rocksdb/cache/cache_reservation_manager.cc
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

189 lines
7 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.
#include "cache/cache_reservation_manager.h"
#include <cassert>
#include <cstddef>
#include <cstring>
#include <memory>
#include "cache/cache_entry_roles.h"
#include "rocksdb/cache.h"
#include "rocksdb/slice.h"
#include "rocksdb/status.h"
#include "table/block_based/block_based_table_reader.h"
#include "util/coding.h"
namespace ROCKSDB_NAMESPACE {
CacheReservationManager::CacheReservationManager(std::shared_ptr<Cache> cache,
bool delayed_decrease)
: delayed_decrease_(delayed_decrease),
cache_allocated_size_(0),
memory_used_(0) {
assert(cache != nullptr);
cache_ = cache;
}
CacheReservationManager::~CacheReservationManager() {
for (auto* handle : dummy_handles_) {
cache_->Release(handle, true);
}
}
template <CacheEntryRole R>
Status CacheReservationManager::UpdateCacheReservation(
std::size_t new_mem_used) {
memory_used_ = new_mem_used;
std::size_t cur_cache_allocated_size =
cache_allocated_size_.load(std::memory_order_relaxed);
if (new_mem_used == cur_cache_allocated_size) {
return Status::OK();
} else if (new_mem_used > cur_cache_allocated_size) {
Status s = IncreaseCacheReservation<R>(new_mem_used);
return s;
} else {
// In delayed decrease mode, we don't decrease cache reservation
// untill the memory usage is less than 3/4 of what we reserve
// in the cache.
// We do this because
// (1) Dummy entry insertion is expensive in block cache
// (2) Delayed releasing previously inserted dummy entries can save such
// expensive dummy entry insertion on memory increase in the near future,
// which is likely to happen when the memory usage is greater than or equal
// to 3/4 of what we reserve
if (delayed_decrease_ && new_mem_used >= cur_cache_allocated_size / 4 * 3) {
return Status::OK();
} else {
Status s = DecreaseCacheReservation(new_mem_used);
return s;
}
}
}
// Explicitly instantiate templates for "CacheEntryRole" values we use.
// This makes it possible to keep the template definitions in the .cc file.
template Status CacheReservationManager::UpdateCacheReservation<
CacheEntryRole::kWriteBuffer>(std::size_t new_mem_used);
template Status CacheReservationManager::UpdateCacheReservation<
CacheEntryRole::kCompressionDictionaryBuildingBuffer>(
std::size_t new_mem_used);
// For cache reservation manager unit tests
template Status CacheReservationManager::UpdateCacheReservation<
CacheEntryRole::kMisc>(std::size_t new_mem_used);
template <CacheEntryRole R>
Status CacheReservationManager::MakeCacheReservation(
std::size_t incremental_memory_used,
std::unique_ptr<CacheReservationHandle<R>>* handle) {
assert(handle != nullptr);
Status s =
UpdateCacheReservation<R>(GetTotalMemoryUsed() + incremental_memory_used);
(*handle).reset(new CacheReservationHandle<R>(incremental_memory_used,
shared_from_this()));
return s;
}
template Status
CacheReservationManager::MakeCacheReservation<CacheEntryRole::kMisc>(
std::size_t incremental_memory_used,
std::unique_ptr<CacheReservationHandle<CacheEntryRole::kMisc>>* handle);
template Status CacheReservationManager::MakeCacheReservation<
CacheEntryRole::kFilterConstruction>(
std::size_t incremental_memory_used,
std::unique_ptr<
CacheReservationHandle<CacheEntryRole::kFilterConstruction>>* handle);
template <CacheEntryRole R>
Status CacheReservationManager::IncreaseCacheReservation(
std::size_t new_mem_used) {
Status return_status = Status::OK();
while (new_mem_used > cache_allocated_size_.load(std::memory_order_relaxed)) {
Cache::Handle* handle = nullptr;
return_status = cache_->Insert(GetNextCacheKey(), nullptr, kSizeDummyEntry,
GetNoopDeleterForRole<R>(), &handle);
if (return_status != Status::OK()) {
return return_status;
}
dummy_handles_.push_back(handle);
cache_allocated_size_ += kSizeDummyEntry;
}
return return_status;
}
Status CacheReservationManager::DecreaseCacheReservation(
std::size_t new_mem_used) {
Status return_status = Status::OK();
// Decrease to the smallest multiple of kSizeDummyEntry that is greater than
// or equal to new_mem_used We do addition instead of new_mem_used <=
// cache_allocated_size_.load(std::memory_order_relaxed) - kSizeDummyEntry to
// avoid underflow of size_t when cache_allocated_size_ = 0
while (new_mem_used + kSizeDummyEntry <=
cache_allocated_size_.load(std::memory_order_relaxed)) {
assert(!dummy_handles_.empty());
auto* handle = dummy_handles_.back();
cache_->Release(handle, true);
dummy_handles_.pop_back();
cache_allocated_size_ -= kSizeDummyEntry;
}
return return_status;
}
std::size_t CacheReservationManager::GetTotalReservedCacheSize() {
return cache_allocated_size_.load(std::memory_order_relaxed);
}
std::size_t CacheReservationManager::GetTotalMemoryUsed() {
return memory_used_;
}
Slice CacheReservationManager::GetNextCacheKey() {
// Calling this function will have the side-effect of changing the
// underlying cache_key_ that is shared among other keys generated from this
// fucntion. Therefore please make sure the previous keys are saved/copied
// before calling this function.
cache_key_ = CacheKey::CreateUniqueForCacheLifetime(cache_.get());
return cache_key_.AsSlice();
}
template <CacheEntryRole R>
Cache::DeleterFn CacheReservationManager::TEST_GetNoopDeleterForRole() {
return GetNoopDeleterForRole<R>();
}
template Cache::DeleterFn CacheReservationManager::TEST_GetNoopDeleterForRole<
CacheEntryRole::kFilterConstruction>();
template <CacheEntryRole R>
CacheReservationHandle<R>::CacheReservationHandle(
std::size_t incremental_memory_used,
std::shared_ptr<CacheReservationManager> cache_res_mgr)
: incremental_memory_used_(incremental_memory_used) {
assert(cache_res_mgr != nullptr);
cache_res_mgr_ = cache_res_mgr;
}
template <CacheEntryRole R>
CacheReservationHandle<R>::~CacheReservationHandle() {
assert(cache_res_mgr_ != nullptr);
assert(cache_res_mgr_->GetTotalMemoryUsed() >= incremental_memory_used_);
Status s = cache_res_mgr_->UpdateCacheReservation<R>(
cache_res_mgr_->GetTotalMemoryUsed() - incremental_memory_used_);
s.PermitUncheckedError();
}
// Explicitly instantiate templates for "CacheEntryRole" values we use.
// This makes it possible to keep the template definitions in the .cc file.
template class CacheReservationHandle<CacheEntryRole::kMisc>;
template class CacheReservationHandle<CacheEntryRole::kFilterConstruction>;
} // namespace ROCKSDB_NAMESPACE