rocksdb/tools/simulated_hybrid_file_system.cc
Hui Xiao 06e593376c Group SST write in flush, compaction and db open with new stats (#11910)
Summary:
## Context/Summary
Similar to https://github.com/facebook/rocksdb/pull/11288, https://github.com/facebook/rocksdb/pull/11444, categorizing SST/blob file write according to different io activities allows more insight into the activity.

For that, this PR does the following:
- Tag different write IOs by passing down and converting WriteOptions to IOOptions
- Add new SST_WRITE_MICROS histogram in WritableFileWriter::Append() and breakdown FILE_WRITE_{FLUSH|COMPACTION|DB_OPEN}_MICROS

Some related code refactory to make implementation cleaner:
- Blob stats
   - Replace high-level write measurement with low-level WritableFileWriter::Append() measurement for BLOB_DB_BLOB_FILE_WRITE_MICROS. This is to make FILE_WRITE_{FLUSH|COMPACTION|DB_OPEN}_MICROS  include blob file. As a consequence, this introduces some behavioral changes on it, see HISTORY and db bench test plan below for more info.
   - Fix bugs where BLOB_DB_BLOB_FILE_SYNCED/BLOB_DB_BLOB_FILE_BYTES_WRITTEN include file failed to sync and bytes failed to write.
- Refactor WriteOptions constructor for easier construction with io_activity and rate_limiter_priority
- Refactor DBImpl::~DBImpl()/BlobDBImpl::Close() to bypass thread op verification
- Build table
   - TableBuilderOptions now includes Read/WriteOpitons so BuildTable() do not need to take these two variables
   - Replace the io_priority passed into BuildTable() with TableBuilderOptions::WriteOpitons::rate_limiter_priority. Similar for BlobFileBuilder.
This parameter is used for dynamically changing file io priority for flush, see  https://github.com/facebook/rocksdb/pull/9988?fbclid=IwAR1DtKel6c-bRJAdesGo0jsbztRtciByNlvokbxkV6h_L-AE9MACzqRTT5s for more
   - Update ThreadStatus::FLUSH_BYTES_WRITTEN to use io_activity to track flush IO in flush job and db open instead of io_priority

## Test
### db bench

Flush
```
./db_bench --statistics=1 --benchmarks=fillseq --num=100000 --write_buffer_size=100

rocksdb.sst.write.micros P50 : 1.830863 P95 : 4.094720 P99 : 6.578947 P100 : 26.000000 COUNT : 7875 SUM : 20377
rocksdb.file.write.flush.micros P50 : 1.830863 P95 : 4.094720 P99 : 6.578947 P100 : 26.000000 COUNT : 7875 SUM : 20377
rocksdb.file.write.compaction.micros P50 : 0.000000 P95 : 0.000000 P99 : 0.000000 P100 : 0.000000 COUNT : 0 SUM : 0
rocksdb.file.write.db.open.micros P50 : 0.000000 P95 : 0.000000 P99 : 0.000000 P100 : 0.000000 COUNT : 0 SUM : 0
```

compaction, db oopen
```
Setup: ./db_bench --statistics=1 --benchmarks=fillseq --num=10000 --disable_auto_compactions=1 -write_buffer_size=100 --db=../db_bench

Run:./db_bench --statistics=1 --benchmarks=compact  --db=../db_bench --use_existing_db=1

rocksdb.sst.write.micros P50 : 2.675325 P95 : 9.578788 P99 : 18.780000 P100 : 314.000000 COUNT : 638 SUM : 3279
rocksdb.file.write.flush.micros P50 : 0.000000 P95 : 0.000000 P99 : 0.000000 P100 : 0.000000 COUNT : 0 SUM : 0
rocksdb.file.write.compaction.micros P50 : 2.757353 P95 : 9.610687 P99 : 19.316667 P100 : 314.000000 COUNT : 615 SUM : 3213
rocksdb.file.write.db.open.micros P50 : 2.055556 P95 : 3.925000 P99 : 9.000000 P100 : 9.000000 COUNT : 23 SUM : 66
```

