Hello John, Thanks for that advice. I would like to convert the network to Infiniband but my budget certainly wouldn't cover 18 Infiniband NICs on top of the Infiniband switch. I thought I might be able to justify splashing out on a fancy GigE switch if I could be sure that users would notice the difference. I like the idea of setting up a LAG between the existing Dell 5424 switches but I don't have enough spare ports at the moment. However I will bear that in mind for the future. -Dan. On 01/27/2012 01:38 PM, John Lauro wrote: > If you are considering as much as ?3500 for a switch, you might want to > consider infiniband qdr instead. We don't currently have it here, but are > considering it. From what I can it has lower latency, can do 40gbps, is > reasonably priced (slightly better than 10gbe, not comaring directly to > gb) . That said, you would also have to budget for cards and cables even > though the switch price by itself is not so bad... > > As you have multiple switches (sounds like 5524 or maybe 5424), the first > next to 0 cost change you should do (if not already done) is setup a LAG > between the switches and run 2-4 cables between switches instead of 1. > > >> -----Original Message----- >> From: gluster-users-bounces at gluster.org [mailto:gluster-users- >> bounces at gluster.org] On Behalf Of Dan Bretherton >> Sent: Friday, January 27, 2012 8:05 AM >> To: gluster-users >> Subject: Switch recommendations >> >> Dear All, >> I need to buy a bigger GigE switch for my GlusterFS cluster and I am >> trying to decide whether or not a much more expensive one would be >> justified. I have limited experience with networking so I don't know if >> it would be appropriate to spend ?500, ?1500 or ?3500 for a 48-port >> switch. Those rough costs are based on a comparison of 3 Dell >> Powerconnect switches: the 5548 (bigger version of what we have now), >> the 6248 and the 7048. The servers in the cluster are nothing special - >> mostly Supermicro with SATA drives and 1GigE network adapters. I can >> only justify spending more than ~?500 if I can be sure that users would >> notice the difference. Some of the users' applications do lots of small >> reads and writes, and they do run much more slowly if all the servers >> are not connected to the same switch, as is the case now while I don't >> have a big enough switch. Any advice or comments would be much >> appreciated. >> >> Regards >> Dan. >> _______________________________________________ >> Gluster-users mailing list >> Gluster-users at gluster.org >> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users