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/utilities/cache_dump_load_impl.h

360 lines
12 KiB

// Copyright (c) Facebook, Inc. and its affiliates. 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).
#pragma once
#ifndef ROCKSDB_LITE
#include <unordered_map>
#include "file/random_access_file_reader.h"
#include "file/writable_file_writer.h"
#include "rocksdb/utilities/cache_dump_load.h"
#include "table/block_based/block.h"
#include "table/block_based/block_like_traits.h"
#include "table/block_based/block_type.h"
#include "table/block_based/cachable_entry.h"
#include "table/block_based/parsed_full_filter_block.h"
#include "table/block_based/reader_common.h"
namespace ROCKSDB_NAMESPACE {
// the read buffer size of for the default CacheDumpReader
const unsigned int kDumpReaderBufferSize = 1024; // 1KB
static const unsigned int kSizePrefixLen = 4;
enum CacheDumpUnitType : unsigned char {
kHeader = 1,
kFooter = 2,
kData = 3,
kFilter = 4,
kProperties = 5,
kCompressionDictionary = 6,
kRangeDeletion = 7,
kHashIndexPrefixes = 8,
kHashIndexMetadata = 9,
kMetaIndex = 10,
kIndex = 11,
Remove deprecated block-based filter (#10184) Summary: In https://github.com/facebook/rocksdb/issues/9535, release 7.0, we hid the old block-based filter from being created using the public API, because of its inefficiency. Although we normally maintain read compatibility on old DBs forever, filters are not required for reading a DB, only for optimizing read performance. Thus, it should be acceptable to remove this code and the substantial maintenance burden it carries as useful features are developed and validated (such as user timestamp). This change completely removes the code for reading and writing the old block-based filters, net removing about 1370 lines of code no longer needed. Options removed from testing / benchmarking tools. The prior existence is only evident in a couple of places: * `CacheEntryRole::kDeprecatedFilterBlock` - We can update this public API enum in a major release to minimize source code incompatibilities. * A warning is logged when an old table file is opened that used the old block-based filter. This is provided as a courtesy, and would be a pain to unit test, so manual testing should suffice. Unfortunately, sst_dump does not tell you whether a file uses block-based filter, and the structure of the code makes it very difficult to fix. * To detect that case, `kObsoleteFilterBlockPrefix` (renamed from `kFilterBlockPrefix`) for metaindex is maintained (for now). Other notes: * In some cases where numbers are associated with filter configurations, we have had to update the assigned numbers so that they all correspond to something that exists. * Fixed potential stat counting bug by assuming `filter_checked = false` for cases like `filter == nullptr` rather than assuming `filter_checked = true` * Removed obsolete `block_offset` and `prefix_extractor` parameters from several functions. * Removed some unnecessary checks `if (!table_prefix_extractor() && !prefix_extractor)` because the caller guarantees the prefix extractor exists and is compatible Pull Request resolved: https://github.com/facebook/rocksdb/pull/10184 Test Plan: tests updated, manually test new warning in LOG using base version to generate a DB Reviewed By: riversand963 Differential Revision: D37212647 Pulled By: pdillinger fbshipit-source-id: 06ee020d8de3b81260ffc36ad0c1202cbf463a80
3 years ago
kDeprecatedFilterBlock = 12, // OBSOLETE / DEPRECATED
kFilterMetaBlock = 13,
kBlockTypeMax,
};
// The metadata of a dump unit. After it is serilized, its size is fixed 16
// bytes.
struct DumpUnitMeta {
// sequence number is a monotonically increasing number to indicate the order
// of the blocks being written. Header is 0.
uint32_t sequence_num;
// The Crc32c checksum of its dump unit.
uint32_t dump_unit_checksum;
// The dump unit size after the dump unit is serilized to a string.
uint64_t dump_unit_size;
void reset() {
sequence_num = 0;
dump_unit_checksum = 0;
dump_unit_size = 0;
}
};
// The data structure to hold a block and its information.
