rocksdb/cache/cache_reservation_manager.h
Peter Dillinger 9f7801c5f1 Major Cache refactoring, CPU efficiency improvement (#10975)
Summary:
This is several refactorings bundled into one to avoid having to incrementally re-modify uses of Cache several times. Overall, there are breaking changes to Cache class, and it becomes more of low-level interface for implementing caches, especially block cache. New internal APIs make using Cache cleaner than before, and more insulated from block cache evolution. Hopefully, this is the last really big block cache refactoring, because of rather effectively decoupling the implementations from the uses. This change also removes the EXPERIMENTAL designation on the SecondaryCache support in Cache. It seems reasonably mature at this point but still subject to change/evolution (as I warn in the API docs for Cache).

The high-level motivation for this refactoring is to minimize code duplication / compounding complexity in adding SecondaryCache support to HyperClockCache (in a later PR). Other benefits listed below.

* static_cast lines of code +29 -35 (net removed 6)
* reinterpret_cast lines of code +6 -32 (net removed 26)

## cache.h and secondary_cache.h
* Always use CacheItemHelper with entries instead of just a Deleter. There are several motivations / justifications:
  * Simpler for implementations to deal with just one Insert and one Lookup.
  * Simpler and more efficient implementation because we don't have to track which entries are using helpers and which are using deleters
  * Gets rid of hack to classify cache entries by their deleter. Instead, the CacheItemHelper includes a CacheEntryRole. This simplifies a lot of code (cache_entry_roles.h almost eliminated). Fixes https://github.com/facebook/rocksdb/issues/9428.
  * Makes it trivial to adjust SecondaryCache behavior based on kind of block (e.g. don't re-compress filter blocks).
  * It is arguably less convenient for many direct users of Cache, but direct users of Cache are now rare with introduction of typed_cache.h (below).
  * I considered and rejected an alternative approach in which we reduce customizability by assuming each secondary cache compatible value starts with a Slice referencing the uncompressed block contents (already true or mostly true), but we apparently intend to stack secondary caches. Saving an entry from a compressed secondary to a lower tier requires custom handling offered by SaveToCallback, etc.
* Make CreateCallback part of the helper and introduce CreateContext to work with it (alternative to https://github.com/facebook/rocksdb/issues/10562). This cleans up the interface while still allowing context to be provided for loading/parsing values into primary cache. This model works for async lookup in BlockBasedTable reader (reader owns a CreateContext) under the assumption that it always waits on secondary cache operations to finish. (Otherwise, the CreateContext could be destroyed while async operation depending on it continues.) This likely contributes most to the observed performance improvement because it saves an std::function backed by a heap allocation.
* Use char* for serialized data, e.g. in SaveToCallback, where void* was confusingly used. (We use `char*` for serialized byte data all over RocksDB, with many advantages over `void*`. `memcpy` etc. are legacy APIs that should not be mimicked.)
* Add a type alias Cache::ObjectPtr = void*, so that we can better indicate the intent of the void* when it is to be the object associated with a Cache entry. Related: started (but did not complete) a refactoring to move away from "value" of a cache entry toward "object" or "obj". (It is confusing to call Cache a key-value store (like DB) when it is really storing arbitrary in-memory objects, not byte strings.)
* Remove unnecessary key param from DeleterFn. This is good for efficiency in HyperClockCache, which does not directly store the cache key in memory. (Alternative to https://github.com/facebook/rocksdb/issues/10774)
* Add allocator to Cache DeleterFn. This is a kind of future-proofing change in case we get more serious about using the Cache allocator for memory tracked by the Cache. Right now, only the uncompressed block contents are allocated using the allocator, and a pointer to that allocator is saved as part of the cached object so that the deleter can use it. (See CacheAllocationPtr.) If in the future we are able to "flatten out" our Cache objects some more, it would be good not to have to track the allocator as part of each object.
* Removes legacy `ApplyToAllCacheEntries` and changes `ApplyToAllEntries` signature for Deleter->CacheItemHelper change.

