mirror of
https://github.com/facebook/rocksdb.git
synced 2024-11-29 18:33:58 +00:00
9a034801ce
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
107 lines
4.1 KiB
C++
107 lines
4.1 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).
|
|
|
|
#pragma once
|
|
|
|
#include "utilities/transactions/write_prepared_txn_db.h"
|
|
#include "utilities/transactions/write_unprepared_txn.h"
|
|
|
|
namespace ROCKSDB_NAMESPACE {
|
|
|
|
class WriteUnpreparedTxn;
|
|
|
|
class WriteUnpreparedTxnDB : public WritePreparedTxnDB {
|
|
public:
|
|
using WritePreparedTxnDB::WritePreparedTxnDB;
|
|
|
|
Status Initialize(const std::vector<size_t>& compaction_enabled_cf_indices,
|
|
const std::vector<ColumnFamilyHandle*>& handles) override;
|
|
|
|
Transaction* BeginTransaction(const WriteOptions& write_options,
|
|
const TransactionOptions& txn_options,
|
|
Transaction* old_txn) override;
|
|
|
|
// Struct to hold ownership of snapshot and read callback for cleanup.
|
|
struct IteratorState;
|
|
|
|
using WritePreparedTxnDB::NewIterator;
|
|
Iterator* NewIterator(const ReadOptions& _read_options,
|
|
ColumnFamilyHandle* column_family,
|
|
WriteUnpreparedTxn* txn);
|
|
|
|
private:
|
|
Status RollbackRecoveredTransaction(const DBImpl::RecoveredTransaction* rtxn);
|
|
};
|
|
|
|
class WriteUnpreparedCommitEntryPreReleaseCallback : public PreReleaseCallback {
|
|
// TODO(lth): Reduce code duplication with
|
|
// WritePreparedCommitEntryPreReleaseCallback
|
|
public:
|
|
// includes_data indicates that the commit also writes non-empty
|
|
// CommitTimeWriteBatch to memtable, which needs to be committed separately.
|
|
WriteUnpreparedCommitEntryPreReleaseCallback(
|
|
WritePreparedTxnDB* db, DBImpl* db_impl,
|
|
const std::map<SequenceNumber, size_t>& unprep_seqs,
|
|
size_t data_batch_cnt = 0, bool publish_seq = true)
|
|
: db_(db),
|
|
db_impl_(db_impl),
|
|
unprep_seqs_(unprep_seqs),
|
|
data_batch_cnt_(data_batch_cnt),
|
|
includes_data_(data_batch_cnt_ > 0),
|
|
publish_seq_(publish_seq) {
|
|
assert(unprep_seqs.size() > 0);
|
|
}
|
|
|
|
virtual Status Callback(SequenceNumber commit_seq,
|
|
bool is_mem_disabled __attribute__((__unused__)),
|
|
uint64_t, size_t /*index*/,
|
|
size_t /*total*/) override {
|
|
const uint64_t last_commit_seq = LIKELY(data_batch_cnt_ <= 1)
|
|
? commit_seq
|
|
: commit_seq + data_batch_cnt_ - 1;
|
|
// Recall that unprep_seqs maps (un)prepared_seq => prepare_batch_cnt.
|
|
for (const auto& s : unprep_seqs_) {
|
|
for (size_t i = 0; i < s.second; i++) {
|
|
db_->AddCommitted(s.first + i, last_commit_seq);
|
|
}
|
|
}
|
|
|
|
if (includes_data_) {
|
|
assert(data_batch_cnt_);
|
|
// Commit the data that is accompanied with the commit request
|
|
for (size_t i = 0; i < data_batch_cnt_; i++) {
|
|
// For commit seq of each batch use the commit seq of the last batch.
|
|
// This would make debugging easier by having all the batches having
|
|
// the same sequence number.
|
|
db_->AddCommitted(commit_seq + i, last_commit_seq);
|
|
}
|
|
}
|
|
if (db_impl_->immutable_db_options().two_write_queues && publish_seq_) {
|
|
assert(is_mem_disabled); // implies the 2nd queue
|
|
// Publish the sequence number. We can do that here assuming the callback
|
|
// is invoked only from one write queue, which would guarantee that the
|
|
// publish sequence numbers will be in order, i.e., once a seq is
|
|
// published all the seq prior to that are also publishable.
|
|
db_impl_->SetLastPublishedSequence(last_commit_seq);
|
|
}
|
|
// else SequenceNumber that is updated as part of the write already does the
|
|
// publishing
|
|
return Status::OK();
|
|
}
|
|
|
|
private:
|
|
WritePreparedTxnDB* db_;
|
|
DBImpl* db_impl_;
|
|
const std::map<SequenceNumber, size_t>& unprep_seqs_;
|
|
size_t data_batch_cnt_;
|
|
// Either because it is commit without prepare or it has a
|
|
// CommitTimeWriteBatch
|
|
bool includes_data_;
|
|
// Should the callback also publishes the commit seq number
|
|
bool publish_seq_;
|
|
};
|
|
|
|
} // namespace ROCKSDB_NAMESPACE
|