struct DumpUnit {
// The timestamp when the block is identified, copied, and dumped from block
// cache
uint64_t timestamp;
// The type of the block
CacheDumpUnitType type;
// The key of this block when the block is referenced by this Cache
Slice key;
// The block size
size_t value_len;
// The Crc32c checksum of the block
uint32_t value_checksum;
// Pointer to the block. Note that, in the dump process, it points to a memory
// buffer copied from cache block. The buffer is freed when we process the
// next block. In the load process, we use an std::string to store the
Refactor to avoid confusing "raw block" (#10408) Summary: We have a lot of confusing code because of mixed, sometimes completely opposite uses of of the term "raw block" or "raw contents", sometimes within the same source file. For example, in `BlockBasedTableBuilder`, `raw_block_contents` and `raw_size` generally referred to uncompressed block contents and size, while `WriteRawBlock` referred to writing a block that is already compressed if it is going to be. Meanwhile, in `BlockBasedTable`, `raw_block_contents` either referred to a (maybe compressed) block with trailer, or a maybe compressed block maybe without trailer. (Note: left as follow-up work to use C++ typing to better sort out the various kinds of BlockContents.) This change primarily tries to apply some consistent terminology around the kinds of block representations, avoiding the unclear "raw". (Any meaning of "raw" assumes some bias toward the storage layer or toward the logical data layer.) Preferred terminology: * **Serialized block** - bytes that go into storage. For block-based table (usually the case) this includes the block trailer. WART: block `size` may or may not include the trailer; need to be clear about whether it does or not. * **Maybe compressed block** - like a serialized block, but without the trailer (or no promise of including a trailer). Must be accompanied by a CompressionType. * **Uncompressed block** - "payload" bytes that are either stored with no compression, used as input to compression function, or result of decompression function. * **Parsed block** - an in-memory form of a block in block cache, as it is used by the table reader. Different C++ types are used depending on the block type (see block_like_traits.h). Other refactorings: * Misc corrections/improvements of internal API comments * Remove a few misleading / unhelpful / redundant comments. * Use move semantics in some places to simplify contracts * Use better parameter names to indicate which parameters are used for outputs * Remove some extraneous `extern` * Various clean-ups to `CacheDumperImpl` (mostly unnecessary code) Pull Request resolved: https://github.com/facebook/rocksdb/pull/10408 Test Plan: existing tests Reviewed By: akankshamahajan15 Differential Revision: D38172617 Pulled By: pdillinger fbshipit-source-id: ccb99299f324ac5ca46996d34c5089621a4f260c
2 years ago
// serialized dump_unit read from the reader. So it points to the memory
// address of the begin of the block in this string.
void* value;
DumpUnit() { reset(); }
void reset() {
timestamp = 0;
type = CacheDumpUnitType::kBlockTypeMax;
key.clear();
value_len = 0;
value_checksum = 0;
value = nullptr;
}
};
// The default implementation of the Cache Dumper
class CacheDumperImpl : public CacheDumper {
public:
CacheDumperImpl(const CacheDumpOptions& dump_options,
const std::shared_ptr<Cache>& cache,
std::unique_ptr<CacheDumpWriter>&& writer)
: options_(dump_options), cache_(cache), writer_(std::move(writer)) {}
~CacheDumperImpl() { writer_.reset(); }
Status SetDumpFilter(std::vector<DB*> db_list) override;
IOStatus DumpCacheEntriesToWriter() override;
private:
Refactor to avoid confusing "raw block" (#10408) Summary: We have a lot of confusing code because of mixed, sometimes completely opposite uses of of the term "raw block" or "raw contents", sometimes within the same source file. For example, in `BlockBasedTableBuilder`, `raw_block_contents` and `raw_size` generally referred to uncompressed block contents and size, while `WriteRawBlock` referred to writing a block that is already compressed if it is going to be. Meanwhile, in `BlockBasedTable`, `raw_block_contents` either referred to a (maybe compressed) block with trailer, or a maybe compressed block maybe without trailer. (Note: left as follow-up work to use C++ typing to better sort out the various kinds of BlockContents.) This change primarily tries to apply some consistent terminology around the kinds of block representations, avoiding the unclear "raw". (Any meaning of "raw" assumes some bias toward the storage layer or toward the logical data layer.) Preferred terminology: * **Serialized block** - bytes that go into storage. For block-based table (usually the case) this includes the block trailer. WART: block `size` may or may not include the trailer; need to be