## typed_cache.h
Adds various "typed" interfaces to the Cache as internal APIs, so that most uses of Cache can use simple type safe code without casting and without explicit deleters, etc. Almost all of the non-test, non-glue code uses of Cache have been migrated. (Follow-up work: CompressedSecondaryCache deserves deeper attention to migrate.) This change expands RocksDB's internal usage of metaprogramming and SFINAE (https://en.cppreference.com/w/cpp/language/sfinae).

The existing usages of Cache are divided up at a high level into these new interfaces. See updated existing uses of Cache for examples of how these are used.
* PlaceholderCacheInterface - Used for making cache reservations, with entries that have a charge but no value.
* BasicTypedCacheInterface<TValue> - Used for primary cache storage of objects of type TValue, which can be cleaned up with std::default_delete<TValue>. The role is provided by TValue::kCacheEntryRole or given in an optional template parameter.
* FullTypedCacheInterface<TValue, TCreateContext> - Used for secondary cache compatible storage of objects of type TValue. In addition to BasicTypedCacheInterface constraints, we require TValue::ContentSlice() to return persistable data. This simplifies usage for the normal case of simple secondary cache compatibility (can give you a Slice to the data already in memory). In addition to TCreateContext performing the role of Cache::CreateContext, it is also expected to provide a factory function for creating TValue.
* For each of these, there's a "Shared" version (e.g. FullTypedSharedCacheInterface) that holds a shared_ptr to the Cache, rather than assuming external ownership by holding only a raw `Cache*`.

These interfaces introduce specific handle types for each interface instantiation, so that it's easy to see what kind of object is controlled by a handle. (Ultimately, this might not be worth the extra complexity, but it seems OK so far.)

Note: I attempted to make the cache 'charge' automatically inferred from the cache object type, such as by expecting an ApproximateMemoryUsage() function, but this is not so clean because there are cases where we need to compute the charge ahead of time and don't want to re-compute it.

## block_cache.h
This header is essentially the replacement for the old block_like_traits.h. It includes various things to support block cache access with typed_cache.h for block-based table.

## block_based_table_reader.cc
Before this change, accessing the block cache here was an awkward mix of static polymorphism (template TBlocklike) and switch-case on a dynamic BlockType value. This change mostly unifies on static polymorphism, relying on minor hacks in block_cache.h to distinguish variants of Block. We still check BlockType in some places (especially for stats, which could be improved in follow-up work) but at least the BlockType is a static constant from the template parameter. (No more awkward partial redundancy between static and dynamic info.) This likely contributes to the overall performance improvement, but hasn't been tested in isolation.

The other key source of simplification here is a more unified system of creating block cache objects: for directly populating from primary cache and for promotion from secondary cache. Both use BlockCreateContext, for context and for factory functions.

## block_based_table_builder.cc, cache_dump_load_impl.cc
Before this change, warming caches was super ugly code. Both of these source files had switch statements to basically transition from the dynamic BlockType world to the static TBlocklike world. None of that mess is needed anymore as there's a new, untyped WarmInCache function that handles all the details just as promotion from SecondaryCache would. (Fixes `TODO akanksha: Dedup below code` in block_based_table_builder.cc.)

## Everything else
Mostly just updating Cache users to use new typed APIs when reasonably possible, or changed Cache APIs when not.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/10975

Test Plan:
tests updated

Performance test setup similar to https://github.com/facebook/rocksdb/issues/10626 (by cache size, LRUCache when not "hyper" for HyperClockCache):

