Bladewatch.com

I got an email from a colleague who was planning the upgrade on their Dell PowerEdge R710 which was running ESX, there had been some debate about deciding on how much memory to buy and working out the optimum memory configuration in order to match the channels as they were upgrading from 64GB to 128GB. Anyway he said that he found this very helpful, do check it out, it’s their memory upgrade manual and is very informative.  As the document says if you upgrade the memory but do not balance the memory across the channels you might receive a bios error message, the server will still work and boot but you will get the message every time you restart.

The document is here it’s a pdf for download.

Share and Enjoy

http://www.tilera.com/products/processors/TILE-Gx_Family

This latest generation processor family features devices with 16 to 100 identical processor cores (tiles) interconnected with Tilera’s iMesh™ on-chip network. Each tile consists of a complete, full-featured processor as well as L1 and L2 cache and a non-blocking switch that connects the tiles into the mesh. As with all Tilera® processors, each tile can independently run a full operating system, or multiple tiles taken together can run a multiprocessing OS like SMP Linux.

The TILE-Gx family processor slashes board real estate requirements and system costs by integrating a complete set of memory and I/O controllers, eliminating the need for an external north bridge or south bridge. TileDirect™ technology provides coherent I/O directly into the tile caches to deliver ultimate low-latency packet processing performance. Tilera’s DDC™ (Dynamic Distributed Cache) system for fully coherent cache across the tile array enables scalable performance for threaded and shared memory applications.

This is something I will need to read up about, the 100 core processor possibility sounds outstanding, what I need to further understand is the processor compatibility with existing platforms and therefore the ability to run it in parallel environments, a mixed current and legacy operating environment. For cloud and new build outs, with the application code ported and optimized, systems based on these processors (like those of the Power and Cell/B.E) could be a compelling offering.

Share and Enjoy

I read this blog article about what to carry on a USB stick in order to be able to manage your blade or rack servers and it got me thinking about something I mentioned to both the guys at Dell and HP the last time I met them either at a conference or at one of their blog events. The concept was just a directory structure per model containing the drivers and firmware for a server with a pre-defined baseline, so version 1.0 might be May 2011, which contains everything up to that date, and version 1.1 being June 2011, allowing me to be one month or so behind to prevent any issues from recently released updates and get everything up to a pre-determined baseline, understanding that I might have a mixed new and legacy estate.

For the HP:

• HP Proliant Support Pack 8.60 for x86 and x64 (Windows 2003 x86 and x64)
• HP Proliant Support Pack 7.90 for Windows 2000
• HP firmware – I’ve simply downloaded the online firmware for the servers that I look after and named the firmware after the server name, so DL380 G5 for example
• HP SmartStart ISO so that we can boot the server into the SmartStart cd and run the diagnostics or systems configuration
• HP Firmware CD – again so that we can do the complete system firmware – that is the disk, array controller, integrated lights out, system and network

For the Dell:

• Dell firmware for the servers that I look after – backplane, SCSI/SAS controllers, system, network and DRAC or iDRAC
• Dell drivers for the servers that we look after – Windows 2003 x86 and x64 + Windows 2008
• Dell DSET utility in remote and thick mode so that we can run and submit diagnostics
• Dell OpenManage Administrator tool and agents pack

For the IBM

• IBM drivers for the servers that we look after – Windows 2003 x86 and x64 + Windows 2008
• IBM firmware for the servers we look after – backplane, SCSI/SAS, system, network and RSA
• IBM Update Express ISO and DVD – a new one was released about a week ago
• IBM Dynamic System Analysis – their diagnostic tool both the offline and online or portable edition
• IBM Director agents for hardware monitoring through their IBM Director tool

Other tools/utilities

• Windows 2003 Service Pack 2 – useful if you need to re-service pack a server
• Winternals pack – http://technet.microsoft.com/en-us/library/ee460919.aspx
• Drac/ILO/RSA configuration scripts and utilities

Going forward the tools as well as the vendor utilities will naturally migrate or start to include ESXi tools and utilities as well as those for Hyper-V as they become more mainstream in my working space.

If I miss anything or you have suggestions, do let me know.  I would publish the bundle on the bladewatch site as a free download but I found the associated legal issues involved complex and not wishing to upset anyone or cause emotional trauma, I’ve listed them instead. As ever, they are just a guide and are in no way instructional and if in doubt ask your service provider or vendor for further information and best practice.