blob stats - just to make sure they aren't broken by this PR
```
Integrated Blob DB

Setup: ./db_bench --enable_blob_files=1 --statistics=1 --benchmarks=fillseq --num=10000 --disable_auto_compactions=1 -write_buffer_size=100 --db=../db_bench

Run:./db_bench --enable_blob_files=1 --statistics=1 --benchmarks=compact  --db=../db_bench --use_existing_db=1

pre-PR:
rocksdb.blobdb.blob.file.write.micros P50 : 7.298246 P95 : 9.771930 P99 : 9.991813 P100 : 16.000000 COUNT : 235 SUM : 1600
rocksdb.blobdb.blob.file.synced COUNT : 1
rocksdb.blobdb.blob.file.bytes.written COUNT : 34842

post-PR:
rocksdb.blobdb.blob.file.write.micros P50 : 2.000000 P95 : 2.829360 P99 : 2.993779 P100 : 9.000000 COUNT : 707 SUM : 1614
- COUNT is higher and values are smaller as it includes header and footer write
- COUNT is 3X higher due to each Append() count as one post-PR, while in pre-PR, 3 Append()s counts as one. See https://github.com/facebook/rocksdb/pull/11910/files#diff-32b811c0a1c000768cfb2532052b44dc0b3bf82253f3eab078e15ff201a0dabfL157-L164

rocksdb.blobdb.blob.file.synced COUNT : 1 (stay the same)
rocksdb.blobdb.blob.file.bytes.written COUNT : 34842 (stay the same)
```

```
Stacked Blob DB

Run: ./db_bench --use_blob_db=1 --statistics=1 --benchmarks=fillseq --num=10000 --disable_auto_compactions=1 -write_buffer_size=100 --db=../db_bench

pre-PR:
rocksdb.blobdb.blob.file.write.micros P50 : 12.808042 P95 : 19.674497 P99 : 28.539683 P100 : 51.000000 COUNT : 10000 SUM : 140876
rocksdb.blobdb.blob.file.synced COUNT : 8
rocksdb.blobdb.blob.file.bytes.written COUNT : 1043445

post-PR:
rocksdb.blobdb.blob.file.write.micros P50 : 1.657370 P95 : 2.952175 P99 : 3.877519 P100 : 24.000000 COUNT : 30001 SUM : 67924
- COUNT is higher and values are smaller as it includes header and footer write
- COUNT is 3X higher due to each Append() count as one post-PR, while in pre-PR, 3 Append()s counts as one. See https://github.com/facebook/rocksdb/pull/11910/files#diff-32b811c0a1c000768cfb2532052b44dc0b3bf82253f3eab078e15ff201a0dabfL157-L164

rocksdb.blobdb.blob.file.synced COUNT : 8 (stay the same)
rocksdb.blobdb.blob.file.bytes.written COUNT : 1043445 (stay the same)
```

###  Rehearsal CI stress test
Trigger 3 full runs of all our CI stress tests

###  Performance

Flush
```
TEST_TMPDIR=/dev/shm ./db_basic_bench_pre_pr --benchmark_filter=ManualFlush/key_num:524288/per_key_size:256 --benchmark_repetitions=1000
-- default: 1 thread is used to run benchmark; enable_statistics = true

Pre-pr: avg 507515519.3 ns
497686074,499444327,500862543,501389862,502994471,503744435,504142123,504224056,505724198,506610393,506837742,506955122,507695561,507929036,508307733,508312691,508999120,509963561,510142147,510698091,510743096,510769317,510957074,511053311,511371367,511409911,511432960,511642385,511691964,511730908,

Post-pr: avg 511971266.5 ns, regressed 0.88%
502744835,506502498,507735420,507929724,508313335,509548582,509994942,510107257,510715603,511046955,511352639,511458478,512117521,512317380,512766303,512972652,513059586,513804934,513808980,514059409,514187369,514389494,514447762,514616464,514622882,514641763,514666265,514716377,514990179,515502408,
```

Compaction
```
TEST_TMPDIR=/dev/shm ./db_basic_bench_{pre|post}_pr --benchmark_filter=ManualCompaction/comp_style:0/max_data:134217728/per_key_size:256/enable_statistics:1  --benchmark_repetitions=1000
-- default: 1 thread is used to run benchmark

Pre-pr: avg 495346098.30 ns
492118301,493203526,494201411,494336607,495269217,495404950,496402598,497012157,497358370,498153846

Post-pr: avg 504528077.20, regressed 1.85%. "ManualCompaction" include flush so the isolated regression for compaction should be around 1.85-0.88 = 0.97%
502465338,502485945,502541789,502909283,503438601,504143885,506113087,506629423,507160414,507393007
```

Put with WAL (in case passing WriteOptions slows down this path even without collecting SST write stats)
```
TEST_TMPDIR=/dev/shm ./db_basic_bench_pre_pr --benchmark_filter=DBPut/comp_style:0/max_data:107374182400/per_key_size:256/enable_statistics:1/wal:1  --benchmark_repetitions=1000
-- default: 1 thread is used to run benchmark

Pre-pr: avg 3848.10 ns
3814,3838,3839,3848,3854,3854,3854,3860,3860,3860

Post-pr: avg 3874.20 ns, regressed 0.68%
3863,3867,3871,3874,3875,3877,3877,3877,3880,3881
```