34MB 1thread base.hyper -> kops/s: 0.745 io_bytes/op: 2.52504e+06 miss_ratio: 0.140906 max_rss_mb: 76.4844
34MB 1thread new.hyper -> kops/s: 0.751 io_bytes/op: 2.5123e+06 miss_ratio: 0.140161 max_rss_mb: 79.3594
34MB 1thread base -> kops/s: 0.254 io_bytes/op: 1.36073e+07 miss_ratio: 0.918818 max_rss_mb: 45.9297
34MB 1thread new -> kops/s: 0.252 io_bytes/op: 1.36157e+07 miss_ratio: 0.918999 max_rss_mb: 44.1523
34MB 32thread base.hyper -> kops/s: 7.272 io_bytes/op: 2.88323e+06 miss_ratio: 0.162532 max_rss_mb: 516.602
34MB 32thread new.hyper -> kops/s: 7.214 io_bytes/op: 2.99046e+06 miss_ratio: 0.168818 max_rss_mb: 518.293
34MB 32thread base -> kops/s: 3.528 io_bytes/op: 1.35722e+07 miss_ratio: 0.914691 max_rss_mb: 264.926
34MB 32thread new -> kops/s: 3.604 io_bytes/op: 1.35744e+07 miss_ratio: 0.915054 max_rss_mb: 264.488
233MB 1thread base.hyper -> kops/s: 53.909 io_bytes/op: 2552.35 miss_ratio: 0.0440566 max_rss_mb: 241.984
233MB 1thread new.hyper -> kops/s: 62.792 io_bytes/op: 2549.79 miss_ratio: 0.044043 max_rss_mb: 241.922
233MB 1thread base -> kops/s: 1.197 io_bytes/op: 2.75173e+06 miss_ratio: 0.103093 max_rss_mb: 241.559
233MB 1thread new -> kops/s: 1.199 io_bytes/op: 2.73723e+06 miss_ratio: 0.10305 max_rss_mb: 240.93
233MB 32thread base.hyper -> kops/s: 1298.69 io_bytes/op: 2539.12 miss_ratio: 0.0440307 max_rss_mb: 371.418
233MB 32thread new.hyper -> kops/s: 1421.35 io_bytes/op: 2538.75 miss_ratio: 0.0440307 max_rss_mb: 347.273
233MB 32thread base -> kops/s: 9.693 io_bytes/op: 2.77304e+06 miss_ratio: 0.103745 max_rss_mb: 569.691
233MB 32thread new -> kops/s: 9.75 io_bytes/op: 2.77559e+06 miss_ratio: 0.103798 max_rss_mb: 552.82
1597MB 1thread base.hyper -> kops/s: 58.607 io_bytes/op: 1449.14 miss_ratio: 0.0249324 max_rss_mb: 1583.55
1597MB 1thread new.hyper -> kops/s: 69.6 io_bytes/op: 1434.89 miss_ratio: 0.0247167 max_rss_mb: 1584.02
1597MB 1thread base -> kops/s: 60.478 io_bytes/op: 1421.28 miss_ratio: 0.024452 max_rss_mb: 1589.45
1597MB 1thread new -> kops/s: 63.973 io_bytes/op: 1416.07 miss_ratio: 0.0243766 max_rss_mb: 1589.24
1597MB 32thread base.hyper -> kops/s: 1436.2 io_bytes/op: 1357.93 miss_ratio: 0.0235353 max_rss_mb: 1692.92
1597MB 32thread new.hyper -> kops/s: 1605.03 io_bytes/op: 1358.04 miss_ratio: 0.023538 max_rss_mb: 1702.78
1597MB 32thread base -> kops/s: 280.059 io_bytes/op: 1350.34 miss_ratio: 0.023289 max_rss_mb: 1675.36
1597MB 32thread new -> kops/s: 283.125 io_bytes/op: 1351.05 miss_ratio: 0.0232797 max_rss_mb: 1703.83

Almost uniformly improving over base revision, especially for hot paths with HyperClockCache, up to 12% higher throughput seen (1597MB, 32thread, hyper). The improvement for that is likely coming from much simplified code for providing context for secondary cache promotion (CreateCallback/CreateContext), and possibly from less branching in block_based_table_reader. And likely a small improvement from not reconstituting key for DeleterFn.

