Comments on: CEE the future http://blog.nigelpoulton.com/cee-the-future/#utm_source=feed&utm_medium=feed&utm_campaign=feed with nigel poulton Tue, 20 Dec 2011 19:37:54 +0000 http://wordpress.org/?v=2.8.6 hourly 1 By: Wooch http://blog.nigelpoulton.com/cee-the-future/comment-page-1/#comment-706 Wooch Fri, 29 Jan 2010 01:35:01 +0000 http://blogs.rupturedmonkey.com/?p=547#comment-706 Nigel, interesting stuff as always. But your advocacy for CEE raises a question: if CEE is the wave of the future, might it become a replacement for PCIe? It's fast, easily managed, relatively cheap, etc... How about a PCIe vs CEE knock-down here? Which is really better? Nigel, interesting stuff as always. But your advocacy for CEE raises a question: if CEE is the wave of the future, might it become a replacement for PCIe?
It's fast, easily managed, relatively cheap, etc…

How about a PCIe vs CEE knock-down here? Which is really better?

]]> By: stephen2615 http://blog.nigelpoulton.com/cee-the-future/comment-page-1/#comment-559 stephen2615 Thu, 05 Nov 2009 06:03:17 +0000 http://blogs.rupturedmonkey.com/?p=547#comment-559 Hi Nigel, Sorry for not getting back but I missed your reply.  We havelots of  C7000 running normal stuff mostly san pass through to edge switches.  Some are running NPIV and some are real core/edge.  Some enclosures have 9124e's.     Our network admin wont consider VC for some reason he refuses to give.  Hmmm.. I could give him something to consider..   HP have made is absolutely clear (so far) that FCoE is not part of the blade infrastructure unless you put a CNA into a blade slot that can take a HBA.   I want our server ops people to come up with an architecture that I can agree on but as per normal, no decision is the best one as you can't make the wrong decision.   Things like virtualised blades in the enclosure get messy with what mezz slots to use for what.  The Mezz slot 2 and 3 are the best (in 680's) but our people insist that 10 GbE is the way forward and I have to use slot 1 due to port mapping.  Blades and idiots can be a bad mix.   Anyway, its all hearsay until someone makes a decison on BFS anyway.  If we go BFS, the SAN will go balistic and I can't even consider the size.  I am getting a bit keen on VC now that they have 24 port 8 Gpbs FC. Hi Nigel,
Sorry for not getting back but I missed your reply.  We havelots of  C7000 running normal stuff mostly san pass through to edge switches.  Some are running NPIV and some are real core/edge.  Some enclosures have 9124e’s. 
 
 Our network admin wont consider VC for some reason he refuses to give.  Hmmm.. I could give him something to consider..
 
HP have made is absolutely clear (so far) that FCoE is not part of the blade infrastructure unless you put a CNA into a blade slot that can take a HBA.   I want our server ops people to come up with an architecture that I can agree on but as per normal, no decision is the best one as you can’t make the wrong decision.
 
Things like virtualised blades in the enclosure get messy with what mezz slots to use for what.  The Mezz slot 2 and 3 are the best (in 680’s) but our people insist that 10 GbE is the way forward and I have to use slot 1 due to port mapping.  Blades and idiots can be a bad mix.
 
Anyway, its all hearsay until someone makes a decison on BFS anyway.  If we go BFS, the SAN will go balistic and I can’t even consider the size.  I am getting a bit keen on VC now that they have 24 port 8 Gpbs FC.

]]> By: blogs.rupturedmonkey.com » Blog Archive http://blog.nigelpoulton.com/cee-the-future/comment-page-1/#comment-558 blogs.rupturedmonkey.com » Blog Archive Mon, 02 Nov 2009 22:46:52 +0000 http://blogs.rupturedmonkey.com/?p=547#comment-558 [...] Cisco price – if the guys at the booth are to be believed.  However, as I mention in a previous post, there is more to value than price.  It may turn out to be very short sighted to jump all over [...] [...] Cisco price – if the guys at the booth are to be believed.  However, as I mention in a previous post, there is more to value than price.  It may turn out to be very short sighted to jump all over [...]