clear about whether it does or not. * **Maybe compressed block** - like a serialized block, but without the trailer (or no promise of including a trailer). Must be accompanied by a CompressionType. * **Uncompressed block** - "payload" bytes that are either stored with no compression, used as input to compression function, or result of decompression function. * **Parsed block** - an in-memory form of a block in block cache, as it is used by the table reader. Different C++ types are used depending on the block type (see block_like_traits.h). Other refactorings: * Misc corrections/improvements of internal API comments * Remove a few misleading / unhelpful / redundant comments. * Use move semantics in some places to simplify contracts * Use better parameter names to indicate which parameters are used for outputs * Remove some extraneous `extern` * Various clean-ups to `CacheDumperImpl` (mostly unnecessary code) Pull Request resolved: https://github.com/facebook/rocksdb/pull/10408 Test Plan: existing tests Reviewed By: akankshamahajan15 Differential Revision: D38172617 Pulled By: pdillinger fbshipit-source-id: ccb99299f324ac5ca46996d34c5089621a4f260c
2 years ago
IOStatus WriteBlock(CacheDumpUnitType type, const Slice& key,
const Slice& value);
IOStatus WriteHeader();
IOStatus WriteFooter();
bool ShouldFilterOut(const Slice& key);
std::function<void(const Slice&, void*, size_t, Cache::DeleterFn)>
DumpOneBlockCallBack();
CacheDumpOptions options_;
std::shared_ptr<Cache> cache_;
std::unique_ptr<CacheDumpWriter> writer_;
Meta-internal folly integration with F14FastMap (#9546) Summary: Especially after updating to C++17, I don't see a compelling case for *requiring* any folly components in RocksDB. I was able to purge the existing hard dependencies, and it can be quite difficult to strip out non-trivial components from folly for use in RocksDB. (The prospect of doing that on F14 has changed my mind on the best approach here.) But this change creates an optional integration where we can plug in components from folly at compile time, starting here with F14FastMap to replace std::unordered_map when possible (probably no public APIs for example). I have replaced the biggest CPU users of std::unordered_map with compile-time pluggable UnorderedMap which will use F14FastMap when USE_FOLLY is set. USE_FOLLY is always set in the Meta-internal buck build, and a simulation of that is in the Makefile for public CI testing. A full folly build is not needed, but checking out the full folly repo is much simpler for getting the dependency, and anything else we might want to optionally integrate in the future. Some picky details: * I don't think the distributed mutex stuff is actually used, so it was easy to remove. * I implemented an alternative to `folly::constexpr_log2` (which is much easier in C++17 than C++11) so that I could pull out the hard dependencies on `ConstexprMath.h` * I had to add noexcept move constructors/operators to some types to make F14's complainUnlessNothrowMoveAndDestroy check happy, and I added a macro to make that easier in some common cases. * Updated Meta-internal buck build to use folly F14Map (always) No updates to HISTORY.md nor INSTALL.md as this is not (yet?) considered a production integration for open source users. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9546 Test Plan: CircleCI tests updated so that a couple of them use folly. Most internal unit & stress/crash tests updated to use Meta-internal latest folly. (Note: they should probably use buck but they currently use Makefile.) Example performance improvement: when filter partitions are pinned in cache, they are tracked by PartitionedFilterBlockReader::filter_map_ and we can build a test that exercises that heavily. Build DB with ``` TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -benchmarks=fillrandom -num=10000000 -disable_wal=1 -write_buffer_size=30000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -partition_index_and_filters ``` and test with (simultaneous runs with & without folly, ~20 times each to see convergence) ``` TEST_TMPDIR=/dev/shm/rocksdb ./db_bench_folly -readonly -use_existing_db -benchmarks=readrandom -num=10000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -partition_index_and_filters -duration=40 -pin_l0_filter_and_index_blocks_in_cache ``` Average ops/s no folly: 26229.2 Average ops/s with folly: 26853.3 (+2.4%) Reviewed By: ajkr Differential Revision: D34181736 Pulled By: pdillinger fbshipit-source-id: ffa6ad5104c2880321d8a1aa7187e00ab0d02e94
3 years ago
UnorderedMap<Cache::DeleterFn, CacheEntryRole> role_map_;
SystemClock* clock_;
uint32_t sequence_num_;
// The cache key prefix filter. Currently, we use db_session_id as the prefix,
// so using std::set to store the prefixes as filter is enough. Further
// improvement can be applied like BloomFilter or others to speedup the
// filtering.
