified file Initializing RocksDB Options from command-line flags RocksDB: version 7.3 Date: Wed May 18 18:34:22 2022 CPU: 32 * Intel Xeon Processor (Skylake) CPUCache: 16384 KB Keys: 32 bytes each (+ 0 bytes user-defined timestamp) Values: 512 bytes each (256 bytes after compression) Entries: 5000000 Prefix: 0 bytes Keys per prefix: 0 RawSize: 2594.0 MB (estimated) FileSize: 1373.3 MB (estimated) Write rate: 0 bytes/second Read rate: 0 ops/second Compression: Snappy Compression sampling rate: 0 Memtablerep: SkipListFactory Perf Level: 1 ------------------------------------------------ DB path: [/tmp/prefix_scan_prefetch_main] seekrandom : 553913.576 micros/op 1 ops/sec 120.199 seconds 217 operations; 293.6 MB/s (217 of 217 found) ``` - db_stress with async_io disabled completed succesfully ``` export CRASH_TEST_EXT_ARGS=" --async_io=0" make crash_test -j ``` I**n Progress**: db_stress with async_io is failing and working on debugging/fixing it. Reviewed By: anand1976 Differential Revision: D36459323 Pulled By: akankshamahajan15 fbshipit-source-id: abb1cd944abe712bae3986ae5b16704b3338917c
2 years ago
}
return io_status_;
}
} // namespace ROCKSDB_NAMESPACE