rocksdb/db/db_impl/db_impl_readonly.cc

270 lines
9.6 KiB
C++
Raw Normal View History

// 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).
#include "db/db_impl/db_impl_readonly.h"
#include "db/arena_wrapped_db_iter.h"
#include "db/compacted_db_impl.h"
#include "db/db_impl/db_impl.h"
#include "db/db_iter.h"
#include "db/merge_context.h"
#include "monitoring/perf_context_imp.h"
#include "util/cast_util.h"
namespace ROCKSDB_NAMESPACE {
#ifndef ROCKSDB_LITE
DBImplReadOnly::DBImplReadOnly(const DBOptions& db_options,
const std::string& dbname)
: DBImpl(db_options, dbname) {
ROCKS_LOG_INFO(immutable_db_options_.info_log,
"Opening the db in read only mode");
LogFlush(immutable_db_options_.info_log);
}
DBImplReadOnly::~DBImplReadOnly() {}
// Implementations of the DB interface
Status DBImplReadOnly::Get(const ReadOptions& read_options,
ColumnFamilyHandle* column_family, const Slice& key,
PinnableSlice* pinnable_val) {
assert(pinnable_val != nullptr);
// TODO: stopwatch DB_GET needed?, perf timer needed?
PERF_TIMER_GUARD(get_snapshot_time);
Status s;
SequenceNumber snapshot = versions_->LastSequence();
auto cfh = static_cast_with_check<ColumnFamilyHandleImpl>(column_family);
auto cfd = cfh->cfd();
if (tracer_) {
InstrumentedMutexLock lock(&trace_mutex_);
if (tracer_) {
tracer_->Get(column_family, key);
}
}
SuperVersion* super_version = cfd->GetSuperVersion();
MergeContext merge_context;
Use only "local" range tombstones during Get (#4449) Summary: Previously, range tombstones were accumulated from every level, which was necessary if a range tombstone in a higher level covered a key in a lower level. However, RangeDelAggregator::AddTombstones's complexity is based on the number of tombstones that are currently stored in it, which is wasteful in the Get case, where we only need to know the highest sequence number of range tombstones that cover the key from higher levels, and compute the highest covering sequence number at the current level. This change introduces this optimization, and removes the use of RangeDelAggregator from the Get path. In the benchmark results, the following command was used to initialize the database: ``` ./db_bench -db=/dev/shm/5k-rts -use_existing_db=false -benchmarks=filluniquerandom -write_buffer_size=1048576 -compression_type=lz4 -target_file_size_base=1048576 -max_bytes_for_level_base=4194304 -value_size=112 -key_size=16 -block_size=4096 -level_compaction_dynamic_level_bytes=true -num=5000000 -max_background_jobs=12 -benchmark_write_rate_limit=20971520 -range_tombstone_width=100 -writes_per_range_tombstone=100 -max_num_range_tombstones=50000 -bloom_bits=8 ``` ...and the following command was used to measure read throughput: ``` ./db_bench -db=/dev/shm/5k-rts/ -use_existing_db=true -benchmarks=readrandom -disable_auto_compactions=true -num=5000000 -reads=100000 -threads=32 ``` The filluniquerandom command was only run once, and the resulting database was used to measure read performance before and after the PR. Both binaries were compiled with `DEBUG_LEVEL=0`. Readrandom results before PR: ``` readrandom : 4.544 micros/op 220090 ops/sec; 16.9 MB/s (63103 of 100000 found) ``` Readrandom results after PR: ``` readrandom : 11.147 micros/op 89707 ops/sec; 6.9 MB/s (63103 of 100000 found) ``` So it's actually slower right now, but this PR paves the way for future optimizations (see #4493). ---- Pull Request resolved: https://github.com/facebook/rocksdb/pull/4449 Differential Revision: D10370575 Pulled By: abhimadan fbshipit-source-id: 9a2e152be1ef36969055c0e9eb4beb0d96c11f4d
2018-10-24 19:29:29 +00:00
SequenceNumber max_covering_tombstone_seq = 0;
LookupKey lkey(key, snapshot);
PERF_TIMER_STOP(get_snapshot_time);
return timestamp from get (#6409) Summary: Added new Get() methods that return timestamp. Dummy implementation is given so that classes derived from DB don't need to be touched to provide their implementation. MultiGet is not included. ReadRandom perf test (10 minutes) on the same development machine ram drive with the same DB data shows no regression (within marge of error). The test is adapted from https://github.com/facebook/rocksdb/wiki/RocksDB-In-Memory-Workload-Performance-Benchmarks. base line (commit 72ee067b9): 101.712 micros/op 314602 ops/sec; 36.0 MB/s (5658999 of 5658999 found) This PR: 100.288 micros/op 319071 ops/sec; 36.5 MB/s (5674999 of 5674999 found) ./db_bench --db=r:\rocksdb.github --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --cache_size=2147483648 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=134217728 --max_bytes_for_level_base=1073741824 --disable_wal=0 --wal_dir=r:\rocksdb.github\WAL_LOG --sync=0 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --duration=600 --benchmarks=readrandom --use_existing_db=1 --num=25000000 --threads=32 Pull Request resolved: https://github.com/facebook/rocksdb/pull/6409 Differential Revision: D20200086 Pulled By: riversand963 fbshipit-source-id: 490edd74d924f62bd8ae9c29c2a6bbbb8410ca50