Pull Request resolved: https://github.com/facebook/rocksdb/pull/11910

Reviewed By: ajkr

Differential Revision: D49788060

Pulled By: hx235

fbshipit-source-id: 79e73699cda5be3b66461687e5147c2484fc5eff
2023-12-29 15:29:23 -08:00

245 lines
8.5 KiB
C++

// 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).
#include "util/stop_watch.h"
#include <algorithm>
#include <sstream>
#include <string>
#include "rocksdb/rate_limiter.h"
#include "tools/simulated_hybrid_file_system.h"
namespace ROCKSDB_NAMESPACE {
const int64_t kUsPerSec = 1000000;
const int64_t kDummyBytesPerUs = 1024;
namespace {
// From bytes to read/write, calculate service time needed by an HDD.
// This is used to simulate latency from HDD.
int CalculateServeTimeUs(size_t bytes) {
return 12200 + static_cast<int>(static_cast<double>(bytes) * 0.005215);
}
// There is a bug in rater limiter that would crash with small requests
// Hack to get it around.
void RateLimiterRequest(RateLimiter* rater_limiter, int64_t amount) {
int64_t left = amount * kDummyBytesPerUs;
const int64_t kMaxToRequest = kDummyBytesPerUs * kUsPerSec / 1024;
while (left > 0) {
int64_t to_request = std::min(kMaxToRequest, left);
rater_limiter->Request(to_request, Env::IOPriority::IO_LOW, nullptr);
left -= to_request;
}
}
} // namespace
// The metadata file format: each line is a full filename of a file which is
// warm
SimulatedHybridFileSystem::SimulatedHybridFileSystem(
const std::shared_ptr<FileSystem>& base,
const std::string& metadata_file_name, int throughput_multiplier,
bool is_full_fs_warm)
: FileSystemWrapper(base),
// Limit to 100 requests per second.
rate_limiter_(NewGenericRateLimiter(
int64_t{throughput_multiplier} * kDummyBytesPerUs *
kUsPerSec /* rate_bytes_per_sec */,
1000 /* refill_period_us */)),
metadata_file_name_(metadata_file_name),
name_("SimulatedHybridFileSystem: " + std::string(target()->Name())),
is_full_fs_warm_(is_full_fs_warm) {
IOStatus s = base->FileExists(metadata_file_name, IOOptions(), nullptr);
if (s.IsNotFound()) {
return;
}
std::string metadata;
s = ReadFileToString(base.get(), metadata_file_name, &metadata);
if (!s.ok()) {
fprintf(stderr, "Error reading from file %s: %s",
metadata_file_name.c_str(), s.ToString().c_str());
// Exit rather than assert as this file system is built to run with
// benchmarks, which usually run on release mode.
std::exit(1);
}
std::istringstream input;
input.str(metadata);
std::string line;
while (std::getline(input, line)) {
fprintf(stderr, "Warm file %s\n", line.c_str());
warm_file_set_.insert(line);
}
}
// Need to write out the metadata file to file. See comment of
// SimulatedHybridFileSystem::SimulatedHybridFileSystem() for format of the
// file.
SimulatedHybridFileSystem::~SimulatedHybridFileSystem() {
if (metadata_file_name_.empty()) {
return;
}
std::string metadata;
for (const auto& f : warm_file_set_) {
metadata += f;
metadata += "\n";
}
IOOptions opts;
IOStatus s =
WriteStringToFile(target(), metadata, metadata_file_name_, true, opts);
if (!s.ok()) {
fprintf(stderr, "Error writing to file %s: %s", metadata_file_name_.c_str(),
s.ToString().c_str());
}
}
IOStatus SimulatedHybridFileSystem::NewRandomAccessFile(
const std::string& fname, const FileOptions& file_opts,
std::unique_ptr<FSRandomAccessFile>* result, IODebugContext* dbg) {
Temperature temperature = Temperature::kUnknown;
if (is_full_fs_warm_) {
temperature = Temperature::kWarm;
} else {
const std::lock_guard<std::mutex> lock(mutex_);
if (warm_file_set_.find(fname) != warm_file_set_.end()) {
temperature = Temperature::kWarm;
}
assert(temperature == file_opts.temperature);
}
IOStatus s = target()->NewRandomAccessFile(fname, file_opts, result, dbg);
result->reset(
new SimulatedHybridRaf(std::move(*result), rate_limiter_, temperature));
return s;
}
