[openstack-dev] 答复: 答复: 答复: [neutron] Neutron scaling datapoints?
Hi Vilobh, Thanks a lot for the info. I’ll refine previous spec and propose a new one soon. Actually the AgentGroup code, as well as the db-base AgentGroup driver is almost done. It works well when testing in my OpenStack setup. Will also post the code, which may help understand why AgentGroup and tooz is useful for neutron scalability and performance. Hope to get more comments from neutron team. Thanks. Best, Robin 发件人: Vilobh Meshram [mailto:vilobhmeshram.openst...@gmail.com] 发送时间: 2015年4月15日 2:50 收件人: OpenStack Development Mailing List (not for usage questions) 主题: Re: [openstack-dev] 答复: 答复: [neutron] Neutron scaling datapoints? Hi Robin, The idea sounds good to me too. I am working on refactoring ServiceGroup code. Tooz has a nice compatibility matrix which can be found here [2] which you might find useful. -Vilobh [1] Servicegroup code refactoring : https://review.openstack.org/#/c/172502/ [2] Tooz compatibility matrix : http://docs.openstack.org/developer/tooz/compatibility.html On Tue, Apr 14, 2015 at 6:07 AM, Wangbibo wangb...@huawei.commailto:wangb...@huawei.com wrote: Hi Kevin and Joshua, Thanks for the review. Glad to see that oslo puts distributed coordination into its scope now. Per out of date info [1] (oslo doesn’t do it, while each project should do it separately ), specific backend (zk/memcached) manipulating is included in spec[2], as nova ServiceGroup did. Now we have tooz, then that part should be moved out of AgentGroup and let tooz take it over. Neutron AgentGroup spec needs an update, as what nova ServiceGroup refactor is doing. [3] Per spec[3], tooz doesn’t intend to eliminate or replace ServiceGroup completely. They are integrated and co-work to provide nova ServiceGroup functionalities. That may answer the question from Kevin and Kyle, about relationship between AgentGroup and tooz. Let’s jump into [3][4]: 1) Service Group still exists; 2) Add Tooz driver for ServiceGroup, to take over zk/redis/… backend; 3) Db-based ServiceGroup driver is retained. Db driver was introduced for backward compatibility (with db-based liveness monitoring which existed for a long time before ServiceGroup was added). Since this driver uses tables and a data model that is intrinsically tied to the internal of nova, tooz cannot take it over. 4) Zk/memcached ServiceGroup drivers are temporarily retained, but will be deprecated in future; 5) Eventually, there would be two ServiceGroup drivers: db driver tooz driver; Actually, things are the same for neutron, except that we don’t need to consider zk/memcached driver deprecation. I would like to refine current spec and propose a ”Agent Group and using tooz” spec, following the outlines above. What do you think, Kevin and Joshua? Thanks. ☺ Best, Robin [1] https://wiki.openstack.org/wiki/NovaZooKeeperHeartbeat [2] https://review.openstack.org/#/c/168921/ [3] https://review.openstack.org/#/c/138607/11/specs/liberty/approved/service-group-using-tooz.rst [4] ServiceGroup refactor code: https://review.openstack.org/#/c/172502/ 发件人: Wangbibo [mailto:wangb...@huawei.commailto:wangb...@huawei.com] 发送时间: 2015年4月13日 16:52 收件人: OpenStack Development Mailing List (not for usage questions) 主题: [openstack-dev] 答复: [neutron] Neutron scaling datapoints? Hi Kevin, Totally agree with you that heartbeat from each agent is something that we cannot eliminate currently. Agent status depends on it, and further scheduler and HA depends on agent status. I proposed a Liberty spec for introducing open framework/pluggable agent status drivers.[1][2] It allows us to use some other 3rd party backend to monitor agent status, such as zookeeper, memcached. Meanwhile, it guarantees backward compatibility so that users could still use db-based status monitoring mechanism as their default choice. Base on that, we may do further optimization on issues Attila and you mentioned. Thanks. [1] BP - https://blueprints.launchpad.net/neutron/+spec/agent-group-and-status-drivers [2] Liberty Spec proposed - https://review.openstack.org/#/c/168921/ Best, Robin 发件人: Kevin Benton [mailto:blak...@gmail.com] 发送时间: 2015年4月11日 12:35 收件人: OpenStack Development Mailing List (not for usage questions) 主题: Re: [openstack-dev] [neutron] Neutron scaling datapoints? Which periodic updates did you have in mind to eliminate? One of the few remaining ones I can think of is sync_routers but it would be great if you can enumerate the ones you observed because eliminating overhead in agents is something I've been working on as well. One of the most common is the heartbeat from each agent. However, I don't think we can't eliminate them because they are used to determine if the agents are still alive for scheduling purposes. Did you have something else in mind to determine if an agent is alive? On Fri, Apr 10, 2015 at 2:18 AM, Attila Fazekas afaze...@redhat.commailto:afaze...@redhat.com wrote: I'm 99.9% sure
[openstack-dev] 答复: 答复: [neutron] Neutron scaling datapoints?