2020-03-02 23:58:32 +00:00
if (super_version->mem->Get(lkey, pinnable_val->GetSelf(),
/*timestamp=*/nullptr, &s, &merge_context,
Use only "local" range tombstones during Get (#4449) Summary: Previously, range tombstones were accumulated from every level, which was necessary if a range tombstone in a higher level covered a key in a lower level. However, RangeDelAggregator::AddTombstones's complexity is based on the number of tombstones that are currently stored in it, which is wasteful in the Get case, where we only need to know the highest sequence number of range tombstones that cover the key from higher levels, and compute the highest covering sequence number at the current level. This change introduces this optimization, and removes the use of RangeDelAggregator from the Get path. In the benchmark results, the following command was used to initialize the database: ``` ./db_bench -db=/dev/shm/5k-rts -use_existing_db=false -benchmarks=filluniquerandom -write_buffer_size=1048576 -compression_type=lz4 -target_file_size_base=1048576 -max_bytes_for_level_base=4194304 -value_size=112 -key_size=16 -block_size=4096 -level_compaction_dynamic_level_bytes=true -num=5000000 -max_background_jobs=12 -benchmark_write_rate_limit=20971520 -range_tombstone_width=100 -writes_per_range_tombstone=100 -max_num_range_tombstones=50000 -bloom_bits=8 ``` ...and the following command was used to measure read throughput: ``` ./db_bench -db=/dev/shm/5k-rts/ -use_existing_db=true -benchmarks=readrandom -disable_auto_compactions=true -num=5000000 -reads=100000 -threads=32 ``` The filluniquerandom command was only run once, and the resulting database was used to measure read performance before and after the PR. Both binaries were compiled with `DEBUG_LEVEL=0`. Readrandom results before PR: ``` readrandom : 4.544 micros/op 220090 ops/sec; 16.9 MB/s (63103 of 100000 found) ``` Readrandom results after PR: ``` readrandom : 11.147 micros/op 89707 ops/sec; 6.9 MB/s (63103 of 100000 found) ``` So it's actually slower right now, but this PR paves the way for future optimizations (see #4493). ---- Pull Request resolved: https://github.com/facebook/rocksdb/pull/4449 Differential Revision: D10370575 Pulled By: abhimadan fbshipit-source-id: 9a2e152be1ef36969055c0e9eb4beb0d96c11f4d
2018-10-24 19:29:29 +00:00
&max_covering_tombstone_seq, read_options)) {
pinnable_val->PinSelf();
RecordTick(stats_, MEMTABLE_HIT);
} else {
PERF_TIMER_GUARD(get_from_output_files_time);
return timestamp from get (#6409) Summary: Added new Get() methods that return timestamp. Dummy implementation is given so that classes derived from DB don't need to be touched to provide their implementation. MultiGet is not included. ReadRandom perf test (10 minutes) on the same development machine ram drive with the same DB data shows no regression (within marge of error). The test is adapted from https://github.com/facebook/rocksdb/wiki/RocksDB-In-Memory-Workload-Performance-Benchmarks. base line (commit 72ee067b9): 101.712 micros/op 314602 ops/sec; 36.0 MB/s (5658999 of 5658999 found) This PR: 100.288 micros/op 319071 ops/sec; 36.5 MB/s (5674999 of 5674999 found) ./db_bench --db=r:\rocksdb.github --num_levels=6 --key_size=20 --prefix_size=20 --keys_per_prefix=0 --value_size=100 --cache_size=2147483648 --cache_numshardbits=6 --compression_type=none --compression_ratio=1 --min_level_to_compress=-1 --disable_seek_compaction=1 --hard_rate_limit=2 --write_buffer_size=134217728 --max_write_buffer_number=2 --level0_file_num_compaction_trigger=8 --target_file_size_base=134217728 --max_bytes_for_level_base=1073741824 --disable_wal=0 --wal_dir=r:\rocksdb.github\WAL_LOG --sync=0 --verify_checksum=1 --delete_obsolete_files_period_micros=314572800 --max_background_compactions=4 --max_background_flushes=0 --level0_slowdown_writes_trigger=16 --level0_stop_writes_trigger=24 --statistics=0 --stats_per_interval=0 --stats_interval=1048576 --histogram=0 --use_plain_table=1 --open_files=-1 --mmap_read=1 --mmap_write=0 --memtablerep=prefix_hash --bloom_bits=10 --bloom_locality=1 --duration=600 --benchmarks=readrandom --use_existing_db=1 --num=25000000 --threads=32 Pull Request resolved: https://github.com/facebook/rocksdb/pull/6409 Differential Revision: D20200086 Pulled By: riversand963 fbshipit-source-id: 490edd74d924f62bd8ae9c29c2a6bbbb8410ca50
