-
Notifications
You must be signed in to change notification settings - Fork 7
xp load ded
Matthieu Simonin edited this page Oct 7, 2016
·
1 revision
- physical nodes
- 1 control, 1 network, 20 computes, 1 util, 1 conductor-node, 1 haproxy-node, 1 memcached-node, 1 rabbitmq-node, 1 mariadb-node
- control
- neutron_server, nova_scheduler, nova_novncproxy, nova_consoleauth, nova_api, glance_api, glance_registry, keystone, cron, kolla_toolbox, heka, cadvisor, docker_registry, collectd
- network
- neutron_metadata_agent, neutron_l3_agent, neutron_dhcp_agent, neutron_openvswitch_agent, neutron_openvswitch_agent, openvswitch_db, keepalived, cron, kolla_toolbox, heka, cadvisor, collectd
- compute
- nova_ssh, nova_libvirt, nova_compute_fake_1, …, nova_compute_fake_#fake, openvswitch_db, openvswitch_vswitchd, neutron_openvswitch_agent, neutron_openvswitch_agent_fake_1, …, neutron_openvswitch_agent_fake_#fake, cron, kolla_toolbox, heka, cadvisor, , collectd
- util
- cadvisor, grafana, influx (rally)
- conductor-node
- nova_conductor, cadvisor, collectd, heka
- haproxy-node
- haproxy, cadvisor, collectd, heka
- memcached-node
- memcached, cadvisor, collectd, heka
- rabbitmq-node
- rabbitmq, cadvisor, collectd, heka
- mariadb-node
- mariadb, cadvisor, collectd, heka
First, find the name of the host machine.
cd results
vagrant up load-ded
for #compute in {100, 200, 500, 1000}
deploy fresh openstack
boot_and_delete concurrency=50 times=2*#compute
wait 5 min
boot_and_list concurrency=50 times=2*#compute
wait 5 min
gather result
More briefly :
./kolla-g5k benchs vanilla-boot-delete-then-boot-list.txt --times xxx --concurrency yyy --wait 300
> cat vanilla-boot-delete-then-boot-list.txt
disco-rally-boot-and-delete.json
disco-rally-boot-and-list.json