Reviewed By: anand1976

Differential Revision: D42417818

Pulled By: pdillinger

fbshipit-source-id: f86bfdd584dce27c028b151ba56818ad14f7a432
2023-01-11 14:20:40 -08:00

318 lines
13 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 <atomic>
#include <cstddef>
#include <cstdint>
#include <memory>
#include <mutex>
#include <vector>
#include "cache/cache_entry_roles.h"
#include "cache/cache_key.h"
#include "cache/typed_cache.h"
#include "rocksdb/slice.h"
#include "rocksdb/status.h"
#include "util/coding.h"
namespace ROCKSDB_NAMESPACE {
// CacheReservationManager is an interface for reserving cache space for the
// memory used
class CacheReservationManager {
public:
// CacheReservationHandle is for managing the lifetime of a cache reservation
// for an incremental amount of memory used (i.e, incremental_memory_used)
class CacheReservationHandle {
public:
virtual ~CacheReservationHandle() {}
};
virtual ~CacheReservationManager() {}
virtual Status UpdateCacheReservation(std::size_t new_memory_used) = 0;
// TODO(hx235): replace the usage of
// `UpdateCacheReservation(memory_used_delta, increase)` with
// `UpdateCacheReservation(new_memory_used)` so that we only have one
// `UpdateCacheReservation` function
virtual Status UpdateCacheReservation(std::size_t memory_used_delta,
bool increase) = 0;
virtual Status MakeCacheReservation(
std::size_t incremental_memory_used,
std::unique_ptr<CacheReservationManager::CacheReservationHandle>
*handle) = 0;
virtual std::size_t GetTotalReservedCacheSize() = 0;
virtual std::size_t GetTotalMemoryUsed() = 0;
};
// CacheReservationManagerImpl implements interface CacheReservationManager
// for reserving cache space for the memory used by inserting/releasing dummy
// entries in the cache.
//
// This class is NOT thread-safe, except that GetTotalReservedCacheSize()
// can be called without external synchronization.
template <CacheEntryRole R>
class CacheReservationManagerImpl
: public CacheReservationManager,
public std::enable_shared_from_this<CacheReservationManagerImpl<R>> {
public:
class CacheReservationHandle
: public CacheReservationManager::CacheReservationHandle {
public:
CacheReservationHandle(
std::size_t incremental_memory_used,
std::shared_ptr<CacheReservationManagerImpl> cache_res_mgr);
~CacheReservationHandle() override;
private:
std::size_t incremental_memory_used_;
std::shared_ptr<CacheReservationManagerImpl> cache_res_mgr_;
};
// Construct a CacheReservationManagerImpl
// @param cache The cache where dummy entries are inserted and released for
// reserving cache space
// @param delayed_decrease If set true, then dummy entries won't be released
// immediately when memory usage decreases.
// Instead, it will be released when the memory usage
// decreases to 3/4 of what we have reserved so far.
// This is for saving some future dummy entry
// insertion when memory usage increases are likely to
// happen in the near future.
//
// REQUIRED: cache is not nullptr
explicit CacheReservationManagerImpl(std::shared_ptr<Cache> cache,
bool delayed_decrease = false);
// no copy constructor, copy assignment, move constructor, move assignment
CacheReservationManagerImpl(const CacheReservationManagerImpl &) = delete;
CacheReservationManagerImpl &operator=(const CacheReservationManagerImpl &) =
delete;
CacheReservationManagerImpl(CacheReservationManagerImpl &&) = delete;
CacheReservationManagerImpl &operator=(CacheReservationManagerImpl &&) =
delete;
~CacheReservationManagerImpl() override;
// One of the two ways of reserving/releasing cache space,
// see MakeCacheReservation() for the other.
//
// Use ONLY one of these two ways to prevent unexpected behavior.
//
// Insert and release dummy entries in the cache to
// match the size of total dummy entries with the least multiple of
// kSizeDummyEntry greater than or equal to new_mem_used
//