std::set<std::string> prefix_filter_;
};
// The default implementation of CacheDumpedLoader
class CacheDumpedLoaderImpl : public CacheDumpedLoader {
public:
CacheDumpedLoaderImpl(const CacheDumpOptions& dump_options,
const BlockBasedTableOptions& /*toptions*/,
const std::shared_ptr<SecondaryCache>& secondary_cache,
std::unique_ptr<CacheDumpReader>&& reader)
: options_(dump_options),
secondary_cache_(secondary_cache),
reader_(std::move(reader)) {}
~CacheDumpedLoaderImpl() {}
IOStatus RestoreCacheEntriesToSecondaryCache() override;
private:
IOStatus ReadDumpUnitMeta(std::string* data, DumpUnitMeta* unit_meta);
IOStatus ReadDumpUnit(size_t len, std::string* data, DumpUnit* unit);
IOStatus ReadHeader(std::string* data, DumpUnit* dump_unit);
IOStatus ReadCacheBlock(std::string* data, DumpUnit* dump_unit);
CacheDumpOptions options_;
std::shared_ptr<SecondaryCache> secondary_cache_;
std::unique_ptr<CacheDumpReader> reader_;
Meta-internal folly integration with F14FastMap (#9546) Summary: Especially after updating to C++17, I don't see a compelling case for *requiring* any folly components in RocksDB. I was able to purge the existing hard dependencies, and it can be quite difficult to strip out non-trivial components from folly for use in RocksDB. (The prospect of doing that on F14 has changed my mind on the best approach here.) But this change creates an optional integration where we can plug in components from folly at compile time, starting here with F14FastMap to replace std::unordered_map when possible (probably no public APIs for example). I have replaced the biggest CPU users of std::unordered_map with compile-time pluggable UnorderedMap which will use F14FastMap when USE_FOLLY is set. USE_FOLLY is always set in the Meta-internal buck build, and a simulation of that is in the Makefile for public CI testing. A full folly build is not needed, but checking out the full folly repo is much simpler for getting the dependency, and anything else we might want to optionally integrate in the future. Some picky details: * I don't think the distributed mutex stuff is actually used, so it was easy to remove. * I implemented an alternative to `folly::constexpr_log2` (which is much easier in C++17 than C++11) so that I could pull out the hard dependencies on `ConstexprMath.h` * I had to add noexcept move constructors/operators to some types to make F14's complainUnlessNothrowMoveAndDestroy check happy, and I added a macro to make that easier in some common cases. * Updated Meta-internal buck build to use folly F14Map (always) No updates to HISTORY.md nor INSTALL.md as this is not (yet?) considered a production integration for open source users. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9546 Test Plan: CircleCI tests updated so that a couple of them use folly. Most internal unit & stress/crash tests updated to use Meta-internal latest folly. (Note: they should probably use buck but they currently use Makefile.) Example performance improvement: when filter partitions are pinned in cache, they are tracked by PartitionedFilterBlockReader::filter_map_ and we can build a test that exercises that heavily. Build DB with ``` TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -benchmarks=fillrandom -num=10000000 -disable_wal=1 -write_buffer_size=30000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -partition_index_and_filters ``` and test with (simultaneous runs with & without folly, ~20 times each to see convergence) ``` TEST_TMPDIR=/dev/shm/rocksdb ./db_bench_folly -readonly -use_existing_db -benchmarks=readrandom -num=10000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -partition_index_and_filters -duration=40 -pin_l0_filter_and_index_blocks_in_cache ``` Average ops/s no folly: 26229.2 Average ops/s with folly: 26853.3 (+2.4%) Reviewed By: ajkr Differential Revision: D34181736 Pulled By: pdillinger fbshipit-source-id: ffa6ad5104c2880321d8a1aa7187e00ab0d02e94
3 years ago
UnorderedMap<Cache::DeleterFn, CacheEntryRole> role_map_;
};
// The default implementation of CacheDumpWriter. We write the blocks to a file
// sequentially.
class ToFileCacheDumpWriter : public CacheDumpWriter {
public:
explicit ToFileCacheDumpWriter(
std::unique_ptr<WritableFileWriter>&& file_writer)
: file_writer_(std::move(file_writer)) {}
~ToFileCacheDumpWriter() { Close().PermitUncheckedError(); }
