Add NUMA topology constraint to solver scheduler
This adds the numa topology constraint that matches the numa topology filter in filter scheduler. Change-Id: Id80f0a7364132204c82f50e61b0c77913485ae04
This commit is contained in:
Xinyuan Huang
parent
010cc4fe58
commit
e63eaf53ce
|
|
@ -0,0 +1,72 @@
|
|||
# Copyright (c) 2015 Cisco Systems Inc.
|
||||
# All Rights Reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License"); you may
|
||||
# not use this file except in compliance with the License. You may obtain
|
||||
# a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
|
||||
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
|
||||
# License for the specific language governing permissions and limitations
|
||||
# under the License.
|
||||
|
||||
import copy
|
||||
|
||||
from oslo_log import log as logging
|
||||
|
||||
from nova.scheduler.filters import numa_topology_filter
|
||||
from nova_solverscheduler.scheduler.solvers import constraints
|
||||
|
||||
LOG = logging.getLogger(__name__)
|
||||
|
||||
|
||||
class NUMATopologyConstraint(constraints.BaseLinearConstraint):
|
||||
"""Constraint on requested NUMA topology."""
|
||||
|
||||
def __init__(self):
|
||||
super(NUMATopologyConstraint, self).__init__()
|
||||
self.host_filter = numa_topology_filter.NUMATopologyFilter()
|
||||
|
||||
def _get_acceptable_instance_num(self, host_state, filter_properties,
|
||||
max_num):
|
||||
instance = filter_properties['request_spec']['instance_properties']
|
||||
acceptable_num = 0
|
||||
while acceptable_num < max_num:
|
||||
if self.host_filter.host_passes(host_state, filter_properties):
|
||||
acceptable_num += 1
|
||||
host_state.consume_from_instance(instance)
|
||||
else:
|
||||
break
|
||||
return acceptable_num
|
||||
|
||||
def get_constraint_matrix(self, hosts, filter_properties):
|
||||
num_hosts = len(hosts)
|
||||
num_instances = filter_properties.get('num_instances')
|
||||
|
||||
constraint_matrix = [[True for j in xrange(num_instances)]
|
||||
for i in xrange(num_hosts)]
|
||||
|
||||
for i in xrange(num_hosts):
|
||||
host_state = copy.deepcopy(hosts[i])
|
||||
acceptable_instance_num = self._get_acceptable_instance_num(
|
||||
host_state, filter_properties, num_instances)
|
||||
|
||||
if acceptable_instance_num < num_instances:
|
||||
inacceptable_num = num_instances - acceptable_instance_num
|
||||
constraint_matrix[i] = (
|
||||
[True for j in xrange(acceptable_instance_num)] +
|
||||
[False for j in xrange(inacceptable_num)])
|
||||
|
||||
LOG.debug("%(host)s can accept %(num)s requested instances "
|
||||
"according to NUMATopologyConstraint.",
|
||||
{'host': hosts[i],
|
||||
'num': acceptable_instance_num})
|
||||
|
||||
numa_topology_limit = host_state.limits.get('numa_topology')
|
||||
if numa_topology_limit:
