Files
scylladb/mutation_reader.cc
Avi Kivity 9ac730dcc9 mutation_reader: make restricting_mutation_reader even more restricting
While limiting the number of concurrently executing sstable readers reduces
our memory load, the queued readers, although consuming a small amount of
memory, can still grow without bounds.

To limit the damage, add two limits on the queue:
 - a timeout, which is equal to the read timeout
 - a queue length limit, which is equal to 2% of the shard memory divided
   by an estimate of the queued request size (1kb)

Together, these limits bound the amount of memory needed by queued disk
requests in case the disk can't keep up.
Message-Id: <1467206055-30769-1-git-send-email-avi@scylladb.com>
2016-06-29 15:17:35 +02:00

231 lines
7.9 KiB
C++

/*
* Copyright (C) 2015 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/>.
*/
#include <boost/range/algorithm/heap_algorithm.hpp>
#include <boost/range/algorithm/reverse.hpp>
#include "mutation_reader.hh"
#include "core/future-util.hh"
#include "utils/move.hh"
namespace stdx = std::experimental;
template<typename T>
T move_and_clear(T& obj) {
T x = std::move(obj);
obj = T();
return x;
}
// Combines multiple mutation_readers into one.
class combined_reader final : public mutation_reader::impl {
std::vector<mutation_reader> _readers;
struct mutation_and_reader {
streamed_mutation m;
mutation_reader* read;
};
std::vector<mutation_and_reader> _ptables;
// comparison function for std::make_heap()/std::push_heap()
static bool heap_compare(const mutation_and_reader& a, const mutation_and_reader& b) {
auto&& s = a.m.schema();
// order of comparison is inverted, because heaps produce greatest value first
return b.m.decorated_key().less_compare(*s, a.m.decorated_key());
}
std::vector<streamed_mutation> _current;
std::vector<mutation_reader*> _next;
private:
future<> prepare_next() {
return parallel_for_each(_next, [this] (mutation_reader* mr) {
return (*mr)().then([this, mr] (streamed_mutation_opt next) {
if (next) {
_ptables.emplace_back(mutation_and_reader { std::move(*next), mr });
boost::range::push_heap(_ptables, &heap_compare);
}
});
}).then([this] {
_next.clear();
});
}
// Produces next mutation or disengaged optional if there are no more.
future<streamed_mutation_opt> next() {
if (_current.empty() && !_next.empty()) {
return prepare_next().then([this] { return next(); });
}
if (_ptables.empty()) {
return make_ready_future<streamed_mutation_opt>();
};
while (!_ptables.empty()) {
boost::range::pop_heap(_ptables, &heap_compare);
auto& candidate = _ptables.back();
streamed_mutation& m = candidate.m;
if (!_current.empty() && !_current.back().decorated_key().equal(*m.schema(), m.decorated_key())) {
// key has changed, so emit accumulated mutation
boost::range::push_heap(_ptables, &heap_compare);
return make_ready_future<streamed_mutation_opt>(merge_mutations(move_and_clear(_current)));
}
_current.emplace_back(std::move(m));
_next.emplace_back(candidate.read);
_ptables.pop_back();
}
return make_ready_future<streamed_mutation_opt>(merge_mutations(move_and_clear(_current)));
}
public:
combined_reader(std::vector<mutation_reader> readers)
: _readers(std::move(readers))
{
_next.reserve(_readers.size());
_current.reserve(_readers.size());
_ptables.reserve(_readers.size());
for (auto&& r : _readers) {
_next.emplace_back(&r);
}
}
virtual future<streamed_mutation_opt> operator()() override {
return next();
}
};
mutation_reader
make_combined_reader(std::vector<mutation_reader> readers) {
return make_mutation_reader<combined_reader>(std::move(readers));
}
mutation_reader
make_combined_reader(mutation_reader&& a, mutation_reader&& b) {
std::vector<mutation_reader> v;
v.reserve(2);
v.push_back(std::move(a));
v.push_back(std::move(b));
return make_combined_reader(std::move(v));
}
class reader_returning final : public mutation_reader::impl {
streamed_mutation _m;
bool _done = false;
public:
reader_returning(streamed_mutation m) : _m(std::move(m)) {
}
virtual future<streamed_mutation_opt> operator()() override {
if (_done) {
return make_ready_future<streamed_mutation_opt>();
} else {
_done = true;
return make_ready_future<streamed_mutation_opt>(std::move(_m));
}
}
};
mutation_reader make_reader_returning(mutation m) {
return make_mutation_reader<reader_returning>(streamed_mutation_from_mutation(std::move(m)));
}
mutation_reader make_reader_returning(streamed_mutation m) {
return make_mutation_reader<reader_returning>(std::move(m));
}
class reader_returning_many final : public mutation_reader::impl {
std::vector<streamed_mutation> _m;
bool _done = false;
public:
reader_returning_many(std::vector<streamed_mutation> m) : _m(std::move(m)) {
boost::range::reverse(_m);
}
virtual future<streamed_mutation_opt> operator()() override {
if (_m.empty()) {
return make_ready_future<streamed_mutation_opt>();
}
auto m = std::move(_m.back());
_m.pop_back();
return make_ready_future<streamed_mutation_opt>(std::move(m));
}
};
mutation_reader make_reader_returning_many(std::vector<mutation> mutations, query::clustering_key_filtering_context ck_filtering) {
std::vector<streamed_mutation> streamed_mutations;
streamed_mutations.reserve(mutations.size());
for (auto& m : mutations) {
const query::clustering_row_ranges& ck_ranges = ck_filtering.get_ranges(m.key());
auto mp = mutation_partition(std::move(m.partition()), *m.schema(), ck_ranges);
auto sm = streamed_mutation_from_mutation(mutation(m.schema(), m.decorated_key(), std::move(mp)));
streamed_mutations.emplace_back(std::move(sm));
}
return make_mutation_reader<reader_returning_many>(std::move(streamed_mutations));
}
mutation_reader make_reader_returning_many(std::vector<streamed_mutation> mutations) {
return make_mutation_reader<reader_returning_many>(std::move(mutations));
}
class empty_reader final : public mutation_reader::impl {
public:
virtual future<streamed_mutation_opt> operator()() override {
return make_ready_future<streamed_mutation_opt>();
}
};
mutation_reader make_empty_reader() {
return make_mutation_reader<empty_reader>();
}
class restricting_mutation_reader : public mutation_reader::impl {
const restricted_mutation_reader_config& _config;
unsigned _weight = 0;
bool _waited = false;
mutation_reader _base;
public:
restricting_mutation_reader(const restricted_mutation_reader_config& config, unsigned weight, mutation_reader&& base)
: _config(config), _weight(weight), _base(std::move(base)) {
if (_config.sem->waiters() >= _config.max_queue_length) {
_config.raise_queue_overloaded_exception();
}
}
~restricting_mutation_reader() {
if (_waited) {
_config.sem->signal(_weight);
}
}
future<streamed_mutation_opt> operator()() override {
// FIXME: we should defer freeing until the mutation is freed, perhaps,
// rather than just returned
if (_waited) {
return _base();
}
auto waited = _config.timeout.count() != 0
? _config.sem->wait(_config.timeout, _weight)
: _config.sem->wait(_weight);
return waited.then([this] {
_waited = true;
return _base();
});
}
};
mutation_reader
make_restricted_reader(const restricted_mutation_reader_config& config, unsigned weight, mutation_reader&& base) {
return make_mutation_reader<restricting_mutation_reader>(config, weight, std::move(base));
}