Refactor to avoid confusing "raw block" (#10408) Summary: We have a lot of confusing code because of mixed, sometimes completely opposite uses of of the term "raw block" or "raw contents", sometimes within the same source file. For example, in `BlockBasedTableBuilder`, `raw_block_contents` and `raw_size` generally referred to uncompressed block contents and size, while `WriteRawBlock` referred to writing a block that is already compressed if it is going to be. Meanwhile, in `BlockBasedTable`, `raw_block_contents` either referred to a (maybe compressed) block with trailer, or a maybe compressed block maybe without trailer. (Note: left as follow-up work to use C++ typing to better sort out the various kinds of BlockContents.) This change primarily tries to apply some consistent terminology around the kinds of block representations, avoiding the unclear "raw". (Any meaning of "raw" assumes some bias toward the storage layer or toward the logical data layer.) Preferred terminology: * **Serialized block** - bytes that go into storage. For block-based table (usually the case) this includes the block trailer. WART: block `size` may or may not include the trailer; need to be clear about whether it does or not. * **Maybe compressed block** - like a serialized block, but without the trailer (or no promise of including a trailer). Must be accompanied by a CompressionType. * **Uncompressed block** - "payload" bytes that are either stored with no compression, used as input to compression function, or result of decompression function. * **Parsed block** - an in-memory form of a block in block cache, as it is used by the table reader. Different C++ types are used depending on the block type (see block_like_traits.h). Other refactorings: * Misc corrections/improvements of internal API comments * Remove a few misleading / unhelpful / redundant comments. * Use move semantics in some places to simplify contracts * Use better parameter names to indicate which parameters are used for outputs * Remove some extraneous `extern` * Various clean-ups to `CacheDumperImpl` (mostly unnecessary code) Pull Request resolved: https://github.com/facebook/rocksdb/pull/10408 Test Plan: existing tests Reviewed By: akankshamahajan15 Differential Revision: D38172617 Pulled By: pdillinger fbshipit-source-id: ccb99299f324ac5ca46996d34c5089621a4f260c
2 years ago
// Write the serialized metadata to the file
virtual IOStatus WriteMetadata(const Slice& metadata) override {
assert(file_writer_ != nullptr);
std::string prefix;
PutFixed32(&prefix, static_cast<uint32_t>(metadata.size()));
IOStatus io_s = file_writer_->Append(Slice(prefix));
if (!io_s.ok()) {
return io_s;
}
io_s = file_writer_->Append(metadata);
return io_s;
}
Refactor to avoid confusing "raw block" (#10408) Summary: We have a lot of confusing code because of mixed, sometimes completely opposite uses of of the term "raw block" or "raw contents", sometimes within the same source file. For example, in `BlockBasedTableBuilder`, `raw_block_contents` and `raw_size` generally referred to uncompressed block contents and size, while `WriteRawBlock` referred to writing a block that is already compressed if it is going to be. Meanwhile, in `BlockBasedTable`, `raw_block_contents` either referred to a (maybe compressed) block with trailer, or a maybe compressed block maybe without trailer. (Note: left as follow-up work to use C++ typing to better sort out the various kinds of BlockContents.) This change primarily tries to apply some consistent terminology around the kinds of block representations, avoiding the unclear "raw". (Any meaning of "raw" assumes some bias toward the storage layer or toward the logical data layer.) Preferred terminology: * **Serialized block** - bytes that go into storage. For block-based table (usually the case) this includes the block trailer. WART: block `size` may or may not include the trailer; need to be clear about whether it does or not. * **Maybe compressed block** - like a serialized block, but without the trailer (or no promise of including a trailer). Must be accompanied by a CompressionType. * **Uncompressed block** - "payload" bytes that are either stored with no compression, used as input to compression function, or result of decompression function. * **Parsed block** - an in-memory form of a block in block cache, as it is used by the table reader. Different C++ types are used depending on the block type (see block_like_traits.h). Other refactorings: * Misc corrections/improvements of internal API comments * Remove a few misleading / unhelpful / redundant comments. * Use move semantics in some places to simplify contracts * Use better parameter names to indicate which parameters are used for outputs * Remove some extraneous `extern` * Various clean-ups to `CacheDumperImpl` (mostly unnecessary code) Pull Request resolved: https://github.com/facebook/rocksdb/pull/10408 Test Plan: existing tests Reviewed By: akankshamahajan15 Differential Revision: D38172617 Pulled By: pdillinger fbshipit-source-id: ccb99299f324ac5ca46996d34c5089621a4f260c
2 years ago
// Write the serialized data to the file
virtual IOStatus WritePacket(const Slice& data) override {
assert(file_writer_ != nullptr);
std::string prefix;
PutFixed32(&prefix, static_cast<uint32_t>(data.size()));
IOStatus io_s = file_writer_->Append(Slice(prefix));
if (!io_s.ok()) {
return io_s;
}
io_s = file_writer_->Append(data);
return io_s;
}
// Reset the writer
virtual IOStatus Close() override {
file_writer_.reset();
return IOStatus::OK();
}
private:
std::unique_ptr<WritableFileWriter> file_writer_;
};
// The default implementation of CacheDumpReader. It is implemented based on
// RandomAccessFileReader. Note that, we keep an internal variable to remember
// the current offset.
