rocksdb/table/block_based/block_based_table_reader_impl.h
Peter Dillinger 0050a73a4f New stable, fixed-length cache keys (#9126)
Summary:
This change standardizes on a new 16-byte cache key format for
block cache (incl compressed and secondary) and persistent cache (but
not table cache and row cache).

The goal is a really fast cache key with practically ideal stability and
uniqueness properties without external dependencies (e.g. from FileSystem).
A fixed key size of 16 bytes should enable future optimizations to the
concurrent hash table for block cache, which is a heavy CPU user /
bottleneck, but there appears to be measurable performance improvement
even with no changes to LRUCache.

This change replaces a lot of disjointed and ugly code handling cache
keys with calls to a simple, clean new internal API (cache_key.h).
(Preserving the old cache key logic under an option would be very ugly
and likely negate the performance gain of the new approach. Complete
replacement carries some inherent risk, but I think that's acceptable
with sufficient analysis and testing.)

The scheme for encoding new cache keys is complicated but explained
in cache_key.cc.

Also: EndianSwapValue is moved to math.h to be next to other bit
operations. (Explains some new include "math.h".) ReverseBits operation
added and unit tests added to hash_test for both.

Fixes https://github.com/facebook/rocksdb/issues/7405 (presuming a root cause)

Pull Request resolved: https://github.com/facebook/rocksdb/pull/9126

Test Plan:
### Basic correctness
Several tests needed updates to work with the new functionality, mostly
because we are no longer relying on filesystem for stable cache keys
so table builders & readers need more context info to agree on cache
keys. This functionality is so core, a huge number of existing tests
exercise the cache key functionality.

### Performance
Create db with
`TEST_TMPDIR=/dev/shm ./db_bench -bloom_bits=10 -benchmarks=fillrandom -num=3000000 -partition_index_and_filters`
And test performance with
`TEST_TMPDIR=/dev/shm ./db_bench -readonly -use_existing_db -bloom_bits=10 -benchmarks=readrandom -num=3000000 -duration=30 -cache_index_and_filter_blocks -cache_size=250000 -threads=4`
using DEBUG_LEVEL=0 and simultaneous before & after runs.
Before ops/sec, avg over 100 runs: 121924
After ops/sec, avg over 100 runs: 125385 (+2.8%)

### Collision probability
I have built a tool, ./cache_bench -stress_cache_key to broadly simulate host-wide cache activity
over many months, by making some pessimistic simplifying assumptions:
* Every generated file has a cache entry for every byte offset in the file (contiguous range of cache keys)
* All of every file is cached for its entire lifetime

We use a simple table with skewed address assignment and replacement on address collision
to simulate files coming & going, with quite a variance (super-Poisson) in ages. Some output
with `./cache_bench -stress_cache_key -sck_keep_bits=40`:

```
Total cache or DBs size: 32TiB  Writing 925.926 MiB/s or 76.2939TiB/day
Multiply by 9.22337e+18 to correct for simulation losses (but still assume whole file cached)
```

These come from default settings of 2.5M files per day of 32 MB each, and
`-sck_keep_bits=40` means that to represent a single file, we are only keeping 40 bits of
the 128-bit cache key.  With file size of 2\*\*25 contiguous keys (pessimistic), our simulation
is about 2\*\*(128-40-25) or about 9 billion billion times more prone to collision than reality.

More default assumptions, relatively pessimistic:
* 100 DBs in same process (doesn't matter much)
* Re-open DB in same process (new session ID related to old session ID) on average
every 100 files generated
* Restart process (all new session IDs unrelated to old) 24 times per day

After enough data, we get a result at the end:

```
(keep 40 bits)  17 collisions after 2 x 90 days, est 10.5882 days between (9.76592e+19 corrected)
```

If we believe the (pessimistic) simulation and the mathematical generalization, we would need to run a billion machines all for 97 billion days to expect a cache key collision. To help verify that our generalization ("corrected") is robust, we can make our simulation more precise with `-sck_keep_bits=41` and `42`, which takes more running time to get enough data:

```
(keep 41 bits)  16 collisions after 4 x 90 days, est 22.5 days between (1.03763e+20 corrected)
(keep 42 bits)  19 collisions after 10 x 90 days, est 47.3684 days between (1.09224e+20 corrected)
```

The generalized prediction still holds. With the `-sck_randomize` option, we can see that we are beating "random" cache keys (except offsets still non-randomized) by a modest amount (roughly 20x less collision prone than random), which should make us reasonably comfortable even in "degenerate" cases:

```
197 collisions after 1 x 90 days, est 0.456853 days between (4.21372e+18 corrected)
```

I've run other tests to validate other conditions behave as expected, never behaving "worse than random" unless we start chopping off structured data.

