1. Cluster by ip , e.g a cluster with pc placed in different locations and
each location does a different task based on its ip (the task has not been
specified )
sure, I've certainly used IP to provide a useful mnemonic for the "role"
of hosts within a cluster. but I'm worried about your use of
Nifty Tom Mitchell wrote:
On Sun, Jan 25, 2009 at 08:55:41AM -0600, Gerry Creager wrote:
.
In other words, it's not the switch. I tend to "think out loud" to
expose all possible failure modes, a process I learned at NASA/Johnson
when I worked on Space Station's Medical Operations. In ma
Scott Atchley wrote:
On Jan 25, 2009, at 10:13 AM, Gerry Creager wrote:
-bash-3.2# ethtool -K rx off
no offload settings changed
You missed the interface here. You should try:
-bash-3.2# ethtool -K eth1 rx off
Yeah. Too much morning. Not enough coffee. Back to it.
gerry
-bash-3.2# etht
Joe Landman wrote:
Gerry Creager wrote:
Joe Landman wrote:
I wonder if the switch could be implicated. We have seen some
(cheap) GbE switches not support (in practice) jumbo frames
(irrespective of literature).
Been there, done that. HP claims to be able to handle packets up to
9000 bytes
On Sun, Jan 25, 2009 at 08:55:41AM -0600, Gerry Creager wrote:
.
>
> In other words, it's not the switch. I tend to "think out loud" to
> expose all possible failure modes, a process I learned at NASA/Johnson
> when I worked on Space Station's Medical Operations. In manned
> spaceflight
Answers inline.
Joshua
-- Original Message --
Received: 12:49 AM CST, 01/23/2009
From: amjad ali
To: Beowulf Mailing List
Subject: [Beowulf] programming guidence request
> Hello All,
> I am developing my parallel CFD code on a small cluster. My system has
> openmpi installed based on g
On Thu, 22 Jan 2009 22:40:25 -0800
Greg Lindahl wrote:
> On Fri, Jan 23, 2009 at 11:03:46AM +0500, amjad ali wrote:
>
> > (1) Which debugger would be easy and effective to use for above?
>
> print *,
>
openmpi spawns several processes so you options are pretty much print or
pausing you progra
Hi ,
I am wondering if any of the following has been done :
1. Cluster by ip , e.g a cluster with pc placed in different locations and
each location does a different task based on its ip (the task has not been
specified )
2. Mobile clusters : any one by any small chance looked into clusters over
I remember having read that "enterprise" drives are built to run nonstop (
24x7 ), whereas "consumer" disks are expected to be used briefly (2-3hrs?)
and then powered down daily.
So, the differences could be firmware, cooling & airflow, bearing
types (fluid vs ball) etc.
Regards
Anand
On Thu,
On Jan 25, 2009, at 10:13 AM, Gerry Creager wrote:
-bash-3.2# ethtool -K rx off
no offload settings changed
You missed the interface here. You should try:
-bash-3.2# ethtool -K eth1 rx off
-bash-3.2# ethtool -k eth1
Offload parameters for eth1:
rx-checksumming: on
tx-checksumming: on
scatte
Gerry Creager wrote:
Joe Landman wrote:
I wonder if the switch could be implicated. We have seen some (cheap)
GbE switches not support (in practice) jumbo frames (irrespective of
literature).
Been there, done that. HP claims to be able to handle packets up to
9000 bytes of payload. (9122 t
Joe Landman wrote:
I wonder if the switch could be implicated. We have seen some (cheap)
GbE switches not support (in practice) jumbo frames (irrespective of
literature).
Been there, done that. HP claims to be able to handle packets up to
9000 bytes of payload. (9122 total, IIRC)
Nifty T
Nifty Tom Mitchell wrote:
On Sat, Jan 24, 2009 at 09:36:09AM -0600, Gerry Creager wrote:
Couple of follow-up notes.
MTU=4500: Had one node fall over with the same overflow errors.
MTU=3000: A WRF model is running, but single timesteps are executing
2.5x slower than MTU=1500
I'll go snag t
We've run some forensics with a real testset. It's not the HP Procurve
switch.
We've also seen good jumbo results with some of the managed Linksys
48-port gigabit switches.
In other words, it's not the switch. I tend to "think out loud" to
expose all possible failure modes, a process I lea
14 matches
Mail list logo
