# -*- encoding: utf-8 -*-
# Copyright (c) 2015 Intel Corp
#
# Authors: Junjie-Huang <junjie.huang@intel.com>
#
# 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.
#
"""
*Good Thermal Strategy*:
Towards to software defined infrastructure, the power and thermal
intelligences is being adopted to optimize workload, which can help
improve efficiency, reduce power, as well as to improve datacenter PUE
and lower down operation cost in data center.
Outlet (Exhaust Air) Temperature is one of the important thermal
telemetries to measure thermal/workload status of server.
"""
import datetime
from oslo_log import log
from watcher._i18n import _
from watcher.common import exception as wexc
from watcher.datasource import ceilometer as ceil
from watcher.datasource import gnocchi as gnoc
from watcher.decision_engine.model import element
from watcher.decision_engine.strategy.strategies import base
LOG = log.getLogger(__name__)
[docs]class OutletTempControl(base.ThermalOptimizationBaseStrategy):
"""[PoC] Outlet temperature control using live migration
*Description*
It is a migration strategy based on the outlet temperature of compute
hosts. It generates solutions to move a workload whenever a server's
outlet temperature is higher than the specified threshold.
*Requirements*
* Hardware: All computer hosts should support IPMI and PTAS technology
* Software: Ceilometer component ceilometer-agent-ipmi running
in each compute host, and Ceilometer API can report such telemetry
``hardware.ipmi.node.outlet_temperature`` successfully.
* You must have at least 2 physical compute hosts to run this strategy.
*Limitations*
- This is a proof of concept that is not meant to be used in production
- We cannot forecast how many servers should be migrated. This is the
reason why we only plan a single virtual machine migration at a time.
So it's better to use this algorithm with `CONTINUOUS` audits.
- It assume that live migrations are possible
*Spec URL*
https://github.com/openstack/watcher-specs/blob/master/specs/mitaka/approved/outlet-temperature-based-strategy.rst
""" # noqa
# The meter to report outlet temperature in ceilometer
MIGRATION = "migrate"
METRIC_NAMES = dict(
ceilometer=dict(
host_outlet_temp='hardware.ipmi.node.outlet_temperature'),
gnocchi=dict(
host_outlet_temp='hardware.ipmi.node.outlet_temperature'),
)
def __init__(self, config, osc=None):
"""Outlet temperature control using live migration
:param config: A mapping containing the configuration of this strategy
:type config: dict
:param osc: an OpenStackClients object, defaults to None
:type osc: :py:class:`~.OpenStackClients` instance, optional
"""
super(OutletTempControl, self).__init__(config, osc)
self._ceilometer = None
self._gnocchi = None
@classmethod
[docs] def get_name(cls):
return "outlet_temperature"
@classmethod
[docs] def get_display_name(cls):
return _("Outlet temperature based strategy")
@classmethod
[docs] def get_translatable_display_name(cls):
return "Outlet temperature based strategy"
@property
def period(self):
return self.input_parameters.get('period', 30)
@classmethod
[docs] def get_schema(cls):
# Mandatory default setting for each element
return {
"properties": {
"threshold": {
"description": "temperature threshold for migration",
"type": "number",
"default": 35.0
},
"period": {
"description": "The time interval in seconds for "
"getting statistic aggregation",
"type": "number",
"default": 30
},
"granularity": {
"description": "The time between two measures in an "
"aggregated timeseries of a metric.",
"type": "number",
"default": 300
},
},
}
@property
def ceilometer(self):
if self._ceilometer is None:
self._ceilometer = ceil.CeilometerHelper(osc=self.osc)
return self._ceilometer
@ceilometer.setter
def ceilometer(self, c):
self._ceilometer = c
@property
def gnocchi(self):
if self._gnocchi is None:
self._gnocchi = gnoc.GnocchiHelper(osc=self.osc)
return self._gnocchi
@gnocchi.setter
def gnocchi(self, g):
self._gnocchi = g
@property
def granularity(self):
return self.input_parameters.get('granularity', 300)
[docs] def calc_used_resource(self, node):
"""Calculate the used vcpus, memory and disk based on VM flavors"""
instances = self.compute_model.get_node_instances(node)
vcpus_used = 0
memory_mb_used = 0
disk_gb_used = 0
for instance in instances:
vcpus_used += instance.vcpus
memory_mb_used += instance.memory
disk_gb_used += instance.disk
return vcpus_used, memory_mb_used, disk_gb_used
[docs] def group_hosts_by_outlet_temp(self):
"""Group hosts based on outlet temp meters"""
nodes = self.compute_model.get_all_compute_nodes()
size_cluster = len(nodes)
if size_cluster == 0:
raise wexc.ClusterEmpty()
hosts_need_release = []
hosts_target = []
metric_name = self.METRIC_NAMES[
