[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-ha-dev
Subject:    Re: [Linux-ha-dev] [PATCH] New heartbeat
From:       Alan Robertson <alanr () unix ! sh>
Date:       2004-03-25 14:47:13
Message-ID: 4062F0F1.7000806 () unix ! sh
[Download RAW message or body]

Zhu, Yi wrote:
> Alan Robertson wrote:
> 
> 
>>It *is* faster.  Let me give an example...
>>
>>If you send out one request for each of 100 machines, and
>>each takes 100ms to complete, then this will take about 10
>>seconds to complete - because you perform them all serially...
>>
>>If you send out the broadcast request, and every client sends
>>their answer at the same time, then they all complete in about 100ms
>>total time. (Actually, it's probably a little slower than that because
>>now it's bounded above by the single worst case reply time,
>>not by 100 * the average time - but it's still MUCH faster than doing
>>it serially). 
>>
>>So, it is much more scaleable.
> 
> 
> OK. Agreed.
> 
> 
>>But, as you point out, it is more complex.
>>
>>I can see that having both the synchronous single-query case
>>and the asynchronous cluster-wide query would be useful.
>>
>>They could probably both use the same messages types, though
>>and be handled the same way on the server side though...
> 
> 
> They are now in the same type and handled in the same way on the
> server side. Do you think I can check in the patch now or after 1.2.1
> released?

Now is fine.


-- 
     Alan Robertson <alanr@unix.sh>

"Openness is the foundation and preservative of friendship...  Let me claim 
from you at all times your undisguised opinions." - William Wilberforce

_______________________________________________________
Linux-HA-Dev: Linux-HA-Dev@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
[prev in list] [next in list] [prev in thread] [next in thread]