Intel Communities : All Content - Server Solutions Insider

Ask a "Nehalem" Expert

webadmin@intel.com — Sat, 28 Nov 2009 02:49:22 GMT

On the new Nehalem based boxes we are seeing a discrepancy reported from the OS on the installed memory. I have used (12) 4GB dimms and (6) 8GB dimms and both show the exact same amount of phyical memory avaible. The issue is that the OS is only reporting that I have 47.144GB? Any help would be greatly appreciated.

Webinar - Dell, Red Hat, and Intel help you plan migration from SPARC/Solaris

michihiro.koyama@intel.com — Wed, 25 Nov 2009 19:16:44 GMT

It is not late to register for a joint webinar delivered by Red Hat, Intel, and Dell.

This webinar delivers how to plan and execute your migration from SPARC/Solaris to Dell/Intel/Red Hat, a popular content also delivered in seminars the companies did in North America in October.

A clear roadmap showing the estimated timeframe and costs for your migration
Effective training for you and your IT staff
Proven best practices to ensure a smooth implementation
Join Red Hat, Intel, and Dell on December 2 2009 at 2pm ET to see how these open source pioneers can help you move from a RISC/UNIX environment to Red Hat Enterprise Linux.

HI

webadmin@intel.com — Fri, 20 Nov 2009 00:18:12 GMT

RHEL 4 and 5 are supported. See http://www.intel.com/support/motherboards/server/S5000VSA/sb/CS-022698.htm

If Fedora is similar enough, just follow the instructions for RHEL at http://www.intel.com/support/motherboards/server/sb/CS-022358.htm

S5000 VSA Board RAID 5 Activation Key Problem

webadmin@intel.com — Sat, 07 Nov 2009 19:00:18 GMT

Hello Alex,

I am glad to hear that you managed to solve your problem with the activation key.

Anyway, always happy to help wherever needed.

All the best,

Kind Regards,

Aryan.

Customers on the move - Solaris to RHEL migration guide

michihiro.koyama@intel.com — Thu, 29 Oct 2009 22:25:40 GMT

Just completed 8 city roadshow with Dell and Red Hat, talking to customers face to face over lunch about how a migration should be planned and done from UNIX/RISC to RHEL/Intel/Dell.

I had opportunities to attend two of the seminars in different cities, meeting few different customers. Customers came to our seminar to look for a template to quickly develop a plan that could also be approved quickly and get the project going to counter the enormous management pressure they get to do more with less.

Most of the customers I had conversations with were from medium size business.

One customer clearly said, it is no brainer, there is no reason to advocate to purchase new AIX/Power or Solaris/SPARC machines any more.

The other customer said, much of the knowledge accumulated to manage UNIX servers can immediately applied for migration activities while he saw the needs to get the UNIX administrators trained on RHEL through the training programs offered by Red Hat.

Consistently, there were the nuance that customers needed to act now, for a short term result.

For those who needs to get going on planning, Red Hat has put together a very good migration guide. The document helps you start thinking about and resourcing your migration. Register, to get started. Also, a lighter version to get an idea.

Please contact RHEL, Dell, or Intel representative for the next steps, and if you don’t have such resource, let me know. I’ll help dispatch one to you.

SE7520AF2 server mother board system fault

webadmin@intel.com — Fri, 23 Oct 2009 21:33:54 GMT

Older boards sometimes continued to report an error condition, even after the condition was cleared. Use the SEL viewer

http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=7470&lang=eng

to view the SEL, and see if there's any other issues reported. If not, save a local copy of the SEL, then clear it. Do an AC power cycle, and see if that clears the fan issue.

IBM x3550 M2, x3650 M2, HS22, and dx360

phillip.mcfee@blackhillscorp.com — Wed, 21 Oct 2009 15:40:13 GMT

Just got done getting the HS22 to Boot From SAN (SVC). We recently received a HS22 blade (7870-AC1) and tried to configure it to Boot From SAN. After several attempts of trying to configure the Qlogic card (QMI2572) a support call was placed to IBM since the Qlogic card was unable to save any settings. Another Qlogic card was also ordered to rule out a defective card and it had the same issue. I wouldn't even recommend trying to Boot From SAN without these firmware levels.

