Re: [Beowulf] IBM LCM16 capacity

2013-09-08 Thread John Hearns
John, thst is a good idea. And can use onboard controllers for power up/down also. 500 nodes equals 500 extra bulky cables! Ps. Anyone remember the Sun nodes where the BMC controllers had two dedicated Ethernet ports plus onboard bridge - so you could daisy chain them up a rack? Good idea. But ma

Re: [Beowulf] IBM LCM16 capacity

2013-09-08 Thread John McCulloch
The console switch was a customer request, but we can recommend the management port remote KVM. Thanks, John -Original Message- From: Reuti [mailto:re...@staff.uni-marburg.de] Sent: Sunday, September 08, 2013 6:13 AM To: John McCulloch Cc: 'beowulf@beowulf.org' Subject: Re: [Beowulf] IB

Re: [Beowulf] IBM LCM16 capacity

2013-09-08 Thread Reuti
Am 07.09.2013 um 19:55 schrieb John McCulloch: > Can anyone relate experience with implementing the IBM LCM16 KVM switch in a > tiered configuration? We are spec’ing approximately 500 node x86 cluster. > > Since you can daisy chain up to 16 servers per port, I would expect the > supported max

[Beowulf] [hpc-announce] Call for Participation: Workshop on Heterogeneous Unconventional Cluster Architectures and Applications (HUCAA)

2013-09-08 Thread Holger Fröning
[Apologies if you got multiple copies of this email. If you'd like to opt out of these announcements, information on how to unsubscribe is available at the bottom of this email.] (apologies if you receive multiple HUCAA publicity emails)

[Beowulf] IBM LCM16 capacity

2013-09-08 Thread John McCulloch
Can anyone relate experience with implementing the IBM LCM16 KVM switch in a tiered configuration? We are spec'ing approximately 500 node x86 cluster. Since you can daisy chain up to 16 servers per port, I would expect the supported maximum servers to be greater than 256 with two or three tiere