]]> By: Nigel Poulton http://blog.nigelpoulton.com/cee-the-future/comment-page-1/#comment-557 Nigel Poulton Fri, 30 Oct 2009 10:53:47 +0000 http://blogs.rupturedmonkey.com/?p=547#comment-557 stephen<br /><br />what HP configuration do you have in mind?  Is this using VC Flex-10 on c7000 BladeSystem?<br /><br />I know HP appear to be taking the "steady as she gows" approach with FCoE.<br /><br />If you lay out your config and what your gripe is I will pith in with thoughts and dont mind pining people at HP for opinions.. stephen

what HP configuration do you have in mind?  Is this using VC Flex-10 on c7000 BladeSystem?

I know HP appear to be taking the "steady as she gows" approach with FCoE.

If you lay out your config and what your gripe is I will pith in with thoughts and dont mind pining people at HP for opinions..

]]> By: stephen2615 http://blog.nigelpoulton.com/cee-the-future/comment-page-1/#comment-556 stephen2615 Fri, 30 Oct 2009 03:04:26 +0000 http://blogs.rupturedmonkey.com/?p=547#comment-556 I just discovered that HP don't support Boot From SAN with FCoE.  Anyone know when thats likely?  That makes no difference what switch technology you use (Brocade or Cisco). I just discovered that HP don’t support Boot From SAN with FCoE.  Anyone know when thats likely?  That makes no difference what switch technology you use (Brocade or Cisco).

]]> By: blogs.rupturedmonkey.com » Blog Archive » Emulex UCNA at SNW Europe http://blog.nigelpoulton.com/cee-the-future/comment-page-1/#comment-555 blogs.rupturedmonkey.com » Blog Archive » Emulex UCNA at SNW Europe Tue, 27 Oct 2009 22:47:17 +0000 http://blogs.rupturedmonkey.com/?p=547#comment-555 [...] blogs.rupturedmonkey.com The greatest storage blog in the world…. « CEE the future [...] [...] blogs.rupturedmonkey.com The greatest storage blog in the world…. « CEE the future [...]

]]> By: Jon Toor http://blog.nigelpoulton.com/cee-the-future/comment-page-1/#comment-554 Jon Toor Mon, 26 Oct 2009 17:57:03 +0000 http://blogs.rupturedmonkey.com/?p=547#comment-554 Nigel, I totally agree with your premise. <em>"FCoE vs whatever"</em> is not the right debate because it does not get at the key questions. As you point out, <strong>"The change in focus needs to be towards smarter and not just faster."</strong> To me, that means discussing the management capabilities of the different solutions, and how they enable smarter, more agile control over the infrastructure. Debating the merits of a protocol is fine, but it doesn't answer fundamental questions such as: Is there a single point of mgmt for I/O across all servers? How easy is it to deploy and manage redundant I/O? Can I quickly migrate I/O to another server if needed? How do I converge connectivity yet maintain physical isolation for the networks that require that? Is QoS simple to establish (and then to maintain as my environment changes)? These questions are simple, but they're the kinds of questions that that impact day-to-day operations. And they get at your point. Solutions need to be smarter, not just faster. Because in the end, "smarter" is what makes it easer to be more efficient. Which after all is the whole point of virtualization, right?     <strong></strong>   Nigel,
I totally agree with your premise.
"FCoE vs whatever" is not the right debate because it does not get at the key questions. As you point out,
"The change in focus needs to be towards smarter and not just faster."
To me, that means discussing the management capabilities of the different solutions, and how they enable smarter, more agile control over the infrastructure.
Debating the merits of a protocol is fine, but it doesn’t answer fundamental questions such as:

Is there a single point of mgmt for I/O across all servers?
How easy is it to deploy and manage redundant I/O?
Can I quickly migrate I/O to another server if needed?
How do I converge connectivity yet maintain physical isolation for the networks that require that?
Is QoS simple to establish (and then to maintain as my environment changes)?

These questions are simple, but they’re the kinds of questions that that impact day-to-day operations.
And they get at your point. Solutions need to be smarter, not just faster. Because in the end, "smarter" is what makes it easer to be more efficient. Which after all is the whole point of virtualization, right?
 
 

 

]]>