2020-03-02 23:58:32 +00:00
super_version->current->Get(read_options, lkey, pinnable_val,
/*timestamp=*/nullptr, &s, &merge_context,
&max_covering_tombstone_seq);
RecordTick(stats_, MEMTABLE_MISS);
}
RecordTick(stats_, NUMBER_KEYS_READ);
size_t size = pinnable_val->size();
RecordTick(stats_, BYTES_READ, size);
RecordInHistogram(stats_, BYTES_PER_READ, size);
PERF_COUNTER_ADD(get_read_bytes, size);
return s;
}
Iterator* DBImplReadOnly::NewIterator(const ReadOptions& read_options,
ColumnFamilyHandle* column_family) {
auto cfh = static_cast_with_check<ColumnFamilyHandleImpl>(column_family);
auto cfd = cfh->cfd();
SuperVersion* super_version = cfd->GetSuperVersion()->Ref();
SequenceNumber latest_snapshot = versions_->LastSequence();
SequenceNumber read_seq =
read_options.snapshot != nullptr
? reinterpret_cast<const SnapshotImpl*>(read_options.snapshot)
->number_
: latest_snapshot;
ReadCallback* read_callback = nullptr; // No read callback provided.
auto db_iter = NewArenaWrappedDbIterator(
env_, read_options, *cfd->ioptions(), super_version->mutable_cf_options,
read_seq,
super_version->mutable_cf_options.max_sequential_skip_in_iterations,
super_version->version_number, read_callback);
auto internal_iter =
NewInternalIterator(read_options, cfd, super_version, db_iter->GetArena(),
Properly report IO errors when IndexType::kBinarySearchWithFirstKey is used (#6621) Summary: Context: Index type `kBinarySearchWithFirstKey` added the ability for sst file iterator to sometimes report a key from index without reading the corresponding data block. This is useful when sst blocks are cut at some meaningful boundaries (e.g. one block per key prefix), and many seeks land between blocks (e.g. for each prefix, the ranges of keys in different sst files are nearly disjoint, so a typical seek needs to read a data block from only one file even if all files have the prefix). But this added a new error condition, which rocksdb code was really not equipped to deal with: `InternalIterator::value()` may fail with an IO error or Status::Incomplete, but it's just a method returning a Slice, with no way to report error instead. Before this PR, this type of error wasn't handled at all (an empty slice was returned), and kBinarySearchWithFirstKey implementation was considered a prototype. Now that we (LogDevice) have experimented with kBinarySearchWithFirstKey for a while and confirmed that it's really useful, this PR is adding the missing error handling. It's a pretty inconvenient situation implementation-wise. The error needs to be reported from InternalIterator when trying to access value. But there are ~700 call sites of `InternalIterator::value()`, most of which either can't hit the error condition (because the iterator is reading from memtable or from index or something) or wouldn't benefit from the deferred loading of the value (e.g. compaction iterator that reads all values anyway). Adding error handling to all these call sites would needlessly bloat the code. So instead I made the deferred value loading optional: only the call sites that may use deferred loading have to call the new method `PrepareValue()` before calling `value()`. The feature is enabled with a new bool argument `allow_unprepared_value` to a bunch of methods that create iterators (it wouldn't make sense to put it in ReadOptions because it's completely internal to iterators, with virtually no user-visible effect). Lmk if you have better ideas. Note that the deferred value loading only happens for *internal* iterators. The user-visible iterator (DBIter) always prepares the value before returning from Seek/Next/etc. We could go further and add an API to defer that value loading too, but that's most likely not useful for LogDevice, so it doesn't seem worth the complexity for now. Pull Request resolved: https://github.com/facebook/rocksdb/pull/6621 Test Plan: make -j5 check . Will also deploy to some logdevice test clusters and look at stats. Reviewed By: siying Differential Revision: D20786930 Pulled By: al13n321 fbshipit-source-id: 6da77d918bad3780522e918f17f4d5513d3e99ee
2020-04-16 00:37:23 +00:00
db_iter->GetRangeDelAggregator(), read_seq,
/* allow_unprepared_value */ true);
db_iter->SetIterUnderDBIter(internal_iter);
return db_iter;
}
Status DBImplReadOnly::NewIterators(
const ReadOptions& read_options,
const std::vector<ColumnFamilyHandle*>& column_families,
std::vector<Iterator*>* iterators) {
ReadCallback* read_callback = nullptr; // No read callback provided.