Share and Enjoy

http://www.hp.com/hpinfo/newsroom/press/2011/110621xa.html

Maximize employee productivity with collaboration

HP provides SMBs with a broad range of technologies that reduce costs, improve efficiency and ensure collaboration for greater employee productivity. New solutions enable clients to:

• Increase productivity with the HP ProLiant ML110 G7, an entry-level server that is simple to deploy and manage. It supports basic office applications such as web messaging, small databases, file and print, as well as small vertical applications.
• Scale performance with the HP ProLiant DL120 G7 server, a powerful, entry-level rack-optimized server running a wide range of dedicated applications, including IT infrastructure applications, web, messaging, file and print operations, small internet applications and shared web access.
• Simplify deployment and management with the HP V1810-48G web-managed switch. This 48-port switch easily integrates into existing multivendor networks and accommodates increasing performance requirements as businesses grow and develop.
• Reduce energy costs with the easy-to-install, standards-based HP V1410 unmanaged Fast Ethernet switch series. The first IEEE Energy Efficiency Ethernet compliant switches are operational out of the box to offer SMBs an affordable entry-level networking solution.
• Increase performance with the HP P2000 G3 Modular Smart Array (MSA). Among the industry’s first entry-level storage solutions to support the VMware API for Array Integration and VMware vCenter, the HP P2000 MSA allows SMBs to get enterprise-class performance and VMware manageability. In addition, HP Insight Control Storage Module Manager for VMware vCenter enables management and monitoring of server, storage and networking resources for virtual machines within the vCenter console.

These portfolio of announcements furthers HP offerings in the SMB space, not only in the server space but also in the storage and networking space to offer complete solutions from a range of their server, storage and networking businesses. It’s great to see their continued innovation of their rack and tower servers as well as their storage offering in their Modular Smart Array platform, do check it out.

Share and Enjoy

http://www.theaustralian.com.au/australian-it/it-business/nations-health-in-good-hands/story-e6frganx-1226078137767

BEFORE the Nursing Board of Victoria (NBV) became the Australian Health Practitioner Regulation Agency (AHPRA) in July last year, its Melbourne data centre was nationally recognised for its sophistication.

AHPRA is now the organisation responsible for the registration and accreditation of 500,000 health professionals in Australia. It replaced all the previously independent state boards separated by professions.

As NBV chief information officer, Michael Hoffman undertook a major virtualisation and consolidation project to deliver greater reliability and uptime.

Check out this article talking about how this agency has used virtualization to improve reliability and adaptability of its platforms, it’s always great to read how businesses and end users are benefiting from the platform, in this case VMware vSphere 4.

Share and Enjoy

http://www.businesscloud9.com/content/can-tilera-take-intel-ultimate-cloud-computing-processor/5694

Perhaps the biggest challenge yet to the dominance of the Intel x/86 / x64 general purpose processor is announced this week by US processor designer, Tilera.

The company is introducing what it claims, perhaps with a tad over-zealous grandiosity, is the ultimate Cloud Computing processor. The TILE Gx 3000 family, to give it its more prosaic title, is based around a 64-bit core, and comes in 36, 64 and 100-core versions. Many will feel that its biggest claim to fame, however, is the company’s expectation that it will consume 80% lower Wattage than Intel’s just-announced Sandy Bridge Xeon, coupled with a claimed 10x performance per Watt advantage over the Intel device.

Check out this interesting post about the Tilera processor, I confess that it is not something I am familiar with but it does sound cool, and I wonder if this could be an alternative to the x86 offering. I note with interest the core count and that it works with Linux, I wonder if the challenge with it could be porting your applications, going forward though as cloud becomes more mainstream this might not be as big a challenge to all concerned.

Share and Enjoy

http://www.marketwatch.com/story/ibm-launches-smart-cloud-for-disaster-recovery-and-data-archive-2011-06-20

ARMONK, N.Y., June 20, 2011 /PRNewswire via COMTEX/ — IBM IBM -0.06% today announced new cloud services that enable clients to protect, store and retrieve their most critical data in minutes versus days in the event of a disaster.

Disasters – whether from hurricanes, earthquakes or IT failures – can cause significant damage to businesses in the loss of crucial data and revenue. Analyst estimates state that the loss of revenue during downtime is tens of thousands of dollars for every hour.

Businesses need to assess threat risks and create a resilience strategy that helps ensure uninterrupted business operations and enables speed and agility when an unforeseen event strikes. The new IBM SmartCloud Resilience services allow customers to protect their data and applications faster, cheaper and in a more flexible manner than they could before within a traditional data center environment.

