Make the underlying map to be a boost::intrusive::unordered_set<timestamped_val> instead of std::unordered_set<Key, timestamped_val>. Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
353 lines
12 KiB
C++
353 lines
12 KiB
C++
/*
|
|
* Copyright (C) 2016 ScyllaDB
|
|
*/
|
|
|
|
/*
|
|
* This file is part of Scylla.
|
|
*
|
|
* Scylla is free software: you can redistribute it and/or modify
|
|
* it under the terms of the GNU Affero General Public License as published by
|
|
* the Free Software Foundation, either version 3 of the License, or
|
|
* (at your option) any later version.
|
|
*
|
|
* Scylla is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
* GNU General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU General Public License
|
|
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
|
|
*/
|
|
|
|
#pragma once
|
|
|
|
#include <chrono>
|
|
#include <unordered_map>
|
|
#include <boost/intrusive/unordered_set.hpp>
|
|
|
|
#include <seastar/core/timer.hh>
|
|
|
|
#include "utils/exceptions.hh"
|
|
|
|
namespace bi = boost::intrusive;
|
|
|
|
namespace utils {
|
|
// Simple variant of the "LoadingCache" used for permissions in origin.
|
|
|
|
typedef lowres_clock loading_cache_clock_type;
|
|
|
|
template<typename _Tp, typename _Key, typename _Hash, typename _EqualPred>
|
|
class timestamped_val : public bi::unordered_set_base_hook<> {
|
|
public:
|
|
typedef _Key key_type;
|
|
typedef _Tp value_type;
|
|
|
|
private:
|
|
std::experimental::optional<_Tp> _opt_value;
|
|
loading_cache_clock_type::time_point _loaded;
|
|
loading_cache_clock_type::time_point _last_read;
|
|
_Key _key;
|
|
|
|
public:
|
|
struct key_eq {
|
|
bool operator()(const _Key& k, const timestamped_val& c) const {
|
|
return _EqualPred()(k, c.key());
|
|
}
|
|
|
|
bool operator()(const timestamped_val& c, const _Key& k) const {
|
|
return _EqualPred()(c.key(), k);
|
|
}
|
|
};
|
|
|
|
timestamped_val(const _Key& key)
|
|
: _loaded(loading_cache_clock_type::now())
|
|
, _last_read(_loaded)
|
|
, _key(key) {}
|
|
|
|
timestamped_val(_Key&& key)
|
|
: _loaded(loading_cache_clock_type::now())
|
|
, _last_read(_loaded)
|
|
, _key(std::move(key)) {}
|
|
|
|
timestamped_val(const timestamped_val&) = default;
|
|
timestamped_val(timestamped_val&&) = default;
|
|
|
|
// Make sure copy/move-assignments don't go through the template below
|
|
timestamped_val& operator=(const timestamped_val&) = default;
|
|
timestamped_val& operator=(timestamped_val&) = default;
|
|
timestamped_val& operator=(timestamped_val&&) = default;
|
|
|
|
template <typename U>
|
|
timestamped_val& operator=(U&& new_val) {
|
|
_opt_value = std::forward<U>(new_val);
|
|
_loaded = loading_cache_clock_type::now();
|
|
return *this;
|
|
}
|
|
|
|
const _Tp& value() {
|
|
_last_read = loading_cache_clock_type::now();
|
|
return _opt_value.value();
|
|
}
|
|
|
|
explicit operator bool() const noexcept {
|
|
return bool(_opt_value);
|
|
}
|
|
|
|
loading_cache_clock_type::time_point last_read() const noexcept {
|
|
return _last_read;
|
|
}
|
|
|
|
loading_cache_clock_type::time_point loaded() const noexcept {
|
|
return _loaded;
|
|
}
|
|
|
|
const _Key& key() const {
|
|
return _key;
|
|
}
|
|
|
|
friend bool operator==(const timestamped_val& a, const timestamped_val& b){
|
|
return _EqualPred()(a.key(), b.key());
|
|
}
|
|
|
|
friend std::size_t hash_value(const timestamped_val& v) {
|
|
return _Hash()(v.key());
|
|
}
|
|
};
|
|
|
|
class shared_mutex {
|
|
private:
|
|
lw_shared_ptr<semaphore> _mutex_ptr;
|
|
|
|
public:
|
|
shared_mutex() : _mutex_ptr(make_lw_shared<semaphore>(1)) {}
|
|
semaphore& get() const noexcept {
|
|
return *_mutex_ptr;
|
|
}
|
|
};
|
|
|
|
template<typename _Key,
|
|
typename _Tp,
|
|
typename _Hash = std::hash<_Key>,
|
|
typename _Pred = std::equal_to<_Key>,
|
|
typename _Alloc = std::allocator<timestamped_val<_Tp, _Key, _Hash, _Pred>>,
|
|
typename SharedMutexMapAlloc = std::allocator<std::pair<const _Key, shared_mutex>>>
|
|
class loading_cache {
|
|
private:
|
|
typedef timestamped_val<_Tp, _Key, _Hash, _Pred> ts_value_type;
|
|
typedef bi::unordered_set<ts_value_type> set_type;
|
|
typedef std::unordered_map<_Key, shared_mutex, _Hash, _Pred, SharedMutexMapAlloc> write_mutex_map_type;