// Insert dummy entries if new_memory_used > cache_allocated_size_;
//
// Release dummy entries if new_memory_used < cache_allocated_size_
// (and new_memory_used < cache_allocated_size_ * 3/4
// when delayed_decrease is set true);
//
// Keey dummy entries the same if (1) new_memory_used == cache_allocated_size_
// or (2) new_memory_used is in the interval of
// [cache_allocated_size_ * 3/4, cache_allocated_size) when delayed_decrease
// is set true.
//
// @param new_memory_used The number of bytes used by new memory
// The most recent new_memoy_used passed in will be returned
// in GetTotalMemoryUsed() even when the call return non-ok status.
//
// Since the class is NOT thread-safe, external synchronization on the
// order of calling UpdateCacheReservation() is needed if you want
// GetTotalMemoryUsed() indeed returns the latest memory used.
//
// @return On inserting dummy entries, it returns Status::OK() if all dummy
// entry insertions succeed.
// Otherwise, it returns the first non-ok status;
// On releasing dummy entries, it always returns Status::OK().
// On keeping dummy entries the same, it always returns Status::OK().
Status UpdateCacheReservation(std::size_t new_memory_used) override;
Status UpdateCacheReservation(std::size_t /* memory_used_delta */,
bool /* increase */) override {
return Status::NotSupported();
}
// One of the two ways of reserving cache space and releasing is done through
// destruction of CacheReservationHandle.
// See UpdateCacheReservation() for the other way.
//
// Use ONLY one of these two ways to prevent unexpected behavior.
//
// Insert dummy entries in the cache for the incremental memory usage
// to match the size of total dummy entries with the least multiple of
// kSizeDummyEntry greater than or equal to the total memory used.
//
// A CacheReservationHandle is returned as an output parameter.
// The reserved dummy entries are automatically released on the destruction of
// this handle, which achieves better RAII per cache reservation.
//
// WARNING: Deallocate all the handles of the CacheReservationManager object
// before deallocating the object to prevent unexpected behavior.
//
// @param incremental_memory_used The number of bytes increased in memory
// usage.
//
// Calling GetTotalMemoryUsed() afterward will return the total memory
// increased by this number, even when calling MakeCacheReservation()
// returns non-ok status.
//
// Since the class is NOT thread-safe, external synchronization in
// calling MakeCacheReservation() is needed if you want
// GetTotalMemoryUsed() indeed returns the latest memory used.
//
// @param handle An pointer to std::unique_ptr<CacheReservationHandle> that
// manages the lifetime of the cache reservation represented by the
// handle.
//
// @return It returns Status::OK() if all dummy
// entry insertions succeed.
// Otherwise, it returns the first non-ok status;
//
// REQUIRES: handle != nullptr
Status MakeCacheReservation(
std::size_t incremental_memory_used,
std::unique_ptr<CacheReservationManager::CacheReservationHandle> *handle)
override;
// Return the size of the cache (which is a multiple of kSizeDummyEntry)
// successfully reserved by calling UpdateCacheReservation().
//
// When UpdateCacheReservation() returns non-ok status,
// calling GetTotalReservedCacheSize() after that might return a slightly
// smaller number than the actual reserved cache size due to
// the returned number will always be a multiple of kSizeDummyEntry
// and cache full might happen in the middle of inserting a dummy entry.
std::size_t GetTotalReservedCacheSize() override;
// Return the latest total memory used indicated by the most recent call of
// UpdateCacheReservation(std::size_t new_memory_used);
std::size_t GetTotalMemoryUsed() override;
static constexpr std::size_t GetDummyEntrySize() { return kSizeDummyEntry; }
