connection_pool: fix excessive idle CPU usage while waiting for requests

[ThCompiler]

While using ConnectionPool, it was found that it uses 100% cpu for its threads when at rest. After examining its operation, it was realized that the _request_process_loop runs in an infinite loop, without blocking on waiting for the next request. To reduce the load, I suggest using already implemented wait for the next element in the queue's from package queue .

To check the solution, I implemented a simple test that calculates the n-th element of the Fibonacci sequence using a ConnectionPool running in the background:

importqueueimporttimefromtarantoolimportConnectionfromtarantool.connection_poolimportConnectionPoolasOriginalConnectionPooldeffib(n):
a=0b=1for_inrange(n):
a, b=b, a+breturnaclassConnectionPool(OriginalConnectionPool):
# Override of the standard pool loop, which loaded the processor at 100% # with an infinite constant loopdef_request_process_loop( # noqa: max-complexity=8self, key, unit, last_refresh,
):
""" Request process background loop for a pool server. Started in a separate thread, one thread per server. :param key: Result of :meth:`~tarantool.connection_pool._make_key`. :type key: :obj:`str` :param unit: Server metainfo. :type unit: :class:`~tarantool.connection_pool.PoolUnit` :param last_refresh: Time of last metainfo refresh. :type last_refresh: :obj:`float` """whileunit.request_processing_enabled:
try:
task=unit.input_queue.get(timeout=self.refresh_delay)
exceptqueue.Empty:
task=Noneiftask:
method=getattr(Connection, task.method_name)
try:
resp=method(unit.conn, *task.args, **task.kwargs)
exceptExceptionasexc:
unit.output_queue.put(exc)
else:
unit.output_queue.put(resp)
now=time.time()
ifnow-last_refresh>self.refresh_delay:
self._refresh_state(key)
last_refresh=time.time()
deftest_bench_updated_pool(benchmark):
pool=ConnectionPool(
addrs=[
{
'host': '127.0.0.1',
'port': 3301,
},
{
'host': '127.0.0.1',
'port': 3302,
},
{
'host': '127.0.0.1',
'port': 3303,
},
],
user='guest',
)
benchmark(fib, 1000000)
pool.close()
deftest_bench_original_pool(benchmark):
pool=OriginalConnectionPool(
addrs=[
{
'host': '127.0.0.1',
'port': 3301,
},
{
'host': '127.0.0.1',
'port': 3302,
},
{
'host': '127.0.0.1',
'port': 3303,
},
],
user='guest',
)
benchmark(fib, 1000000)
pool.close()

I ran the test with the changes and the old ConnectionPool behavior. So i got the following test results:

I also looked at the distribution of processor activity using py-spy for each of the tests:

• for current behavior:

• for new behavior:

As you can see from the py-spy snapshots, after the change, the processor is almost fully used for calculating Fibonacci sequence, while with the current behavior, most of the time is spent on _request_process_loop.

[bigbes]

lgtm

[oleg-jukovec]

Please, update the commit message (see above) and add a new entry to the CHANGELOG.md.

[ThCompiler]

I had to update the build os version for the readthedocs to 24.04, as it stopped supporting 20.04. 24.04 was selected based on a similar version update in #330.

[ThCompiler]

I checked failed jobs on CI.

Windows system jobs fail because the default version of setuptools is 65.5.0 in the CI image. When dependencies are installed in a step, setuptools updates to the latest version 82.0.1 which meets the requirements of >=75.3.2. The pkg_resources has been removed in this version of setuptools. https://setuptools.pypa.io/en/latest/. As a quick solution, I can set the upper bound for the setuptools version to <82.0 in the testing requirements.

In Linux, a jobs that use the master Tarantool branch fail because the maximum decimal precision has been increased to 76 in Tarantool 3.5. tarantool/tarantool#11497. I can try to fix this in the implementation of the decimal type for msgpack.

Should I make these changes in this pull request?

[ThCompiler]

I decided to have a separate PR(#342) for the CI fix because there are a lot of changes related to the new decimal precision. Could you check this PR please? After it we will be able to successfully finish workflow in this PR.

[Mockird31]

Thanks for the patch!