|
|
typedef loading_cache<_Key, _Tp, _Hash, _Pred, _Alloc> _MyType;
|
|
typedef typename set_type::bucket_traits bi_set_bucket_traits;
|
|
|
|
static constexpr int num_buckets = 1021; // A prime number close to 1024
|
|
|
|
public:
|
|
typedef _Tp value_type;
|
|
typedef _Key key_type;
|
|
typedef typename set_type::iterator iterator;
|
|
|
|
template<typename Func>
|
|
loading_cache(size_t max_size, std::chrono::milliseconds expiry, std::chrono::milliseconds refresh, logging::logger& logger, Func&& load)
|
|
: _buckets(num_buckets)
|
|
, _set(bi_set_bucket_traits(_buckets.data(), _buckets.size()))
|
|
, _max_size(max_size)
|
|
, _expiry(expiry)
|
|
, _refresh(refresh)
|
|
, _logger(logger)
|
|
, _load(std::forward<Func>(load)) {
|
|
|
|
// If expiration period is zero - caching is disabled
|
|
if (!caching_enabled()) {
|
|
return;
|
|
}
|
|
|
|
// Sanity check: if expiration period is given then non-zero refresh period and maximal size are required
|
|
if (_refresh == std::chrono::milliseconds(0) || _max_size == 0) {
|
|
throw exceptions::configuration_exception("loading_cache: caching is enabled but refresh period and/or max_size are zero");
|
|
}
|
|
|
|
_timer.set_callback([this] { on_timer(); });
|
|
_timer.arm(_refresh);
|
|
}
|
|
|
|
~loading_cache() {
|
|
_set.clear_and_dispose([] (ts_value_type* ptr) { loading_cache::destroy_ts_value(ptr); });
|
|
}
|
|
|
|
future<_Tp> get(const _Key& k) {
|
|
// If caching is disabled - always load in the foreground
|
|
if (!caching_enabled()) {
|
|
return _load(k);
|
|
}
|
|
|
|
// If the key is not in the cache yet, then find_or_create() is going to
|
|
// create a new uninitialized value in the map. If the value is already
|
|
// in the cache (the fast path) simply return the value. Otherwise, take
|
|
// the mutex and try to load the value (the slow path).
|
|
iterator ts_value_it = find_or_create(k);
|
|
if (*ts_value_it) {
|
|
return make_ready_future<_Tp>(ts_value_it->value());
|
|
} else {
|
|
return slow_load(k);
|
|
}
|
|
}
|
|
|
|
private:
|
|
bool caching_enabled() const {
|
|
return _expiry != std::chrono::milliseconds(0);
|
|
}
|
|
|
|
/// Look for the entry with the given key. It it doesn't exist - create a new one and add it to the _set.
|
|
///
|
|
/// \param k The key to look for
|
|
///
|
|
/// \return An iterator to the value with the given key (always dirrerent from _set.end())
|
|
template <typename KeyType>
|
|
iterator find_or_create(KeyType&& k) {
|
|
iterator i = _set.find(k, _Hash(), typename ts_value_type::key_eq());
|
|
if (i == _set.end()) {
|
|
ts_value_type* new_ts_val = _Alloc().allocate(1);
|
|
new(new_ts_val) ts_value_type(std::forward<KeyType>(k));
|
|
auto p = _set.insert(*new_ts_val);
|
|
i = p.first;
|
|
}
|
|
|
|
return i;
|
|
}
|
|
|
|
static void destroy_ts_value(ts_value_type* val) {
|
|
val->~ts_value_type();
|
|
_Alloc().deallocate(val, 1);
|
|
}
|
|
|
|
future<_Tp> slow_load(const _Key& k) {
|
|
// If the key is not in the cache yet, then _write_mutex_map[k] is going
|
|
// to create a new value with the initialized mutex. The mutex is going
|
|
// to serialize the producers and only the first one is going to
|
|
// actually issue a load operation and initialize the value with the
|
|
// received result. The rest are going to see (and read) the initialized
|
|
// value when they enter the critical section.
|
|
shared_mutex sm = _write_mutex_map[k];
|
|
return with_semaphore(sm.get(), 1, [this, k] {
|
|
iterator ts_value_it = find_or_create(k);
|
|
if (*ts_value_it) {
|
|
return make_ready_future<_Tp>(ts_value_it->value());
|
|
}
|
|
_logger.trace("{}: storing the value for the first time", k);
|
|
return _load(k).then([this, k] (_Tp t) {
|
|
// we have to "re-read" the _set here because the value may have been evicted by now
|
|
iterator ts_value_it = find_or_create(std::move(k));
|
|
*ts_value_it = std::move(t);
|
|
return make_ready_future<_Tp>(ts_value_it->value());
|
|
});
|
|
}).finally([sm] {});
|
|
}
|
|
|
|
future<> reload(ts_value_type& ts_val) {
|
|
return _load(ts_val.key()).then_wrapped([this, &ts_val] (auto&& f) {
|
|
// The exceptions are related to the load operation itself.