class FromFileCacheDumpReader : public CacheDumpReader {
public:
explicit FromFileCacheDumpReader(
std::unique_ptr<RandomAccessFileReader>&& reader)
: file_reader_(std::move(reader)),
offset_(0),
buffer_(new char[kDumpReaderBufferSize]) {}
~FromFileCacheDumpReader() { delete[] buffer_; }
virtual IOStatus ReadMetadata(std::string* metadata) override {
uint32_t metadata_len = 0;
IOStatus io_s = ReadSizePrefix(&metadata_len);
if (!io_s.ok()) {
return io_s;
}
return Read(metadata_len, metadata);
}
virtual IOStatus ReadPacket(std::string* data) override {
uint32_t data_len = 0;
IOStatus io_s = ReadSizePrefix(&data_len);
if (!io_s.ok()) {
return io_s;
}
return Read(data_len, data);
}
private:
IOStatus ReadSizePrefix(uint32_t* len) {
std::string prefix;
IOStatus io_s = Read(kSizePrefixLen, &prefix);
if (!io_s.ok()) {
return io_s;
}
Slice encoded_slice(prefix);
if (!GetFixed32(&encoded_slice, len)) {
return IOStatus::Corruption("Decode size prefix string failed");
}
return IOStatus::OK();
}
IOStatus Read(size_t len, std::string* data) {
assert(file_reader_ != nullptr);
IOStatus io_s;
unsigned int bytes_to_read = static_cast<unsigned int>(len);
unsigned int to_read = bytes_to_read > kDumpReaderBufferSize
? kDumpReaderBufferSize
: bytes_to_read;
while (to_read > 0) {
io_s = file_reader_->Read(IOOptions(), offset_, to_read, &result_,
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
buffer_, nullptr,
Env::IO_TOTAL /* rate_limiter_priority */);
if (!io_s.ok()) {
return io_s;
}
if (result_.size() < to_read) {
return IOStatus::Corruption("Corrupted cache dump file.");
}
data->append(result_.data(), result_.size());
offset_ += to_read;
bytes_to_read -= to_read;
to_read = bytes_to_read > kDumpReaderBufferSize ? kDumpReaderBufferSize
: bytes_to_read;
}
return io_s;
}
std::unique_ptr<RandomAccessFileReader> file_reader_;
Slice result_;
size_t offset_;
char* buffer_;
};
// The cache dump and load helper class
class CacheDumperHelper {
public:
Refactor to avoid confusing "raw block" (#10408) Summary: We have a lot of confusing code because of mixed, sometimes completely opposite uses of of the term "raw block" or "raw contents", sometimes within the same source file. For example, in `BlockBasedTableBuilder`, `raw_block_contents` and `raw_size` generally referred to uncompressed block contents and size, while `WriteRawBlock` referred to writing a block that is already compressed if it is going to be. Meanwhile, in `BlockBasedTable`, `raw_block_contents` either referred to a (maybe compressed) block with trailer, or a maybe compressed block maybe without trailer. (Note: left as follow-up work to use C++ typing to better sort out the various kinds of BlockContents.) This change primarily tries to apply some consistent terminology around the kinds of block representations, avoiding the unclear "raw". (Any meaning of "raw" assumes some bias toward the storage layer or toward the logical data layer.) Preferred terminology: * **Serialized block** - bytes that go into storage. For block-based table (usually the case) this includes the block trailer. WART: block `size` may or may not include the trailer; need to be clear about whether it does or not. * **Maybe compressed block** - like a serialized block, but without the trailer (or no promise of including a trailer). Must be accompanied by a CompressionType. * **Uncompressed block** - "payload" bytes that are either stored with no compression, used as input to compression function, or result of decompression function. * **Parsed block** - an in-memory form of a block in block cache, as it is used by the table reader. Different C++ types are used depending on the block type (see block_like_traits.h). Other refactorings: * Misc corrections/improvements of internal API comments * Remove a few misleading / unhelpful / redundant comments. * Use move semantics in some places to simplify contracts * Use better parameter names to indicate which parameters are used for outputs * Remove some extraneous `extern` * Various clean-ups to `CacheDumperImpl` (mostly unnecessary code) Pull Request resolved: https://github.com/facebook/rocksdb/pull/10408 Test Plan: existing tests Reviewed By: akankshamahajan15 Differential Revision: D38172617 Pulled By: pdillinger fbshipit-source-id: ccb99299f324ac5ca46996d34c5089621a4f260c
2 years ago
// serialize the dump_unit_meta to a string, it is fixed 16 bytes size.
