rocksdb/util/thread_operation.h
Hui Xiao 9a034801ce Group rocksdb.sst.read.micros stat by different user read IOActivity + misc (#11444)
Summary:
**Context/Summary:**
- Similar to https://github.com/facebook/rocksdb/pull/11288 but for user read such as `Get(), MultiGet(), DBIterator::XXX(), Verify(File)Checksum()`.
   - For this, I refactored some user-facing `MultiGet` calls in `TransactionBase` and various types of `DB` so that it does not call a user-facing `Get()` but `GetImpl()` for passing the `ReadOptions::io_activity` check (see PR conversation)
   - New user read stats breakdown are guarded by `kExceptDetailedTimers` since measurement shows they have 4-5% regression to the upstream/main.

- Misc
   - More refactoring: with https://github.com/facebook/rocksdb/pull/11288, we complete passing `ReadOptions/IOOptions` to FS level. So we can now replace the previously [added](https://github.com/facebook/rocksdb/pull/9424) `rate_limiter_priority` parameter in `RandomAccessFileReader`'s `Read/MultiRead/Prefetch()` with `IOOptions::rate_limiter_priority`
   - Also, `ReadAsync()` call time is measured in `SST_READ_MICRO` now

Pull Request resolved: https://github.com/facebook/rocksdb/pull/11444

Test Plan:
- CI fake db crash/stress test
- Microbenchmarking

**Build** `make clean && ROCKSDB_NO_FBCODE=1 DEBUG_LEVEL=0 make -jN db_basic_bench`
- google benchmark version: 604f6fd3f4
- db_basic_bench_base: upstream
- db_basic_bench_pr: db_basic_bench_base + this PR
- asyncread_db_basic_bench_base: upstream + [db basic bench patch for IteratorNext](https://github.com/facebook/rocksdb/compare/main...hx235:rocksdb:micro_bench_async_read)
- asyncread_db_basic_bench_pr: asyncread_db_basic_bench_base + this PR

**Test**

Get
```
TEST_TMPDIR=/dev/shm ./db_basic_bench_{null_stat|base|pr} --benchmark_filter=DBGet/comp_style:0/max_data:134217728/per_key_size:256/enable_statistics:1/negative_query:0/enable_filter:0/mmap:1/threads:1 --benchmark_repetitions=1000
```

Result
```
Coming soon
```

AsyncRead
```
TEST_TMPDIR=/dev/shm ./asyncread_db_basic_bench_{base|pr} --benchmark_filter=IteratorNext/comp_style:0/max_data:134217728/per_key_size:256/enable_statistics:1/async_io:1/include_detailed_timers:0 --benchmark_repetitions=1000 > syncread_db_basic_bench_{base|pr}.out
```

Result
```
Base:
1956,1956,1968,1977,1979,1986,1988,1988,1988,1990,1991,1991,1993,1993,1993,1993,1994,1996,1997,1997,1997,1998,1999,2001,2001,2002,2004,2007,2007,2008,

PR (2.3% regression, due to measuring `SST_READ_MICRO` that wasn't measured before):
1993,2014,2016,2022,2024,2027,2027,2028,2028,2030,2031,2031,2032,2032,2038,2039,2042,2044,2044,2047,2047,2047,2048,2049,2050,2052,2052,2052,2053,2053,
```

Reviewed By: ajkr

Differential Revision: D45918925

Pulled By: hx235

fbshipit-source-id: 58a54560d9ebeb3a59b6d807639692614dad058a
2023-08-08 17:26:50 -07:00

