2013-03-19 21:52:14 |
Joe Gordon |
bug |
|
|
added bug |
2013-03-19 21:53:20 |
Joe Gordon |
description |
sing Devstack (Grizzly-3) and nova SHA1 98f9f735d7e6ac32a8736971f4aab0d81c56131e
euca-describe-instances (POST /services/Cloud/ CloudController:DescribeInstances) is 13 times slower than nova list with 700 VMS.
This is because s3_image_get_by_uuid is called three times per instance and
block_device_mapping_get_all_by_instance,
block_device_mapping_get_all_by_instance,
aggregate_metadata_get_by_host,
get_ec2_instance_id_by_uuid
are called once per hose.
So for 700 VMs that is 4900 DB calls! |
sing Devstack (Grizzly-3) and nova SHA1 98f9f735d7e6ac32a8736971f4aab0d81c56131e
euca-describe-instances (POST /services/Cloud/ CloudController:DescribeInstances) is 13 times slower than nova list with 700 VMS.
This is because s3_image_get_by_uuid is called three times per instance and
block_device_mapping_get_all_by_instance,
block_device_mapping_get_all_by_instance,
aggregate_metadata_get_by_host,
get_ec2_instance_id_by_uuid
are called once per hose.
So for 700 VMs that is 4900 DB calls!
In comparison 'nova list' does 2 DB calls no matter how many VMs there are. |
|
2013-03-19 22:52:36 |
Joe Gordon |
description |
sing Devstack (Grizzly-3) and nova SHA1 98f9f735d7e6ac32a8736971f4aab0d81c56131e
euca-describe-instances (POST /services/Cloud/ CloudController:DescribeInstances) is 13 times slower than nova list with 700 VMS.
This is because s3_image_get_by_uuid is called three times per instance and
block_device_mapping_get_all_by_instance,
block_device_mapping_get_all_by_instance,
aggregate_metadata_get_by_host,
get_ec2_instance_id_by_uuid
are called once per hose.
So for 700 VMs that is 4900 DB calls!
In comparison 'nova list' does 2 DB calls no matter how many VMs there are. |
using Devstack (Grizzly-3) and nova SHA1 98f9f735d7e6ac32a8736971f4aab0d81c56131e
and running as admin user.
euca-describe-instances (POST /services/Cloud/ CloudController:DescribeInstances) is 13 times slower than nova list with 700 VMS.
This is because s3_image_get_by_uuid is called three times per instance and
block_device_mapping_get_all_by_instance,
block_device_mapping_get_all_by_instance,
aggregate_metadata_get_by_host,
get_ec2_instance_id_by_uuid
are called once per hose.
So for 700 VMs that is 4900 DB calls!
In comparison 'nova list' does 2 DB calls no matter how many VMs there are.
When running the same thing as demo user I get the same thing |
|
2013-03-19 23:28:24 |
Vish Ishaya |
tags |
|
grizzly-rc-potential |
|
2013-03-19 23:28:36 |
Vish Ishaya |
nova: importance |
Undecided |
High |
|
2013-03-20 00:59:51 |
OpenStack Infra |
nova: status |
New |
In Progress |
|
2013-03-20 00:59:51 |
OpenStack Infra |
nova: assignee |
|
Vish Ishaya (vishvananda) |
|
2013-03-21 19:08:12 |
OpenStack Infra |
nova: assignee |
Vish Ishaya (vishvananda) |
Joe Gordon (jogo) |
|
2013-03-21 23:42:32 |
OpenStack Infra |
nova: status |
In Progress |
Fix Committed |
|
2013-03-25 10:45:18 |
Thierry Carrez |
nova: importance |
High |
Medium |
|
2013-03-26 21:40:01 |
Vish Ishaya |
nova: milestone |
|
grizzly-rc2 |
|
2013-03-26 23:12:56 |
OpenStack Infra |
nova: status |
Fix Committed |
Fix Released |
|
2013-04-04 11:24:59 |
Thierry Carrez |
nova: milestone |
grizzly-rc2 |
2013.1 |
|