I can certainly see the demand for cloud services in the disaster recovery space, being able to have a mixed environment where I can fail over to a private or public cloud solution which allows me to offer similar or near production ready environments to continue operations could be an important development in the cloud space. The ability to keep the lights on, to offer service full or functional can be the difference between keeping customers, making revenue and not, anything the cloud providers can do to help this has to be a good thing. I suppose the challenge will be data security combined with confidence in the platform and integration with existing platforms, how ready as an organization am I to transfer my platforms to a cloud type system and how does that translate to my regulatory requirements as well as the practicalities of my infrastructure and application design.  Combining this with a simple understanding of in a disaster recovery situation what services do we need to keep the lights on, email, file and print, active directory and the internal or external applications – which of these can be cloud based, what level of cloud (private/public) and how do we synchronize content, or replicate functionality between environments, does the infrastructure simply switch from internal hosting to infrastructure and application or IT down a wire?

Share and Enjoy

http://www.oracle.com/us/corporate/press/406121

Oracle today announced the Oracle Optimized Solution for Enterprise Cloud Infrastructure, a fully integrated, pre-tested solution which helps accelerate enterprise Infrastructure-as-a-Service (IaaS) cloud deployments.

The new solution leverages Oracle’s unique ability to engineer and integrate key infrastructure components and speed virtual machine deployment times by up to 98 percent, lowering TCO and improving productivity for enterprises and cloud service providers.

The Oracle Optimized Solution for Enterprise Cloud Infrastructure combines Oracle’s Sun Blade server modules, Oracle Solaris or Oracle Linux, Oracle’s ZFS Storage Appliance, and Oracle VM, the only VM fully certified and supported with Oracle Database, Oracle Fusion Middleware and Oracle Applications. With Oracle Enterprise Manager 11g the solution can be managed and supported as a single system.

It will be interesting to see how customers perceive and respond to this announcement from Oracle, certainly being able to buy in or consume the services from Oracle’s own cloud solution, using a combination of software and platform as a service through a pay on use type model could improve accessibility to new customers to their platforms and solutions, whilst allowing existing customers to gain the extra capacity to meet their business needs whether to meet their disaster recovery, development or virtualization and storage projects. Exciting times are ahead, I’m off to check out the Oracle site for more information.

Share and Enjoy

http://www.fujitsu.com/global/news/pr/archives/month/2011/20110620-02.html

Tokyo, June 20, 2011 — RIKEN and Fujitsu have taken first place on the 37th TOP500 list announced today at the 26th International Supercomputing Conference (ISC’11) held in Hamburg, Germany. This ranking is based on a performance measurement of the “K computer⁽¹⁾,” currently under their joint development.

The TOP500-ranked K computer system, currently in the configuration stage, has 672 computer racks equipped with a current total of 68,544 CPUs. This half-build system achieved the world’s best LINPACK⁽²⁾ benchmark performance of 8.162 petaflops (quadrillion floating-point operations per second), to place it at the head of the TOP500 list. In addition, the system has recorded high standards with a computing efficiency ratio of 93.0%. This is the first time since June 2004 that the Japanese supercomputer “Earth Simulator” has been ranked first on the TOP500 list.

Well done to Fujitsu, it’s always interesting to read what range of technologies and software they have used to achieve these impressive results for high performance computing. Do check out the article for more information.

Share and Enjoy

Dell iDRAC firmware version 1.70

Dell have released an updated version of their iDRAC firmware bringing it to 1.70 now. I have tested it on their R610 and R710 servers without issue, due check it out, it brings some feature enhancements and bug fixes according to their readme or release notes below:

Improved Web GUI performance.
* Improved vFlash performance.
* Out of Band Hardware and Firmware Inventory support.
* Console launch without SSO.
* Work notes.
* Improved System FRU data (Added Product Name, serial number, Asset tag,and Version in Product info Area of FRU).
* Virtual MAC support.
* Identify in Command Line Interface (CLI).
* OEM SEL Event filtering.
* Plain English SEL events.
* Default web certificate update through Command Line Interface.
* Display Life Cycle Controller firmware version display on CLI and GUI.
* Support for platforms PowerEdge(TM) R415, and EqualLogic(TM) DX6000G.
* Support for operating system Redhat Enterprise Linux 6 x64.

The same update may apply to their other rack and blade servers, do check their site for more information.

Broadcom NetXtreme II Family of Adapters

They have also released update drivers for their family of NetXtreme II network adaptors here, do check them out, the release notes mentions improvements to teaming. If you’re upgrading the drivers do be sure to check the firmware here.

Share and Enjoy