lvs-users
|
To: | "LinuxVirtualServer.org users mailing list." <lvs-users@xxxxxxxxxxxxxxxxxxxxxx> |
---|---|
Subject: | Re: Is active-active LVS really capable of doubling theoretical bandwidth? |
From: | malcolm <lists@xxxxxxxxxxxxxxxx> |
Date: | Fri, 03 Nov 2006 08:42:13 +0000 |
Michael Spiegle wrote: I did some reading on active-active and while I think its a great idea, I didn't quite understand how it was capable of doubling the throughput as mentioned in the PDF document. Here's how I see it: Which PDF is that? You could just use one Active/Passive LVS with 802.3ad bonding over 2 * GIG NICs?Lets say you have an active-active LVS pair. Each individual LVS has a gigabit fiber card in it. Since an active-active configuration should be able to utilize both cards at the same time, you have 2gbit of THEORETICAL bandwidth (1 + 1 = 2). Or Active/Active with two separate VIPs.Deliberately avoiding actually answering your question because I don't know anything about it :-). -- Regards, Malcolm Turnbull. |
<Prev in Thread] | Current Thread | [Next in Thread> |
---|---|---|
|
Previous by Date: | Re: FIN_WAIT with direct routing., Andreas Lundqvist |
---|---|
Next by Date: | RE: First-time LVS user, having problem with LVS-DR, David Fix |
Previous by Thread: | Is active-active LVS really capable of doubling theoretical bandwidth?, Michael Spiegle |
Next by Thread: | Re: Is active-active LVS really capable of doubling theoretical bandwidth?, Joseph Mack NA3T |
Indexes: | [Date] [Thread] [Top] [All Lists] |