|
||||
hosts[i].limits['numa_topology'] = numa_topology_limit
|
||||
|
||||
return constraint_matrix
|
||||
|
|
@ -0,0 +1,112 @@
|
|||
# Copyright (c) 2015 Cisco Systems Inc.
|
||||
# All Rights Reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License"); you may
|
||||
# not use this file except in compliance with the License. You may obtain
|
||||
# a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
|
||||
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
|
||||
# License for the specific language governing permissions and limitations
|
||||
# under the License.
|
||||
|
||||
import mock
|
||||
from oslo_serialization import jsonutils
|
||||
|
||||
from nova import objects
|
||||
from nova.objects import base as obj_base
|
||||
from nova import test
|
||||
from nova.tests.unit import fake_instance
|
||||
from nova_solverscheduler.scheduler.solvers.constraints \
|
||||
import numa_topology_constraint
|
||||
from nova_solverscheduler.tests.scheduler import solver_scheduler_fakes \
|
||||
as fakes
|
||||
|
||||
NUMA_TOPOLOGY = objects.NUMATopology(
|
||||
cells=[
|
||||
objects.NUMACell(id=0, cpuset=set([1, 2]), memory=512,
|
||||
cpu_usage=0, memory_usage=0, mempages=[],
|
||||
siblings=[], pinned_cpus=set([])),
|
||||
objects.NUMACell(id=1, cpuset=set([3, 4]), memory=512,
|
||||
cpu_usage=0, memory_usage=0, mempages=[],
|
||||
siblings=[], pinned_cpus=set([]))
|
||||
]
|
||||
)
|
||||
|
||||
|
||||
class TestNUMATopologyConstraint(test.NoDBTestCase):
|
||||
|
||||
def setUp(self):
|
||||
super(TestNUMATopologyConstraint, self).setUp()
|
||||
self.constraint_cls = numa_topology_constraint.NUMATopologyConstraint
|
||||
|
||||
def _gen_fake_hosts(self):
|
||||
host1 = fakes.FakeSolverSchedulerHostState('host1', 'node1',
|
||||
{
|
||||
'numa_topology': objects.NUMATopology(
|
||||
cells=[
|
||||
objects.NUMACell(id=0, cpuset=set([1, 2]),
|
||||
memory=1024, cpu_usage=0, memory_usage=0,
|
||||
mempages=[], siblings=[], pinned_cpus=set([])),
|
||||
objects.NUMACell(id=1, cpuset=set([3, 4]),
|
||||
memory=1024, cpu_usage=0, memory_usage=0,
|
||||
mempages=[], siblings=[], pinned_cpus=set([]))]),
|
||||
'pci_stats': None
|
||||
})
|
||||
host2 = fakes.FakeSolverSchedulerHostState('host2', 'node1',
|
||||
{
|
||||
'numa_topology': objects.NUMATopology(
|
||||
cells=[
|
||||
objects.NUMACell(id=0, cpuset=set([1, 2]),
|
||||
memory=1024, cpu_usage=0, memory_usage=0,
|
||||
mempages=[], siblings=[], pinned_cpus=set([])),
|
||||
objects.NUMACell(id=1, cpuset=set([3, 4]),
|
||||
memory=512, cpu_usage=0, memory_usage=0,
|
||||
mempages=[], siblings=[], pinned_cpus=set([]))]),
|
||||
'pci_stats': None
|
||||
})
|
||||
host3 = fakes.FakeSolverSchedulerHostState('host3', 'node1',
|
||||
{
|
||||
'numa_topology': objects.NUMATopology(
|
||||
cells=[
|
||||
objects.NUMACell(id=0, cpuset=set([1, 2]),
|
||||
memory=512, cpu_usage=0, memory_usage=0,
|
||||
mempages=[], siblings=[], pinned_cpus=set([])),
|
||||
objects.NUMACell(id=1, cpuset=set([3]),
|
||||
memory=512, cpu_usage=0, memory_usage=0,
|
||||
mempages=[], siblings=[], pinned_cpus=set([]))]),
|
||||
'pci_stats': None
|
||||
})
|
||||
hosts = [host1, host2, host3]
|
||||
return hosts
|
||||
|
||||
def test_get_constraint_matrix(self):
|
||||
self.flags(ram_allocation_ratio=1)
|
||||
self.flags(cpu_allocation_ratio=2)
|
||||
|
||||
instance_topology = objects.InstanceNUMATopology(cells=[
|
||||
objects.InstanceNUMACell(id=0, cpuset=set([1, 2]), memory=512),
|
||||
objects.InstanceNUMACell(id=1, cpuset=set([3, 4]), memory=512)]
|
||||
)
|
||||
instance = fake_instance.fake_instance_obj(mock.sentinel.ctx,
|
||||
root_gb=0, ephemeral_gb=0, memory_mb=0, vcpus=0,)
|
||||
instance.numa_topology = instance_topology
|
||||
filter_properties = {
|
||||
'request_spec': {
|
||||
'instance_properties': jsonutils.to_primitive(
|
||||
obj_base.obj_to_primitive(instance))},
|
||||
'num_instances': 2}
|
||||
fake_hosts = self._gen_fake_hosts()
|
||||
|
||||
expected_cons_mat = [
|
||||
[True, True],
|
||||
[True, False],
|
||||
[False, False]]
|
||||
|
||||
cons_mat = self.constraint_cls().get_constraint_matrix(
|
||||
fake_hosts, filter_properties)
|
||||
|
||||
self.assertEqual(expected_cons_mat, cons_mat)
|
||||
Loading…
Reference in New Issue