Reviewed By: zhichao-cao

Differential Revision: D33171746

Pulled By: pdillinger

fbshipit-source-id: f16a57e369ed37be5e7e33525ace848d0537c88f
2021-12-16 17:15:13 -08:00

164 lines
6.4 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).
//
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
#pragma once
#include "table/block_based/block_based_table_reader.h"
#include "table/block_based/reader_common.h"
// The file contains some member functions of BlockBasedTable that
// cannot be implemented in block_based_table_reader.cc because
// it's called by other files (e.g. block_based_iterator.h) and
// are templates.
namespace ROCKSDB_NAMESPACE {
// Convert an index iterator value (i.e., an encoded BlockHandle)
// into an iterator over the contents of the corresponding block.
// If input_iter is null, new a iterator
// If input_iter is not null, update this iter and return it
template <typename TBlockIter>
TBlockIter* BlockBasedTable::NewDataBlockIterator(
const ReadOptions& ro, const BlockHandle& handle, TBlockIter* input_iter,
BlockType block_type, GetContext* get_context,
BlockCacheLookupContext* lookup_context, Status s,
FilePrefetchBuffer* prefetch_buffer, bool for_compaction) const {
PERF_TIMER_GUARD(new_table_block_iter_nanos);
TBlockIter* iter = input_iter != nullptr ? input_iter : new TBlockIter;
if (!s.ok()) {
iter->Invalidate(s);
return iter;
}
CachableEntry<UncompressionDict> uncompression_dict;
if (rep_->uncompression_dict_reader) {
const bool no_io = (ro.read_tier == kBlockCacheTier);
s = rep_->uncompression_dict_reader->GetOrReadUncompressionDictionary(
prefetch_buffer, no_io, get_context, lookup_context,
&uncompression_dict);
if (!s.ok()) {
iter->Invalidate(s);
return iter;
}
}
const UncompressionDict& dict = uncompression_dict.GetValue()
? *uncompression_dict.GetValue()
: UncompressionDict::GetEmptyDict();
CachableEntry<Block> block;
s = RetrieveBlock(prefetch_buffer, ro, handle, dict, &block, block_type,
get_context, lookup_context, for_compaction,
/* use_cache */ true, /* wait_for_cache */ true);
if (!s.ok()) {
assert(block.IsEmpty());
iter->Invalidate(s);
return iter;
}
assert(block.GetValue() != nullptr);
// Block contents are pinned and it is still pinned after the iterator
// is destroyed as long as cleanup functions are moved to another object,
// when:
// 1. block cache handle is set to be released in cleanup function, or
// 2. it's pointing to immortal source. If own_bytes is true then we are
// not reading data from the original source, whether immortal or not.
// Otherwise, the block is pinned iff the source is immortal.
const bool block_contents_pinned =
block.IsCached() ||
(!block.GetValue()->own_bytes() && rep_->immortal_table);
iter = InitBlockIterator<TBlockIter>(rep_, block.GetValue(), block_type, iter,
block_contents_pinned);
if (!block.IsCached()) {
if (!ro.fill_cache) {
Cache* const block_cache = rep_->table_options.block_cache.get();
if (block_cache) {
// insert a dummy record to block cache to track the memory usage
Cache::Handle* cache_handle = nullptr;
CacheKey key = CacheKey::CreateUniqueForCacheLifetime(block_cache);
s = block_cache->Insert(key.AsSlice(), nullptr,
block.GetValue()->ApproximateMemoryUsage(),
nullptr, &cache_handle);
if (s.ok()) {
assert(cache_handle != nullptr);
iter->RegisterCleanup(&ForceReleaseCachedEntry, block_cache,
cache_handle);
}
}
}
} else {
iter->SetCacheHandle(block.GetCacheHandle());
}
block.TransferTo(iter);
return iter;
}
// Convert an uncompressed data block (i.e CachableEntry<Block>)
// into an iterator over the contents of the corresponding block.
// If input_iter is null, new a iterator
// If input_iter is not null, update this iter and return it
template <typename TBlockIter>
TBlockIter* BlockBasedTable::NewDataBlockIterator(const ReadOptions& ro,
CachableEntry<Block>& block,
TBlockIter* input_iter,
Status s) const {
PERF_TIMER_GUARD(new_table_block_iter_nanos);
TBlockIter* iter = input_iter != nullptr ? input_iter : new TBlockIter;
if (!s.ok()) {
iter->Invalidate(s);
return iter;
}
assert(block.GetValue() != nullptr);
// Block contents are pinned and it is still pinned after the iterator
// is destroyed as long as cleanup functions are moved to another object,
// when:
// 1. block cache handle is set to be released in cleanup function, or
// 2. it's pointing to immortal source. If own_bytes is true then we are
// not reading data from the original source, whether immortal or not.
// Otherwise, the block is pinned iff the source is immortal.
const bool block_contents_pinned =
block.IsCached() ||
(!block.GetValue()->own_bytes() && rep_->immortal_table);
iter = InitBlockIterator<TBlockIter>(rep_, block.GetValue(), BlockType::kData,
iter, block_contents_pinned);
if (!block.IsCached()) {
if (!ro.fill_cache) {
Cache* const block_cache = rep_->table_options.block_cache.get();
if (block_cache) {
// insert a dummy record to block cache to track the memory usage
Cache::Handle* cache_handle = nullptr;
CacheKey key = CacheKey::CreateUniqueForCacheLifetime(block_cache);
s = block_cache->Insert(key.AsSlice(), nullptr,
block.GetValue()->ApproximateMemoryUsage(),
nullptr, &cache_handle);
if (s.ok()) {
assert(cache_handle != nullptr);
iter->RegisterCleanup(&ForceReleaseCachedEntry, block_cache,
cache_handle);
}
}
}
} else {
iter->SetCacheHandle(block.GetCacheHandle());
}
block.TransferTo(iter);
return iter;
}
} // namespace ROCKSDB_NAMESPACE