if (iterators == nullptr) {
return Status::InvalidArgument("iterators not allowed to be nullptr");
}
iterators->clear();
iterators->reserve(column_families.size());
SequenceNumber latest_snapshot = versions_->LastSequence();
SequenceNumber read_seq =
read_options.snapshot != nullptr
? reinterpret_cast<const SnapshotImpl*>(read_options.snapshot)
->number_
: latest_snapshot;
for (auto cfh : column_families) {
auto* cfd = static_cast_with_check<ColumnFamilyHandleImpl>(cfh)->cfd();
auto* sv = cfd->GetSuperVersion()->Ref();
auto* db_iter = NewArenaWrappedDbIterator(
env_, read_options, *cfd->ioptions(), sv->mutable_cf_options, read_seq,
sv->mutable_cf_options.max_sequential_skip_in_iterations,
sv->version_number, read_callback);
auto* internal_iter =
NewInternalIterator(read_options, cfd, sv, db_iter->GetArena(),
Properly report IO errors when IndexType::kBinarySearchWithFirstKey is used (#6621) Summary: Context: Index type `kBinarySearchWithFirstKey` added the ability for sst file iterator to sometimes report a key from index without reading the corresponding data block. This is useful when sst blocks are cut at some meaningful boundaries (e.g. one block per key prefix), and many seeks land between blocks (e.g. for each prefix, the ranges of keys in different sst files are nearly disjoint, so a typical seek needs to read a data block from only one file even if all files have the prefix). But this added a new error condition, which rocksdb code was really not equipped to deal with: `InternalIterator::value()` may fail with an IO error or Status::Incomplete, but it's just a method returning a Slice, with no way to report error instead. Before this PR, this type of error wasn't handled at all (an empty slice was returned), and kBinarySearchWithFirstKey implementation was considered a prototype. Now that we (LogDevice) have experimented with kBinarySearchWithFirstKey for a while and confirmed that it's really useful, this PR is adding the missing error handling. It's a pretty inconvenient situation implementation-wise. The error needs to be reported from InternalIterator when trying to access value. But there are ~700 call sites of `InternalIterator::value()`, most of which either can't hit the error condition (because the iterator is reading from memtable or from index or something) or wouldn't benefit from the deferred loading of the value (e.g. compaction iterator that reads all values anyway). Adding error handling to all these call sites would needlessly bloat the code. So instead I made the deferred value loading optional: only the call sites that may use deferred loading have to call the new method `PrepareValue()` before calling `value()`. The feature is enabled with a new bool argument `allow_unprepared_value` to a bunch of methods that create iterators (it wouldn't make sense to put it in ReadOptions because it's completely internal to iterators, with virtually no user-visible effect). Lmk if you have better ideas. Note that the deferred value loading only happens for *internal* iterators. The user-visible iterator (DBIter) always prepares the value before returning from Seek/Next/etc. We could go further and add an API to defer that value loading too, but that's most likely not useful for LogDevice, so it doesn't seem worth the complexity for now. Pull Request resolved: https://github.com/facebook/rocksdb/pull/6621 Test Plan: make -j5 check . Will also deploy to some logdevice test clusters and look at stats. Reviewed By: siying Differential Revision: D20786930 Pulled By: al13n321 fbshipit-source-id: 6da77d918bad3780522e918f17f4d5513d3e99ee
2020-04-16 00:37:23 +00:00
db_iter->GetRangeDelAggregator(), read_seq,
/* allow_unprepared_value */ true);
db_iter->SetIterUnderDBIter(internal_iter);
iterators->push_back(db_iter);
}
return Status::OK();
}
namespace {
// Return OK if dbname exists in the file system
// or create_if_missing is false