|
|
// We should ignore them for the background reads - if
|
|
// they persist the value will age and will be reloaded in
|
|
// the forground. If the foreground READ fails the error
|
|
// will be propagated up to the user and will fail the
|
|
// corresponding query.
|
|
try {
|
|
ts_val = f.get0();
|
|
} catch (std::exception& e) {
|
|
_logger.debug("{}: reload failed: {}", ts_val.key(), e.what());
|
|
} catch (...) {
|
|
_logger.debug("{}: reload failed: unknown error", ts_val.key());
|
|
}
|
|
});
|
|
}
|
|
|
|
void erase(iterator it) {
|
|
_set.erase_and_dispose(it, [] (ts_value_type* ptr) { loading_cache::destroy_ts_value(ptr); });
|
|
}
|
|
|
|
// We really miss the std::erase_if()... :(
|
|
void drop_expired() {
|
|
auto now = loading_cache_clock_type::now();
|
|
auto i = _set.begin();
|
|
auto e = _set.end();
|
|
|
|
while (i != e) {
|
|
// An entry should be discarded if it hasn't been reloaded for too long or nobody cares about it anymore
|
|
auto since_last_read = now - i->last_read();
|
|
auto since_loaded = now - i->loaded();
|
|
if (_expiry < since_last_read || _expiry < since_loaded) {
|
|
using namespace std::chrono;
|
|
_logger.trace("drop_expired(): {}: dropping the entry: _expiry {}, ms passed since: loaded {} last_read {}", i->key(), _expiry.count(), duration_cast<milliseconds>(since_loaded).count(), duration_cast<milliseconds>(since_last_read).count());
|
|
erase(i++);
|
|
continue;
|
|
}
|
|
++i;
|
|
}
|
|
}
|
|
|
|
// Shrink the cache to the _max_size discarding the least recently used items
|
|
void shrink() {
|
|
if (_max_size != 0 && _set.size() > _max_size) {
|
|
std::vector<iterator> tmp;
|
|
tmp.reserve(_set.size());
|
|
|
|
iterator i = _set.begin();
|
|
while (i != _set.end()) {
|
|
tmp.emplace_back(i++);
|
|
}
|
|
|
|
std::sort(tmp.begin(), tmp.end(), [] (iterator i1, iterator i2) {
|
|
return i1->last_read() < i2->last_read();
|
|
});
|
|
|
|
tmp.resize(_set.size() - _max_size);
|
|
std::for_each(tmp.begin(), tmp.end(), [this] (auto& k) {
|
|
using namespace std::chrono;
|
|
_logger.trace("shrink(): {}: dropping the entry: ms since last_read {}", k->key(), duration_cast<milliseconds>(loading_cache_clock_type::now() - k->last_read()).count());
|
|
this->erase(k);
|
|
});
|
|
}
|
|
}
|
|
|
|
void on_timer() {
|
|
_logger.trace("on_timer(): start");
|
|
|
|
auto timer_start_tp = loading_cache_clock_type::now();
|
|
|
|
// Clear all cached mutexes
|
|
_write_mutex_map.clear();
|
|
|
|
// Clean up items that were not touched for the whole _expiry period.
|
|
drop_expired();
|
|
|
|
// Remove the least recently used items if map is too big.
|
|
shrink();
|
|
|
|
// Reload all those which vlaue needs to be reloaded.
|
|
parallel_for_each(_set.begin(), _set.end(), [this, curr_time = timer_start_tp] (auto& ts_val) {
|
|
_logger.trace("on_timer(): {}: checking the value age", ts_val.key());
|
|
if (ts_val && ts_val.loaded() + _refresh < curr_time) {
|
|
_logger.trace("on_timer(): {}: reloading the value", ts_val.key());
|
|
return this->reload(ts_val);
|
|
}
|
|
return now();
|
|
}).finally([this, timer_start_tp] {
|
|
_logger.trace("on_timer(): rearming");
|
|
_timer.arm(timer_start_tp + _refresh);
|
|
});
|
|
}
|
|
|
|
std::vector<typename set_type::bucket_type> _buckets;
|
|
set_type _set;
|
|
write_mutex_map_type _write_mutex_map;
|
|
size_t _max_size;
|
|
std::chrono::milliseconds _expiry;
|
|
std::chrono::milliseconds _refresh;
|
|
logging::logger& _logger;
|
|
std::function<future<_Tp>(const _Key&)> _load;
|
|
timer<lowres_clock> _timer;
|
|
};
|
|
|
|
}
|
|
|