Re: [sami] Trying to figure out where we are

In previous mail discussion, there are questions on why should VM be
migrated. 
And also, in the side meeting at Quebec, people want to learn the use case
from real scenario. 
I guess Yunfei's use case could be an answer to that. 

Here are some of the states that I can see on the devices that need to be
migrated.

1. States on switches:
1.1  DHCP snooping table on ports;
1.2  IGMP snooping table on ports;
1.3  Dynamic ACL which is created by traffic or authentication;

2. States on FW
2.1  TCP Connect States
2.2  Dynamic ACL

3. States on LB
3.1  Connect States.
3.2  Session States.

4. States on IPS/IDS.
4.1 Cumulative data

We can think about two scenarios:
1. VM migrate within the subnet under the same FW/LB/IPS. In this case,
states also migrate under the same FW/LB/IPS etc.
States need to be migrated include 1.1, 1.2, 1.3.

2. VM migrate between two geographic data center sites, but the two sites
are in the same subnet (refer to David Black's email). In this case, states
also migrate between FW/LB/IPS etc
All the above states need to be migrated.

Best Regards
Gu Yingjie

-----邮件原件-----
发件人: sami-bounces@ietf.org [mailto:sami-bounces@ietf.org] 代表 Thomas
Narten
发送时间: 2011年8月24日 乐乐20:23
收件人: zhangyunfei
抄送: Yingjie Gu(yingjie); 'Melinda Shore'; sami@ietf.org
主题: Re: [sami] Trying to figure out where we are

"zhangyunfei" <zhangyunfei@chinamobile.com> writes:

> I can share a use case here from our point of view in data center
> operation(will detail this in the upcoming draft as Yingjie said): We
> are considering running different services (e.g., VoIP and streaming
> services) by VM in the same cluster withinn data center. We know
> different services show different user visiting behaviors. For
> example, in the midnight, there are few VoIP usage while there are
> still large amount of streaming usage( this is just an example, maybe
> not so accurate to fit with the reality). In such case, we'd like
> merge the existing VoIP and streaming services into fewer machines and
> cut down the machines running only few VoIP usage to reduce the power
> consumption.  In this senario, we need to consider how to migrate the
> existing VM states.

How is this any different than standard VM migration as practiced
today? What special requirements result from the above scenario?

The topic of this effort is "state migration". In order to make
progress, we need to be very specific about what kind of state and in
what devices needs to be transferred. And why existing approaches as
practiced today are insufficient.

If folk cannot (in one or two paragraphs) answer the above questions,
I do not think writing an Internet Draft about the above would be a
good use of anyone's time.

Thomas
_______________________________________________
sami mailing list
sami@ietf.org
https://www.ietf.org/mailman/listinfo/sami