121 lines
4.2 KiB
C++

// 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).
//
// This file defines the structures for thread operation and state.
// Thread operations are used to describe high level action of a
// thread such as doing compaction or flush, while thread state
// are used to describe lower-level action such as reading /
// writing a file or waiting for a mutex. Operations and states
// are designed to be independent. Typically, a thread usually involves
// in one operation and one state at any specific point in time.
#pragma once
#include <string>
#include "rocksdb/thread_status.h"
namespace ROCKSDB_NAMESPACE {
#ifdef ROCKSDB_USING_THREAD_STATUS
// The structure that describes a major thread operation.
struct OperationInfo {
const ThreadStatus::OperationType type;
const std::string name;
};
// The global operation table.
//
// When updating a status of a thread, the pointer of the OperationInfo
// of the current ThreadStatusData will be pointing to one of the
// rows in this global table.
//
// Note that it's not designed to be constant as in the future we
// might consider adding global count to the OperationInfo.
static OperationInfo global_operation_table[] = {
{ThreadStatus::OP_UNKNOWN, ""},
{ThreadStatus::OP_COMPACTION, "Compaction"},
{ThreadStatus::OP_FLUSH, "Flush"},
{ThreadStatus::OP_DBOPEN, "DBOpen"},
{ThreadStatus::OP_GET, "Get"},
{ThreadStatus::OP_MULTIGET, "MultiGet"},
{ThreadStatus::OP_DBITERATOR, "DBIterator"},
{ThreadStatus::OP_VERIFY_DB_CHECKSUM, "VerifyDBChecksum"},
{ThreadStatus::OP_VERIFY_FILE_CHECKSUMS, "VerifyFileChecksums"},
};
struct OperationStageInfo {
const ThreadStatus::OperationStage stage;
const std::string name;
};
// A table maintains the mapping from stage type to stage string.
// Note that the string must be changed accordingly when the
// associated function name changed.
static OperationStageInfo global_op_stage_table[] = {
{ThreadStatus::STAGE_UNKNOWN, ""},
{ThreadStatus::STAGE_FLUSH_RUN, "FlushJob::Run"},
{ThreadStatus::STAGE_FLUSH_WRITE_L0, "FlushJob::WriteLevel0Table"},
{ThreadStatus::STAGE_COMPACTION_PREPARE, "CompactionJob::Prepare"},
{ThreadStatus::STAGE_COMPACTION_RUN, "CompactionJob::Run"},
{ThreadStatus::STAGE_COMPACTION_PROCESS_KV,
"CompactionJob::ProcessKeyValueCompaction"},
{ThreadStatus::STAGE_COMPACTION_INSTALL, "CompactionJob::Install"},
{ThreadStatus::STAGE_COMPACTION_SYNC_FILE,
"CompactionJob::FinishCompactionOutputFile"},
{ThreadStatus::STAGE_PICK_MEMTABLES_TO_FLUSH,
"MemTableList::PickMemtablesToFlush"},
{ThreadStatus::STAGE_MEMTABLE_ROLLBACK,
"MemTableList::RollbackMemtableFlush"},
{ThreadStatus::STAGE_MEMTABLE_INSTALL_FLUSH_RESULTS,
"MemTableList::TryInstallMemtableFlushResults"},
};
// The structure that describes a state.
struct StateInfo {
const ThreadStatus::StateType type;
const std::string name;
};
// The global state table.
//
// When updating a status of a thread, the pointer of the StateInfo
// of the current ThreadStatusData will be pointing to one of the
// rows in this global table.
static StateInfo global_state_table[] = {
{ThreadStatus::STATE_UNKNOWN, ""},
{ThreadStatus::STATE_MUTEX_WAIT, "Mutex Wait"},
};
struct OperationProperty {
int code;
std::string name;
};
static OperationProperty compaction_operation_properties[] = {
{ThreadStatus::COMPACTION_JOB_ID, "JobID"},
{ThreadStatus::COMPACTION_INPUT_OUTPUT_LEVEL, "InputOutputLevel"},
{ThreadStatus::COMPACTION_PROP_FLAGS, "Manual/Deletion/Trivial"},
{ThreadStatus::COMPACTION_TOTAL_INPUT_BYTES, "TotalInputBytes"},
{ThreadStatus::COMPACTION_BYTES_READ, "BytesRead"},
{ThreadStatus::COMPACTION_BYTES_WRITTEN, "BytesWritten"},
};
static OperationProperty flush_operation_properties[] = {
{ThreadStatus::FLUSH_JOB_ID, "JobID"},
{ThreadStatus::FLUSH_BYTES_MEMTABLES, "BytesMemtables"},
{ThreadStatus::FLUSH_BYTES_WRITTEN, "BytesWritten"}};
#else
struct OperationInfo {};
struct StateInfo {};
#endif // ROCKSDB_USING_THREAD_STATUS
} // namespace ROCKSDB_NAMESPACE