Hi Kevin and Joshua, Thanks for the review. Glad to see that oslo puts distributed coordination into its scope now. Per out of date info [1] (oslo doesn’t do it, while each project should do it separately ), specific backend (zk/memcached) manipulating is included in spec[2], as nova ServiceGroup did. Now we have tooz, then that part should be moved out of AgentGroup and let tooz take it over. Neutron AgentGroup spec needs an update, as what nova ServiceGroup refactor is doing. [3] Per spec[3], tooz doesn’t intend to eliminate or replace ServiceGroup completely. They are integrated and co-work to provide nova ServiceGroup functionalities. That may answer the question from Kevin and Kyle, about relationship between AgentGroup and tooz. Let’s jump into [3][4]: 1) Service Group still exists; 2) Add Tooz driver for ServiceGroup, to take over zk/redis/… backend; 3) Db-based ServiceGroup driver is retained. Db driver was introduced for backward compatibility (with db-based liveness monitoring which existed for a long time before ServiceGroup was added). Since this driver uses tables and a data model that is intrinsically tied to the internal of nova, tooz cannot take it over. 4) Zk/memcached ServiceGroup drivers are temporarily retained, but will be deprecated in future; 5) Eventually, there would be two ServiceGroup drivers: db driver tooz driver; Actually, things are the same for neutron, except that we don’t need to consider zk/memcached driver deprecation. I would like to refine current spec and propose a ”Agent Group and using tooz” spec, following the outlines above. What do you think, Kevin and Joshua? Thanks. ☺ Best, Robin [1] https://wiki.openstack.org/wiki/NovaZooKeeperHeartbeat [2] https://review.openstack.org/#/c/168921/ [3] https://review.openstack.org/#/c/138607/11/specs/liberty/approved/service-group-using-tooz.rst [4] ServiceGroup refactor code: https://review.openstack.org/#/c/172502/ 发件人: Wangbibo [mailto:wangb...@huawei.com] 发送时间: 2015年4月13日 16:52 收件人: OpenStack Development Mailing List (not for usage questions) 主题: [openstack-dev] 答复: [neutron] Neutron scaling datapoints? Hi Kevin, Totally agree with you that heartbeat from each agent is something that we cannot eliminate currently. Agent status depends on it, and further scheduler and HA depends on agent status. I proposed a Liberty spec for introducing open framework/pluggable agent status drivers.[1][2] It allows us to use some other 3rd party backend to monitor agent status, such as zookeeper, memcached. Meanwhile, it guarantees backward compatibility so that users could still use db-based status monitoring mechanism as their default choice. Base on that, we may do further optimization on issues Attila and you mentioned. Thanks. [1] BP - https://blueprints.launchpad.net/neutron/+spec/agent-group-and-status-drivers [2] Liberty Spec proposed - https://review.openstack.org/#/c/168921/ Best, Robin 发件人: Kevin Benton [mailto:blak...@gmail.com] 发送时间: 2015年4月11日 12:35 收件人: OpenStack Development Mailing List (not for usage questions) 主题: Re: [openstack-dev] [neutron] Neutron scaling datapoints? Which periodic updates did you have in mind to eliminate? One of the few remaining ones I can think of is sync_routers but it would be great if you can enumerate the ones you observed because eliminating overhead in agents is something I've been working on as well. One of the most common is the heartbeat from each agent. However, I don't think we can't eliminate them because they are used to determine if the agents are still alive for scheduling purposes. Did you have something else in mind to determine if an agent is alive? On Fri, Apr 10, 2015 at 2:18 AM, Attila Fazekas afaze...@redhat.commailto:afaze...@redhat.com wrote: I'm 99.9% sure, for scaling above 100k managed node, we do not really need to split the openstack to multiple smaller openstack, or use significant number of extra controller machine. The problem is openstack using the right tools SQL/AMQP/(zk), but in a wrong way. For example.: Periodic updates can be avoided almost in all cases The new data can be pushed to the agent just when it needed. The agent can know when the AMQP connection become unreliable (queue or connection loose), and needs to do full sync. https://bugs.launchpad.net/neutron/+bug/1438159 Also the agents when gets some notification, they start asking for details via the AMQP - SQL. Why they do not know it already or get it with the notification ? - Original Message - From: Neil Jerram neil.jer...@metaswitch.commailto:neil.jer...@metaswitch.com To: OpenStack Development Mailing List (not for usage questions) openstack-dev@lists.openstack.orgmailto:openstack-dev@lists.openstack.org Sent: Thursday, April 9, 2015 5:01:45 PM Subject: Re: [openstack-dev] [neutron] Neutron scaling datapoints? Hi Joe, Many thanks for your reply! On 09/04/15 03:34, joehuang
[openstack-dev] 答复: [neutron] Neutron scaling datapoints?