IOStatus SimulatedHybridFileSystem::NewWritableFile(
const std::string& fname, const FileOptions& file_opts,
std::unique_ptr<FSWritableFile>* result, IODebugContext* dbg) {
if (file_opts.temperature == Temperature::kWarm) {
const std::lock_guard<std::mutex> lock(mutex_);
warm_file_set_.insert(fname);
}
IOStatus s = target()->NewWritableFile(fname, file_opts, result, dbg);
if (file_opts.temperature == Temperature::kWarm || is_full_fs_warm_) {
result->reset(new SimulatedWritableFile(std::move(*result), rate_limiter_));
}
return s;
}
IOStatus SimulatedHybridFileSystem::DeleteFile(const std::string& fname,
const IOOptions& options,
IODebugContext* dbg) {
{
const std::lock_guard<std::mutex> lock(mutex_);
warm_file_set_.erase(fname);
}
return target()->DeleteFile(fname, options, dbg);
}
IOStatus SimulatedHybridRaf::Read(uint64_t offset, size_t n,
const IOOptions& options, Slice* result,
char* scratch, IODebugContext* dbg) const {
if (temperature_ == Temperature::kWarm) {
SimulateIOWait(n);
}
return target()->Read(offset, n, options, result, scratch, dbg);
}
IOStatus SimulatedHybridRaf::MultiRead(FSReadRequest* reqs, size_t num_reqs,
const IOOptions& options,
IODebugContext* dbg) {
if (temperature_ == Temperature::kWarm) {
for (size_t i = 0; i < num_reqs; i++) {
SimulateIOWait(reqs[i].len);
}
}
return target()->MultiRead(reqs, num_reqs, options, dbg);
}
IOStatus SimulatedHybridRaf::Prefetch(uint64_t offset, size_t n,
const IOOptions& options,
IODebugContext* dbg) {
if (temperature_ == Temperature::kWarm) {
SimulateIOWait(n);
}
return target()->Prefetch(offset, n, options, dbg);
}
void SimulatedHybridRaf::SimulateIOWait(int64_t bytes) const {
int serve_time = CalculateServeTimeUs(bytes);
{
StopWatchNano stop_watch(Env::Default()->GetSystemClock().get(),
/*auto_start=*/true);
RateLimiterRequest(rate_limiter_.get(), serve_time);
int time_passed_us = static_cast<int>(stop_watch.ElapsedNanos() / 1000);
if (time_passed_us < serve_time) {
Env::Default()->SleepForMicroseconds(serve_time - time_passed_us);
}
}
}
void SimulatedWritableFile::SimulateIOWait(int64_t bytes) const {
int serve_time = CalculateServeTimeUs(bytes);
Env::Default()->SleepForMicroseconds(serve_time);
RateLimiterRequest(rate_limiter_.get(), serve_time);
}
IOStatus SimulatedWritableFile::Append(const Slice& data, const IOOptions& ioo,
IODebugContext* idc) {
if (use_direct_io()) {
SimulateIOWait(data.size());
} else {
unsynced_bytes += data.size();
}
return target()->Append(data, ioo, idc);
}
IOStatus SimulatedWritableFile::Append(
const Slice& data, const IOOptions& options,
const DataVerificationInfo& verification_info, IODebugContext* dbg) {
if (use_direct_io()) {
SimulateIOWait(data.size());
} else {
unsynced_bytes += data.size();
}
return target()->Append(data, options, verification_info, dbg);
}
IOStatus SimulatedWritableFile::PositionedAppend(const Slice& data,
uint64_t offset,
const IOOptions& options,
IODebugContext* dbg) {
if (use_direct_io()) {
SimulateIOWait(data.size());
} else {
// This might be overcalculated, but it's probably OK.
unsynced_bytes += data.size();
}
return target()->PositionedAppend(data, offset, options, dbg);
}
IOStatus SimulatedWritableFile::PositionedAppend(
const Slice& data, uint64_t offset, const IOOptions& options,
const DataVerificationInfo& verification_info, IODebugContext* dbg) {
if (use_direct_io()) {
SimulateIOWait(data.size());
} else {
// This might be overcalculated, but it's probably OK.
unsynced_bytes += data.size();
}
return target()->PositionedAppend(data, offset, options, verification_info,
dbg);
}
IOStatus SimulatedWritableFile::Sync(const IOOptions& options,
IODebugContext* dbg) {
if (unsynced_bytes > 0) {
SimulateIOWait(unsynced_bytes);
unsynced_bytes = 0;
}
return target()->Sync(options, dbg);
}
} // namespace ROCKSDB_NAMESPACE