[openib-general] Re: SA cache design

2006-01-11 Thread Michael S. Tsirkin
Quoting Sean Hefty [EMAIL PROTECTED]: I can envision how the sa_cache could eventually build towards a distributed SA. I've been thinking about distributed SA myself for a while now. This seems like an elegant way to improve the SA scalability. How would distributed copies sync with the

Re: [openib-general] Re: SA cache design

2006-01-11 Thread Sean Hefty
Michael S. Tsirkin wrote: I've been thinking about distributed SA myself for a while now. This seems like an elegant way to improve the SA scalability. How would distributed copies sync with the central SA? I guess we could simply use sockets on top of IPoIB - the sync it probably a rare

[openib-general] Re: SA cache design

2006-01-10 Thread Michael S. Tsirkin
Quoting Sean Hefty [EMAIL PROTECTED]: To keep the design as flexible as possible, my plan is to implement the cache in userspace. The interface to the cache would be via MADs. Clients would send their queries to the sa_cache instead of the SA itself. The format of the MADs would be

[openib-general] Re: SA cache design

2006-01-10 Thread Sean Hefty
Michael S. Tsirkin wrote: To keep the design as flexible as possible, my plan is to implement the cache in userspace. The interface to the cache would be via MADs. Clients would send their queries to the sa_cache instead of the SA itself. The format of the MADs would be essentially

[openib-general] Re: SA cache design

2006-01-10 Thread Michael S. Tsirkin
Quoting r. Sean Hefty [EMAIL PROTECTED]: Subject: Re: SA cache design Michael S. Tsirkin wrote: To keep the design as flexible as possible, my plan is to implement the cache in userspace. The interface to the cache would be via MADs. Clients would send their queries to the sa_cache