mirror of
https://github.com/facebook/rocksdb.git
synced 2024-11-30 04:41:49 +00:00
30bc495c03
Summary: Delete range logic is moved from `DBIter` to `MergingIterator`, and `MergingIterator` will seek to the end of a range deletion if possible instead of scanning through each key and check with `RangeDelAggregator`. With the invariant that a key in level L (consider memtable as the first level, each immutable and L0 as a separate level) has a larger sequence number than all keys in any level >L, a range tombstone `[start, end)` from level L covers all keys in its range in any level >L. This property motivates optimizations in iterator: - in `Seek(target)`, if level L has a range tombstone `[start, end)` that covers `target.UserKey`, then for all levels > L, we can do Seek() on `end` instead of `target` to skip some range tombstone covered keys. - in `Next()/Prev()`, if the current key is covered by a range tombstone `[start, end)` from level L, we can do `Seek` to `end` for all levels > L. This PR implements the above optimizations in `MergingIterator`. As all range tombstone covered keys are now skipped in `MergingIterator`, the range tombstone logic is removed from `DBIter`. The idea in this PR is similar to https://github.com/facebook/rocksdb/issues/7317, but this PR leaves `InternalIterator` interface mostly unchanged. **Credit**: the cascading seek optimization and the sentinel key (discussed below) are inspired by [Pebble](https://github.com/cockroachdb/pebble/blob/master/merging_iter.go) and suggested by ajkr in https://github.com/facebook/rocksdb/issues/7317. The two optimizations are mostly implemented in `SeekImpl()/SeekForPrevImpl()` and `IsNextDeleted()/IsPrevDeleted()` in `merging_iterator.cc`. See comments for each method for more detail. One notable change is that the minHeap/maxHeap used by `MergingIterator` now contains range tombstone end keys besides point key iterators. This helps to reduce the number of key comparisons. For example, for a range tombstone `[start, end)`, a `start` and an `end` `HeapItem` are inserted into the heap. When a `HeapItem` for range tombstone start key is popped from the minHeap, we know this range tombstone becomes "active" in the sense that, before the range tombstone's end key is popped from the minHeap, all the keys popped from this heap is covered by the range tombstone's internal key range `[start, end)`. Another major change, *delete range sentinel key*, is made to `LevelIterator`. Before this PR, when all point keys in an SST file are iterated through in `MergingIterator`, a level iterator would advance to the next SST file in its level. In the case when an SST file has a range tombstone that covers keys beyond the SST file's last point key, advancing to the next SST file would lose this range tombstone. Consequently, `MergingIterator` could return keys that should have been deleted by some range tombstone. We prevent this by pretending that file boundaries in each SST file are sentinel keys. A `LevelIterator` now only advance the file iterator once the sentinel key is processed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10449 Test Plan: - Added many unit tests in db_range_del_test - Stress test: `./db_stress --readpercent=5 --prefixpercent=19 --writepercent=20 -delpercent=10 --iterpercent=44 --delrangepercent=2` - Additional iterator stress test is added to verify against iterators against expected state: https://github.com/facebook/rocksdb/issues/10538. This is based on ajkr's previous attempt https://github.com/facebook/rocksdb/pull/5506#issuecomment-506021913. ``` python3 ./tools/db_crashtest.py blackbox --simple --write_buffer_size=524288 --target_file_size_base=524288 --max_bytes_for_level_base=2097152 --compression_type=none --max_background_compactions=8 --value_size_mult=33 --max_key=5000000 --interval=10 --duration=7200 --delrangepercent=3 --delpercent=9 --iterpercent=25 --writepercent=60 --readpercent=3 --prefixpercent=0 --num_iterations=1000 --range_deletion_width=100 --verify_iterator_with_expected_state_one_in=1 ``` - Performance benchmark: I used a similar setup as in the blog [post](http://rocksdb.org/blog/2018/11/21/delete-range.html) that introduced DeleteRange, "a database with 5 million data keys, and 10000 range tombstones (ignoring those dropped during compaction) that were written in regular intervals after 4.5 million data keys were written". As expected, the performance with this PR depends on the range tombstone width. ``` # Setup: TEST_TMPDIR=/dev/shm ./db_bench_main --benchmarks=fillrandom --writes=4500000 --num=5000000 TEST_TMPDIR=/dev/shm ./db_bench_main --benchmarks=overwrite --writes=500000 --num=5000000 --use_existing_db=true --writes_per_range_tombstone=50 # Scan entire DB TEST_TMPDIR=/dev/shm ./db_bench_main --benchmarks=readseq[-X5] --use_existing_db=true --num=5000000 --disable_auto_compactions=true # Short range scan (10 Next()) TEST_TMPDIR=/dev/shm/width-100/ ./db_bench_main --benchmarks=seekrandom[-X5] --use_existing_db=true --num=500000 --reads=100000 --seek_nexts=10 --disable_auto_compactions=true # Long range scan(1000 Next()) TEST_TMPDIR=/dev/shm/width-100/ ./db_bench_main --benchmarks=seekrandom[-X5] --use_existing_db=true --num=500000 --reads=2500 --seek_nexts=1000 --disable_auto_compactions=true ``` Avg over of 10 runs (some slower tests had fews runs): For the first column (tombstone), 0 means no range tombstone, 100-10000 means width of the 10k range tombstones, and 1 means there is a single range tombstone in the entire DB (width is 1000). The 1 tombstone case is to test regression when there's very few range tombstones in the DB, as no range tombstone is likely to take a different code path than with range tombstones. - Scan entire DB | tombstone width | Pre-PR ops/sec | Post-PR ops/sec | ±% | | ------------- | ------------- | ------------- | ------------- | | 0 range tombstone |2525600 (± 43564) |2486917 (± 33698) |-1.53% | | 100 |1853835 (± 24736) |2073884 (± 32176) |+11.87% | | 1000 |422415 (± 7466) |1115801 (± 22781) |+164.15% | | 10000 |22384 (± 227) |227919 (± 6647) |+918.22% | | 1 range tombstone |2176540 (± 39050) |2434954 (± 24563) |+11.87% | - Short range scan | tombstone width | Pre-PR ops/sec | Post-PR ops/sec | ±% | | ------------- | ------------- | ------------- | ------------- | | 0 range tombstone |35398 (± 533) |35338 (± 569) |-0.17% | | 100 |28276 (± 664) |31684 (± 331) |+12.05% | | 1000 |7637 (± 77) |25422 (± 277) |+232.88% | | 10000 |1367 |28667 |+1997.07% | | 1 range tombstone |32618 (± 581) |32748 (± 506) |+0.4% | - Long range scan | tombstone width | Pre-PR ops/sec | Post-PR ops/sec | ±% | | ------------- | ------------- | ------------- | ------------- | | 0 range tombstone |2262 (± 33) |2353 (± 20) |+4.02% | | 100 |1696 (± 26) |1926 (± 18) |+13.56% | | 1000 |410 (± 6) |1255 (± 29) |+206.1% | | 10000 |25 |414 |+1556.0% | | 1 range tombstone |1957 (± 30) |2185 (± 44) |+11.65% | - Microbench does not show significant regression: https://gist.github.com/cbi42/59f280f85a59b678e7e5d8561e693b61 Reviewed By: ajkr Differential Revision: D38450331 Pulled By: cbi42 fbshipit-source-id: b5ef12e8d8c289ed2e163ccdf277f5039b511fca
342 lines
12 KiB
C++
342 lines
12 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).
|
|
|
|
#include "db/db_impl/db_impl_readonly.h"
|
|
|
|
#include "db/arena_wrapped_db_iter.h"
|
|
#include "db/db_impl/compacted_db_impl.h"
|
|
#include "db/db_impl/db_impl.h"
|
|
#include "db/db_iter.h"
|
|
#include "db/merge_context.h"
|
|
#include "logging/logging.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, /*seq_per_batch*/ false,
|
|
/*batch_per_txn*/ true, /*read_only*/ true) {
|
|
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) {
|
|
return Get(read_options, column_family, key, pinnable_val,
|
|
/*timestamp*/ nullptr);
|
|
}
|
|
|
|
Status DBImplReadOnly::Get(const ReadOptions& read_options,
|
|
ColumnFamilyHandle* column_family, const Slice& key,
|
|
PinnableSlice* pinnable_val,
|
|
std::string* timestamp) {
|
|
assert(pinnable_val != nullptr);
|
|
// TODO: stopwatch DB_GET needed?, perf timer needed?