static void EncodeDumpUnitMeta(const DumpUnitMeta& meta, std::string* data) {
assert(data);
PutFixed32(data, static_cast<uint32_t>(meta.sequence_num));
PutFixed32(data, static_cast<uint32_t>(meta.dump_unit_checksum));
PutFixed64(data, meta.dump_unit_size);
}
Refactor to avoid confusing "raw block" (#10408) Summary: We have a lot of confusing code because of mixed, sometimes completely opposite uses of of the term "raw block" or "raw contents", sometimes within the same source file. For example, in `BlockBasedTableBuilder`, `raw_block_contents` and `raw_size` generally referred to uncompressed block contents and size, while `WriteRawBlock` referred to writing a block that is already compressed if it is going to be. Meanwhile, in `BlockBasedTable`, `raw_block_contents` either referred to a (maybe compressed) block with trailer, or a maybe compressed block maybe without trailer. (Note: left as follow-up work to use C++ typing to better sort out the various kinds of BlockContents.) This change primarily tries to apply some consistent terminology around the kinds of block representations, avoiding the unclear "raw". (Any meaning of "raw" assumes some bias toward the storage layer or toward the logical data layer.) Preferred terminology: * **Serialized block** - bytes that go into storage. For block-based table (usually the case) this includes the block trailer. WART: block `size` may or may not include the trailer; need to be clear about whether it does or not. * **Maybe compressed block** - like a serialized block, but without the trailer (or no promise of including a trailer). Must be accompanied by a CompressionType. * **Uncompressed block** - "payload" bytes that are either stored with no compression, used as input to compression function, or result of decompression function. * **Parsed block** - an in-memory form of a block in block cache, as it is used by the table reader. Different C++ types are used depending on the block type (see block_like_traits.h). Other refactorings: * Misc corrections/improvements of internal API comments * Remove a few misleading / unhelpful / redundant comments. * Use move semantics in some places to simplify contracts * Use better parameter names to indicate which parameters are used for outputs * Remove some extraneous `extern` * Various clean-ups to `CacheDumperImpl` (mostly unnecessary code) Pull Request resolved: https://github.com/facebook/rocksdb/pull/10408 Test Plan: existing tests Reviewed By: akankshamahajan15 Differential Revision: D38172617 Pulled By: pdillinger fbshipit-source-id: ccb99299f324ac5ca46996d34c5089621a4f260c
2 years ago
// Serialize the dump_unit to a string.
static void EncodeDumpUnit(const DumpUnit& dump_unit, std::string* data) {
assert(data);
PutFixed64(data, dump_unit.timestamp);
data->push_back(dump_unit.type);
PutLengthPrefixedSlice(data, dump_unit.key);
PutFixed32(data, static_cast<uint32_t>(dump_unit.value_len));
PutFixed32(data, dump_unit.value_checksum);
PutLengthPrefixedSlice(data,
Slice((char*)dump_unit.value, dump_unit.value_len));
}
Refactor to avoid confusing "raw block" (#10408) Summary: We have a lot of confusing code because of mixed, sometimes completely opposite uses of of the term "raw block" or "raw contents", sometimes within the same source file. For example, in `BlockBasedTableBuilder`, `raw_block_contents` and `raw_size` generally referred to uncompressed block contents and size, while `WriteRawBlock` referred to writing a block that is already compressed if it is going to be. Meanwhile, in `BlockBasedTable`, `raw_block_contents` either referred to a (maybe compressed) block with trailer, or a maybe compressed block maybe without trailer. (Note: left as follow-up work to use C++ typing to better sort out the various kinds of BlockContents.) This change primarily tries to apply some consistent terminology around the kinds of block representations, avoiding the unclear "raw". (Any meaning of "raw" assumes some bias toward the storage layer or toward the logical data layer.) Preferred terminology: * **Serialized block** - bytes that go into storage. For block-based table (usually the case) this includes the block trailer. WART: block `size` may or may not include the trailer; need to be clear about whether it does or not. * **Maybe compressed block** - like a serialized block, but without the trailer (or no promise of including a trailer). Must be accompanied by a CompressionType. * **Uncompressed block** - "payload" bytes that are either stored with no compression, used as input to compression function, or result of decompression function. * **Parsed block** - an in-memory form of a block in block cache, as it is used by the table reader. Different C++ types are used depending on the block type (see block_like_traits.h). Other refactorings: * Misc corrections/improvements of internal API comments * Remove a few misleading / unhelpful / redundant comments. * Use move semantics in some places to simplify contracts * Use better parameter names to indicate which parameters are used for outputs * Remove some extraneous `extern` * Various clean-ups to `CacheDumperImpl` (mostly unnecessary code) Pull Request resolved: https://github.com/facebook/rocksdb/pull/10408 Test Plan: existing tests Reviewed By: akankshamahajan15 Differential Revision: D38172617 Pulled By: pdillinger fbshipit-source-id: ccb99299f324ac5ca46996d34c5089621a4f260c