Status OpenForReadOnlyCheckExistence(const DBOptions& db_options,
const std::string& dbname) {
Status s;
if (!db_options.create_if_missing) {
// Attempt to read "CURRENT" file
const std::shared_ptr<FileSystem>& fs = db_options.env->GetFileSystem();
std::string manifest_path;
uint64_t manifest_file_number;
s = VersionSet::GetCurrentManifestPath(dbname, fs.get(), &manifest_path,
&manifest_file_number);
if (!s.ok()) {
return Status::NotFound(CurrentFileName(dbname), "does not exist");
}
}
return s;
}
} // namespace
Status DB::OpenForReadOnly(const Options& options, const std::string& dbname,
DB** dbptr, bool /*error_if_log_file_exist*/) {
// If dbname does not exist in the file system, should not do anything
Status s = OpenForReadOnlyCheckExistence(options, dbname);
if (!s.ok()) {
return s;
}
*dbptr = nullptr;
// Try to first open DB as fully compacted DB
s = CompactedDBImpl::Open(options, dbname, dbptr);
if (s.ok()) {
return s;
}
DBOptions db_options(options);
ColumnFamilyOptions cf_options(options);
std::vector<ColumnFamilyDescriptor> column_families;
column_families.push_back(
ColumnFamilyDescriptor(kDefaultColumnFamilyName, cf_options));
std::vector<ColumnFamilyHandle*> handles;
s = DBImplReadOnly::OpenForReadOnlyWithoutCheck(
db_options, dbname, column_families, &handles, dbptr);
if (s.ok()) {
assert(handles.size() == 1);
// i can delete the handle since DBImpl is always holding a
// reference to default column family
delete handles[0];
}
return s;
}
Status DB::OpenForReadOnly(
const DBOptions& db_options, const std::string& dbname,
const std::vector<ColumnFamilyDescriptor>& column_families,
std::vector<ColumnFamilyHandle*>* handles, DB** dbptr,
bool error_if_log_file_exist) {
// If dbname does not exist in the file system, should not do anything
Status s = OpenForReadOnlyCheckExistence(db_options, dbname);
if (!s.ok()) {
return s;
}
return DBImplReadOnly::OpenForReadOnlyWithoutCheck(
db_options, dbname, column_families, handles, dbptr,
error_if_log_file_exist);
}
Status DBImplReadOnly::OpenForReadOnlyWithoutCheck(
const DBOptions& db_options, const std::string& dbname,
const std::vector<ColumnFamilyDescriptor>& column_families,
std::vector<ColumnFamilyHandle*>* handles, DB** dbptr,
bool error_if_log_file_exist) {
*dbptr = nullptr;
handles->clear();
SuperVersionContext sv_context(/* create_superversion */ true);
DBImplReadOnly* impl = new DBImplReadOnly(db_options, dbname);
impl->mutex_.Lock();
Status s = impl->Recover(column_families, true /* read only */,
error_if_log_file_exist);
if (s.ok()) {
// set column family handles
for (auto cf : column_families) {
auto cfd =
impl->versions_->GetColumnFamilySet()->GetColumnFamily(cf.name);
if (cfd == nullptr) {
s = Status::InvalidArgument("Column family not found: ", cf.name);
break;
}
handles->push_back(new ColumnFamilyHandleImpl(cfd, impl, &impl->mutex_));
}
}
if (s.ok()) {
for (auto cfd : *impl->versions_->GetColumnFamilySet()) {
sv_context.NewSuperVersion();
cfd->InstallSuperVersion(&sv_context, &impl->mutex_);
}
}
impl->mutex_.Unlock();
sv_context.Clean();
if (s.ok()) {
*dbptr = impl;
for (auto* h : *handles) {
impl->NewThreadStatusCfInfo(
static_cast_with_check<ColumnFamilyHandleImpl>(h)->cfd());
}
} else {
for (auto h : *handles) {
delete h;
}
handles->clear();
delete impl;
}
return s;
}
#else // !ROCKSDB_LITE
Status DB::OpenForReadOnly(const Options& /*options*/,
const std::string& /*dbname*/, DB** /*dbptr*/,
bool /*error_if_log_file_exist*/) {
return Status::NotSupported("Not supported in ROCKSDB_LITE.");
}
Status DB::OpenForReadOnly(
const DBOptions& /*db_options*/, const std::string& /*dbname*/,
const std::vector<ColumnFamilyDescriptor>& /*column_families*/,
std::vector<ColumnFamilyHandle*>* /*handles*/, DB** /*dbptr*/,
bool /*error_if_log_file_exist*/) {
return Status::NotSupported("Not supported in ROCKSDB_LITE.");
}
#endif // !ROCKSDB_LITE
} // namespace ROCKSDB_NAMESPACE