|
|
PERF_TIMER_GUARD(get_snapshot_time);
|
|
|
|
assert(column_family);
|
|
if (read_options.timestamp) {
|
|
const Status s = FailIfTsMismatchCf(
|
|
column_family, *(read_options.timestamp), /*ts_for_read=*/true);
|
|
if (!s.ok()) {
|
|
return s;
|
|
}
|
|
} else {
|
|
const Status s = FailIfCfHasTs(column_family);
|
|
if (!s.ok()) {
|
|
return s;
|
|
}
|
|
}
|
|
|
|
// Clear the timestamps for returning results so that we can distinguish
|
|
// between tombstone or key that has never been written
|
|
if (timestamp) {
|
|
timestamp->clear();
|
|
}
|
|
|
|
const Comparator* ucmp = column_family->GetComparator();
|
|
assert(ucmp);
|
|
std::string* ts = ucmp->timestamp_size() > 0 ? timestamp : nullptr;
|
|
|
|
Status s;
|
|
SequenceNumber snapshot = versions_->LastSequence();
|
|
GetWithTimestampReadCallback read_cb(snapshot);
|
|
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;
|
|
SequenceNumber max_covering_tombstone_seq = 0;
|
|
LookupKey lkey(key, snapshot, read_options.timestamp);
|
|
PERF_TIMER_STOP(get_snapshot_time);
|
|
if (super_version->mem->Get(lkey, pinnable_val->GetSelf(),
|
|
/*columns=*/nullptr, ts, &s, &merge_context,
|
|
&max_covering_tombstone_seq, read_options,
|
|
false /* immutable_memtable */, &read_cb)) {
|
|
pinnable_val->PinSelf();
|
|
RecordTick(stats_, MEMTABLE_HIT);
|
|
} else {
|
|
PERF_TIMER_GUARD(get_from_output_files_time);
|
|
PinnedIteratorsManager pinned_iters_mgr;
|
|
super_version->current->Get(
|
|
read_options, lkey, pinnable_val, /*columns=*/nullptr, ts, &s,
|
|
&merge_context, &max_covering_tombstone_seq, &pinned_iters_mgr,
|
|
/*value_found*/ nullptr,
|
|
/*key_exists*/ nullptr, /*seq*/ nullptr, &read_cb,
|
|
/*is_blob*/ nullptr,
|
|
/*do_merge*/ true);
|
|
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) {
|
|
assert(column_family);
|
|
if (read_options.timestamp) {
|
|
const Status s = FailIfTsMismatchCf(
|
|
column_family, *(read_options.timestamp), /*ts_for_read=*/true);
|
|
if (!s.ok()) {
|
|
return NewErrorIterator(s);
|
|
}
|
|
} else {
|
|
const Status s = FailIfCfHasTs(column_family);
|
|
if (!s.ok()) {
|
|
return NewErrorIterator(s);
|
|
}
|
|
}
|
|
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,
|
|
super_version->current, read_seq,
|
|
super_version->mutable_cf_options.max_sequential_skip_in_iterations,
|
|
super_version->version_number, read_callback);
|
|
auto internal_iter = NewInternalIterator(
|
|
db_iter->GetReadOptions(), cfd, super_version, db_iter->GetArena(),
|
|
read_seq, /* allow_unprepared_value */ true, db_iter);
|
|
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) {
|
|
if (read_options.timestamp) {
|
|
for (auto* cf : column_families) {
|
|
assert(cf);
|
|
const Status s = FailIfTsMismatchCf(cf, *(read_options.timestamp),
|
|
/*ts_for_read=*/true);
|
|
if (!s.ok()) {
|
|
return s;
|
|
}
|
|
}
|
|
} else {
|
|
for (auto* cf : column_families) {
|
|
assert(cf);
|
|
const Status s = FailIfCfHasTs(cf);
|
|
if (!s.ok()) {
|
|
return s;
|
|
}
|
|
}
|
|
}
|
|
|
|
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,
|
|
sv->current, read_seq,
|
|
sv->mutable_cf_options.max_sequential_skip_in_iterations,
|
|
sv->version_number, read_callback);
|
|
auto* internal_iter = NewInternalIterator(
|
|
db_iter->GetReadOptions(), cfd, sv, db_iter->GetArena(), read_seq,
|
|
/* allow_unprepared_value */ true, db_iter);
|
|
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 it if
|
|
// create_if_missing
|
|
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);
|
|
} else {
|
|
// Historic behavior that doesn't necessarily make sense
|
|
s = db_options.env->CreateDirIfMissing(dbname);
|
|
}
|
|
return s;
|
|
}
|
|
} // namespace
|
|
|
|
Status DB::OpenForReadOnly(const Options& options, const std::string& dbname,
|
|
DB** dbptr, bool /*error_if_wal_file_exists*/) {
|
|
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_wal_file_exists) {
|
|
// 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_wal_file_exists);
|
|
}
|
|
|
|
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_wal_file_exists) {
|
|
*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_wal_file_exists);
|
|
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_wal_file_exists*/) {
|
|
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_wal_file_exists*/) {
|
|
return Status::NotSupported("Not supported in ROCKSDB_LITE.");
|
|
}
|
|
#endif // !ROCKSDB_LITE
|
|
|
|
} // namespace ROCKSDB_NAMESPACE
|