Firmware levels should be at:
BIOS 1.04, Build P9E130AUS
Diagnostics 1.13, Build P9YT40A
Blade Sys Mgmt Processor 1.05
Qlogic QMI2572, BIOS Revision 2.08
IMM (Integrated Management Module) 1.05, Build YU0032F

Overall we were able to configure the HS22 to Boot From SAN but were only able to do so by loading an OS (internal drives) and applying the updates since Update Express didn't have the latest firmware for the Qlogic card. The time to deploy a blade has increased. The HS21 took less than 30 minutes to have up and runnning when booting from the SAN compared to the 2 1/2 hrs to deploy the HS22. Please share this email as needed.IBM x3550 M2, x3650 M2, HS22, and dx360

Message was edited by: William Lea, adjusted to smaller font and removed Bold.

S5520SC IASC no CPU/IOH/MEM temperature

webadmin@intel.com — Fri, 16 Oct 2009 16:51:36 GMT

You'd have to open the SDR to be sure; but yes, it's possible that when you chose "Other" for the chassis types, those values don't have any listed Lower Critical or Lower Noncritical levels.

X25-E SSD ... new "trim" ata command ?

webadmin@intel.com — Fri, 16 Oct 2009 07:18:25 GMT

I have same question. Will be update for SSD X25-E?

Will be implemented TRIM with free specification (or like in standard ATA8-ACS2) for better implementation in JFFS2, BRTFS, NILFS and ext4?

Thanks for your answer.

Ch-ch-ch-ch-Changes!

eoin.j.mcconnell@intel.com — Fri, 09 Oct 2009 21:40:38 GMT

It has been a little while since I shared some thoughts about moving from RISC due to a 3 month assignment managing the Nehalem-EX product line. One word describes that product, ‘wow’, and the change it will bring to the IT marketplace as we know it. But I’m not here to talk about that….

David Bowie was certainly introspective when he wrote his “Changes” song about his need to constantly look at oneself, previous decisions and the need for frequent reinvention and change. The sentiments reflected within “Changes” can be applied to all aspects of life both personal and business. Reflecting on previous business decisions and looking for newer and better ways to do things should not be seen as the previous decision being wrong, but rather should be rewarded as looking to change and do things better based upon the environment today.

Previous business decisions to deploy your IT solutions on RISC based architectures was most likely the right decision at the time based on the business need, the solutions availability and the architecture available to run that solution. Some of these solutions are likely due for an upgrade due to changing business needs, a better version of the application now being available or have become to expensive to maintain and support on older server system. It is time to make a change and change is likely to include upgrading to next generation of software solution and choosing a new server system that will perform and work with the software solution.

With the rapid pace of technology innovation and evolution over the last number of years, the decision is not necessarily as clear cut as it may have been in the past.

What I wanted to share with you was some information around how Intel’s Xeon microprocessor has evolved and can now compete with the POWER architecture offered from IBM. Some of you may say that this is not possible, but Xeon 5500 is getting some attention as shown with information posted on IBM website

Price/performance is a key consideration for database workloads and $ / tpmC is pretty widely accepted as a good rule of thumb. Its good to see that the IBM System x3950M2 (based on the Xeon 7400 processor) has a $1.99 $ / tpmC compared to IBM Power P570 $3.54 $ / tpmC;-)

Xeon 5500 has a performance per socket leadership against a similar class POWER 6 2S system. This can be seen by comparing results at www.spec.orgfor benchmarks such as SPECJbb2005, SPECint-rate 2006 etc. IBM makes reference to performance per core leadership over the Xeon 5500. A fair statement, but most customers look at overall system level performance to do the require task. I guess my key takeaway is that if you are looking for a solution to run infrastructure type workloads and get the best bang for your buck then the Xeon 5500 delivers best price/performance

There is also some interesting discussion around scalability of Xeon Vs POWER6. Xeon 5500 is used in 2 socket configurations, and not in scalable systems. So it seems to be a little like comparing apples to oranges!. Scalable Xeon platforms are available in the market today from both IBM and Unisys. There are also 15+ designs for scalable platforms from 8 OEMs coming with Intel’s next generation scalable Xeon product, Nehalem-EX. Some good material was shared recently at Intel’s Developer Forum in San Francisco. Look for Mission Critical Server Deployment class at https://intel.wingateweb.com/us09/scheduler/catalog/catalog.jspThis material provides a good overview about how Nehalem-EX provides supportfor high-end computing with a scalable micro architecture, advanced RAS capabilities and how Redhat will support Nehalem-EX scalability. This presentation also shows an example of the innovation of NEC who are developing mission critical Linux solutions based on NEC’s Scalable architecture using Intel Xeon processors.

So is it the time for you to change?. Are existing options like the Xeon 5500 or the Xeon 7400 the right choice for you? Nehalem-EX is coming and I believe will bring a huge change to the marketplace as we know it today.

What do you think?