self.config.datasource]['host_outlet_temp']
for node in nodes.values():
resource_id = node.uuid
outlet_temp = None
if self.config.datasource == "ceilometer":
outlet_temp = self.ceilometer.statistic_aggregation(
resource_id=resource_id,
meter_name=metric_name,
period=self.period,
aggregate='avg'
)
elif self.config.datasource == "gnocchi":
stop_time = datetime.datetime.utcnow()
start_time = stop_time - datetime.timedelta(
seconds=int(self.period))
outlet_temp = self.gnocchi.statistic_aggregation(
resource_id=resource_id,
metric=metric_name,
granularity=self.granularity,
start_time=start_time,
stop_time=stop_time,
aggregation='mean'
)
# some hosts may not have outlet temp meters, remove from target
if outlet_temp is None:
LOG.warning("%s: no outlet temp data", resource_id)
continue
LOG.debug("%s: outlet temperature %f" % (resource_id, outlet_temp))
instance_data = {'node': node, 'outlet_temp': outlet_temp}
if outlet_temp >= self.threshold:
# mark the node to release resources
hosts_need_release.append(instance_data)
else:
hosts_target.append(instance_data)
return hosts_need_release, hosts_target
[docs] def choose_instance_to_migrate(self, hosts):
"""Pick up an active instance to migrate from provided hosts"""
for instance_data in hosts:
mig_source_node = instance_data['node']
instances_of_src = self.compute_model.get_node_instances(
mig_source_node)
for instance in instances_of_src:
try:
# select the first active instance to migrate
if (instance.state !=
element.InstanceState.ACTIVE.value):
LOG.info("Instance not active, skipped: %s",
instance.uuid)
continue
return mig_source_node, instance
except wexc.InstanceNotFound as e:
LOG.exception(e)
LOG.info("Instance not found")
return None
[docs] def filter_dest_servers(self, hosts, instance_to_migrate):
"""Only return hosts with sufficient available resources"""
required_cores = instance_to_migrate.vcpus
required_disk = instance_to_migrate.disk
required_memory = instance_to_migrate.memory
# filter nodes without enough resource
dest_servers = []
for instance_data in hosts:
host = instance_data['node']
# available
cores_used, mem_used, disk_used = self.calc_used_resource(host)
cores_available = host.vcpus - cores_used
disk_available = host.disk - disk_used
mem_available = host.memory - mem_used
if cores_available >= required_cores \
and disk_available >= required_disk \
and mem_available >= required_memory:
dest_servers.append(instance_data)
return dest_servers
[docs] def pre_execute(self):
LOG.debug("Initializing Outlet temperature strategy")
if not self.compute_model:
raise wexc.ClusterStateNotDefined()
if self.compute_model.stale:
raise wexc.ClusterStateStale()
LOG.debug(self.compute_model.to_string())
[docs] def do_execute(self):
# the migration plan will be triggered when the outlet temperature
# reaches threshold
self.threshold = self.input_parameters.threshold
LOG.debug("Initializing Outlet temperature strategy with threshold=%d",
self.threshold)
hosts_need_release, hosts_target = self.group_hosts_by_outlet_temp()
if len(hosts_need_release) == 0:
# TODO(zhenzanz): return something right if there's no hot servers
LOG.debug("No hosts require optimization")
return self.solution
if len(hosts_target) == 0:
LOG.warning("No hosts under outlet temp threshold found")
return self.solution
# choose the server with highest outlet t
hosts_need_release = sorted(hosts_need_release,
reverse=True,
key=lambda x: (x["outlet_temp"]))
instance_to_migrate = self.choose_instance_to_migrate(
hosts_need_release)
# calculate the instance's cpu cores,memory,disk needs
if instance_to_migrate is None:
return self.solution
mig_source_node, instance_src = instance_to_migrate
dest_servers = self.filter_dest_servers(hosts_target, instance_src)
# sort the filtered result by outlet temp
# pick up the lowest one as dest server
if len(dest_servers) == 0:
# TODO(zhenzanz): maybe to warn that there's no resource
# for instance.
LOG.info("No proper target host could be found")
return self.solution
dest_servers = sorted(dest_servers, key=lambda x: (x["outlet_temp"]))
# always use the host with lowerest outlet temperature
mig_destination_node = dest_servers[0]['node']
# generate solution to migrate the instance to the dest server,
if self.compute_model.migrate_instance(
instance_src, mig_source_node, mig_destination_node):
parameters = {'migration_type': 'live',
'source_node': mig_source_node.uuid,
'destination_node': mig_destination_node.uuid}
self.solution.add_action(action_type=self.MIGRATION,
resource_id=instance_src.uuid,
input_parameters=parameters)
[docs] def post_execute(self):
self.solution.model = self.compute_model
# TODO(v-francoise): Add the indicators to the solution
LOG.debug(self.compute_model.to_string())