// For testing only - it is to help ensure the CacheItemHelperForRole<R>
// accessed from CacheReservationManagerImpl and the one accessed from the
// test are from the same translation units
static const Cache::CacheItemHelper *TEST_GetCacheItemHelperForRole();
private:
static constexpr std::size_t kSizeDummyEntry = 256 * 1024;
Slice GetNextCacheKey();
Status ReleaseCacheReservation(std::size_t incremental_memory_used);
Status IncreaseCacheReservation(std::size_t new_mem_used);
Status DecreaseCacheReservation(std::size_t new_mem_used);
using CacheInterface = PlaceholderSharedCacheInterface<R>;
CacheInterface cache_;
bool delayed_decrease_;
std::atomic<std::size_t> cache_allocated_size_;
std::size_t memory_used_;
std::vector<Cache::Handle *> dummy_handles_;
CacheKey cache_key_;
};
class ConcurrentCacheReservationManager
: public CacheReservationManager,
public std::enable_shared_from_this<ConcurrentCacheReservationManager> {
public:
class CacheReservationHandle
: public CacheReservationManager::CacheReservationHandle {
public:
CacheReservationHandle(
std::shared_ptr<ConcurrentCacheReservationManager> cache_res_mgr,
std::unique_ptr<CacheReservationManager::CacheReservationHandle>
cache_res_handle) {
assert(cache_res_mgr && cache_res_handle);
cache_res_mgr_ = cache_res_mgr;
cache_res_handle_ = std::move(cache_res_handle);
}
~CacheReservationHandle() override {
std::lock_guard<std::mutex> lock(cache_res_mgr_->cache_res_mgr_mu_);
cache_res_handle_.reset();
}
private:
std::shared_ptr<ConcurrentCacheReservationManager> cache_res_mgr_;
std::unique_ptr<CacheReservationManager::CacheReservationHandle>
cache_res_handle_;
};
explicit ConcurrentCacheReservationManager(
std::shared_ptr<CacheReservationManager> cache_res_mgr) {
cache_res_mgr_ = std::move(cache_res_mgr);
}
ConcurrentCacheReservationManager(const ConcurrentCacheReservationManager &) =
delete;
ConcurrentCacheReservationManager &operator=(
const ConcurrentCacheReservationManager &) = delete;
ConcurrentCacheReservationManager(ConcurrentCacheReservationManager &&) =
delete;
ConcurrentCacheReservationManager &operator=(
ConcurrentCacheReservationManager &&) = delete;
~ConcurrentCacheReservationManager() override {}
inline Status UpdateCacheReservation(std::size_t new_memory_used) override {
std::lock_guard<std::mutex> lock(cache_res_mgr_mu_);
return cache_res_mgr_->UpdateCacheReservation(new_memory_used);
}
inline Status UpdateCacheReservation(std::size_t memory_used_delta,
bool increase) override {
std::lock_guard<std::mutex> lock(cache_res_mgr_mu_);
std::size_t total_mem_used = cache_res_mgr_->GetTotalMemoryUsed();
Status s;
if (!increase) {
assert(total_mem_used >= memory_used_delta);
s = cache_res_mgr_->UpdateCacheReservation(total_mem_used -
memory_used_delta);
} else {
s = cache_res_mgr_->UpdateCacheReservation(total_mem_used +
memory_used_delta);
}
return s;
}
inline Status MakeCacheReservation(
std::size_t incremental_memory_used,
std::unique_ptr<CacheReservationManager::CacheReservationHandle> *handle)
override {
std::unique_ptr<CacheReservationManager::CacheReservationHandle>
wrapped_handle;
Status s;
{
std::lock_guard<std::mutex> lock(cache_res_mgr_mu_);
s = cache_res_mgr_->MakeCacheReservation(incremental_memory_used,
&wrapped_handle);
}
(*handle).reset(
new ConcurrentCacheReservationManager::CacheReservationHandle(
std::enable_shared_from_this<
ConcurrentCacheReservationManager>::shared_from_this(),
std::move(wrapped_handle)));
return s;
}
inline std::size_t GetTotalReservedCacheSize() override {
return cache_res_mgr_->GetTotalReservedCacheSize();
}
inline std::size_t GetTotalMemoryUsed() override {
std::lock_guard<std::mutex> lock(cache_res_mgr_mu_);
return cache_res_mgr_->GetTotalMemoryUsed();
}
private:
std::mutex cache_res_mgr_mu_;
std::shared_ptr<CacheReservationManager> cache_res_mgr_;
};
} // namespace ROCKSDB_NAMESPACE