Hi Kevin, Totally agree with you that heartbeat from each agent is something that we cannot eliminate currently. Agent status depends on it, and further scheduler and HA depends on agent status. I proposed a Liberty spec for introducing open framework/pluggable agent status drivers.[1][2] It allows us to use some other 3rd party backend to monitor agent status, such as zookeeper, memcached. Meanwhile, it guarantees backward compatibility so that users could still use db-based status monitoring mechanism as their default choice. Base on that, we may do further optimization on issues Attila and you mentioned. Thanks. [1] BP - https://blueprints.launchpad.net/neutron/+spec/agent-group-and-status-drivers [2] Liberty Spec proposed - https://review.openstack.org/#/c/168921/ Best, Robin 发件人: Kevin Benton [mailto:blak...@gmail.com] 发送时间: 2015年4月11日 12:35 收件人: OpenStack Development Mailing List (not for usage questions) 主题: Re: [openstack-dev] [neutron] Neutron scaling datapoints? Which periodic updates did you have in mind to eliminate? One of the few remaining ones I can think of is sync_routers but it would be great if you can enumerate the ones you observed because eliminating overhead in agents is something I've been working on as well. One of the most common is the heartbeat from each agent. However, I don't think we can't eliminate them because they are used to determine if the agents are still alive for scheduling purposes. Did you have something else in mind to determine if an agent is alive? On Fri, Apr 10, 2015 at 2:18 AM, Attila Fazekas afaze...@redhat.commailto:afaze...@redhat.com wrote: I'm 99.9% sure, for scaling above 100k managed node, we do not really need to split the openstack to multiple smaller openstack, or use significant number of extra controller machine. The problem is openstack using the right tools SQL/AMQP/(zk), but in a wrong way. For example.: Periodic updates can be avoided almost in all cases The new data can be pushed to the agent just when it needed. The agent can know when the AMQP connection become unreliable (queue or connection loose), and needs to do full sync. https://bugs.launchpad.net/neutron/+bug/1438159 Also the agents when gets some notification, they start asking for details via the AMQP - SQL. Why they do not know it already or get it with the notification ? - Original Message - From: Neil Jerram neil.jer...@metaswitch.commailto:neil.jer...@metaswitch.com To: OpenStack Development Mailing List (not for usage questions) openstack-dev@lists.openstack.orgmailto:openstack-dev@lists.openstack.org Sent: Thursday, April 9, 2015 5:01:45 PM Subject: Re: [openstack-dev] [neutron] Neutron scaling datapoints? Hi Joe, Many thanks for your reply! On 09/04/15 03:34, joehuang wrote: Hi, Neil, From theoretic, Neutron is like a broadcast domain, for example, enforcement of DVR and security group has to touch each regarding host where there is VM of this project resides. Even using SDN controller, the touch to regarding host is inevitable. If there are plenty of physical hosts, for example, 10k, inside one Neutron, it's very hard to overcome the broadcast storm issue under concurrent operation, that's the bottleneck for scalability of Neutron. I think I understand that in general terms - but can you be more specific about the broadcast storm? Is there one particular message exchange that involves broadcasting? Is it only from the server to agents, or are there 'broadcasts' in other directions as well? (I presume you are talking about control plane messages here, i.e. between Neutron components. Is that right? Obviously there can also be broadcast storm problems in the data plane - but I don't think that's what you are talking about here.) We need layered architecture in Neutron to solve the broadcast domain bottleneck of scalability. The test report from OpenStack cascading shows that through layered architecture Neutron cascading, Neutron can supports up to million level ports and 100k level physical hosts. You can find the report here: http://www.slideshare.net/JoeHuang7/test-report-for-open-stack-cascading-solution-to-support-1-million-v-ms-in-100-data-centers Many thanks, I will take a look at this. Neutron cascading also brings extra benefit: One cascading Neutron can have many cascaded Neutrons, and different cascaded Neutron can leverage different SDN controller, maybe one is ODL, the other one is OpenContrail. Cascading Neutron--- / \ --cascaded Neutron-- --cascaded Neutron- | | -ODL-- OpenContrail And furthermore, if using Neutron cascading in multiple data centers, the DCI controller (Data center inter-connection controller) can also be used under cascading Neutron, to provide NaaS ( network as a service ) across data