2 years ago
// Deserialize the dump_unit_meta from a string
static Status DecodeDumpUnitMeta(const std::string& encoded_data,
DumpUnitMeta* unit_meta) {
assert(unit_meta != nullptr);
Slice encoded_slice = Slice(encoded_data);
if (!GetFixed32(&encoded_slice, &(unit_meta->sequence_num))) {
return Status::Incomplete("Decode dumped unit meta sequence_num failed");
}
if (!GetFixed32(&encoded_slice, &(unit_meta->dump_unit_checksum))) {
return Status::Incomplete(
"Decode dumped unit meta dump_unit_checksum failed");
}
if (!GetFixed64(&encoded_slice, &(unit_meta->dump_unit_size))) {
return Status::Incomplete(
"Decode dumped unit meta dump_unit_size failed");
}
return Status::OK();
}
Refactor to avoid confusing "raw block" (#10408) Summary: We have a lot of confusing code because of mixed, sometimes completely opposite uses of of the term "raw block" or "raw contents", sometimes within the same source file. For example, in `BlockBasedTableBuilder`, `raw_block_contents` and `raw_size` generally referred to uncompressed block contents and size, while `WriteRawBlock` referred to writing a block that is already compressed if it is going to be. Meanwhile, in `BlockBasedTable`, `raw_block_contents` either referred to a (maybe compressed) block with trailer, or a maybe compressed block maybe without trailer. (Note: left as follow-up work to use C++ typing to better sort out the various kinds of BlockContents.) This change primarily tries to apply some consistent terminology around the kinds of block representations, avoiding the unclear "raw". (Any meaning of "raw" assumes some bias toward the storage layer or toward the logical data layer.) Preferred terminology: * **Serialized block** - bytes that go into storage. For block-based table (usually the case) this includes the block trailer. WART: block `size` may or may not include the trailer; need to be clear about whether it does or not. * **Maybe compressed block** - like a serialized block, but without the trailer (or no promise of including a trailer). Must be accompanied by a CompressionType. * **Uncompressed block** - "payload" bytes that are either stored with no compression, used as input to compression function, or result of decompression function. * **Parsed block** - an in-memory form of a block in block cache, as it is used by the table reader. Different C++ types are used depending on the block type (see block_like_traits.h). Other refactorings: * Misc corrections/improvements of internal API comments * Remove a few misleading / unhelpful / redundant comments. * Use move semantics in some places to simplify contracts * Use better parameter names to indicate which parameters are used for outputs * Remove some extraneous `extern` * Various clean-ups to `CacheDumperImpl` (mostly unnecessary code) Pull Request resolved: https://github.com/facebook/rocksdb/pull/10408 Test Plan: existing tests Reviewed By: akankshamahajan15 Differential Revision: D38172617 Pulled By: pdillinger fbshipit-source-id: ccb99299f324ac5ca46996d34c5089621a4f260c
2 years ago
// Deserialize the dump_unit from a string.
static Status DecodeDumpUnit(const std::string& encoded_data,
DumpUnit* dump_unit) {
assert(dump_unit != nullptr);
Slice encoded_slice = Slice(encoded_data);
// Decode timestamp
if (!GetFixed64(&encoded_slice, &dump_unit->timestamp)) {
return Status::Incomplete("Decode dumped unit string failed");
}
// Decode the block type
dump_unit->type = static_cast<CacheDumpUnitType>(encoded_slice[0]);
encoded_slice.remove_prefix(1);
// Decode the key
if (!GetLengthPrefixedSlice(&encoded_slice, &(dump_unit->key))) {
return Status::Incomplete("Decode dumped unit string failed");
}
// Decode the value size
uint32_t value_len;
if (!GetFixed32(&encoded_slice, &value_len)) {
return Status::Incomplete("Decode dumped unit string failed");
}
dump_unit->value_len = static_cast<size_t>(value_len);
// Decode the value checksum
if (!GetFixed32(&encoded_slice, &(dump_unit->value_checksum))) {
return Status::Incomplete("Decode dumped unit string failed");
}
// Decode the block content and copy to the memory space whose pointer
// will be managed by the cache finally.
Slice block;
if (!GetLengthPrefixedSlice(&encoded_slice, &block)) {
return Status::Incomplete("Decode dumped unit string failed");
}
dump_unit->value = (void*)block.data();
assert(block.size() == dump_unit->value_len);
return Status::OK();
}
};
} // namespace ROCKSDB_NAMESPACE
#endif // ROCKSDB_LITE