Bladewatch.com

An interesting article talking about how IBM and the Syracuse University are working on building one of the worlds’ most energy efficient data centers – very cool, it’s great to highlight the concepts and discuss, the best practice and the range of technologies or processes used to achieve this. Do check it out, they’re using a range of solutions to achieve this.

Looking at how we power the data center, looking at the different elements of the infrastructure, the server, the storage, the network as well as the data center in terms of air flow (checking that we have effective and efficient air flow to where it’s needed) as well as effective power distribution using efficient power supplies and possibly DC power? It will all depend on where you are in the data center life cycle.

It’s a multi feed activity, by that I mean, we should be looking at the infrastructure, the application and the usage patterns:

• Data center design – power/air flow and environment -  what operating temperature should we run at? Could we run it hotter and use less power for cooling
• Server/storage/network – can we use more efficient hardware, using more efficient power supplies, lower voltage processors, solid state or the smaller more efficient drives/SAN boot
• Infrastructure configuration – can we power down what we don’t need, can we remove those organizational screensavers to reduce idle cpu utilization, can we configure the infrastructure more effectively – scale down the network when it’s not needed, remove redundant power supplies from servers that don’t need them. This could include drivers/firmware and best practice configurations or settings being applied.
• Application configuration – can we configure the application to exploit more of the processor functionality to gain performance gains, or could we use graphical processors? Could we better code or optimize the code for the platform?
• Infrastructure utilization – can we use technologies like Citrix, virtualization and grid to create shared infrastructure pools which are more heavily utilized, a citrix farm hosting applications rather than separate servers per application, server virtualization to achieve more cpu and memory utilization. Grid to create a shared compute model where we only consume what we need to use.
• Storage efficiency – do we need to keep all the data we have online right now on our disk shelves – can we use data deduplication technologies, offline or nearline storage where we archive off data and improve data utilization? Can we not do more to reduce physical storage consumption?

Share and Enjoy

Sun have released an updated version of OpenSolaris, which includes numerous improvements in the networks, storage, virtualization and performance fields. It’s always great to see further innovation of the OpenSolaris platform particularly with the announcement of further enterprise support creating further possibilities for end user choice in platforms.

The specific innovations are in the press release quoted below – or check out the video:

Major Enhancements Continue in Storage

OpenSolaris 2009.06 provides dozens of enhancements to the breakthrough technology of ZFS and encompasses it with a complete architecture of connectivity and protocol support. New, fully integrated flash storage support in ZFS helps to optimize large scale pools of very high performance storage by designating flash devices as write accelerators and read accelerators. These pools are automatically managed by ZFS to achieve extreme levels of performance across many workloads, making the need for small caches on RAID controllers obsolete.

Native support for Microsoft CIFS has been added as a full peer to NFS, as a high performance kernel with integrated features and support for Microsoft Windows semantics for security, naming and access rights, allowing transparent use and sharing of files across Windows, Linux and Solaris environments. To round out the complete storage capability, Sun has designed new, very high performance support for iSCSI and Fiberchannel block protocols into the Solaris kernel, allowing systems running OpenSolaris to participate as a client and a target for virtually any storage topology.

All of these storage features are integrated into the Solaris platform and take full advantage of its core functionality including: fault management, networking, multi-threaded scaling, performance, security and resource management capabilities. More information on the storage enhancements in OpenSolaris is available at: http://opensolaris.com/learn.

Comprehensive Approach to Virtualization

With this announcement, Sun continues to deliver on a holistic, built-in virtualization design for networking, storage and application abstraction, raising performance and scale to new highs for the industry. Building on the advances in networking storage virtualization, the OpenSolaris platform delivers key server virtualization technologies in the form of Solaris Containers, Logical Domains (LDoms) for Sun CMT systems and the Xen-based hypervisor to give users a complete virtualization platform built directly into the OpenSolaris OS. One of the most widely deployed virtualization technologies in the world, Solaris Containers provide lightweight, agile, software-defined boundaries that can be used to create virtual servers for consolidating hundreds of existing enterprise-class workloads onto a single system. More information on the built-in virtualization capabilities of OpenSolaris is available at: http://opensolaris.com/learn.

OpenSolaris Performance

OpenSolaris is fully supported on thousands of systems, including all shipping Sun x86 and SPARC(R) systems, delivering world-record performance and reliability, with the highest levels of service and support. Since it was first launched in May 2008, OpenSolaris has set 24 world records and industry benchmarks clearly demonstrating its versatility. The trend continues for OpenSolaris 2009.06. By leveraging the advantages of the built-in innovation and enhancements, OpenSolaris 2009.06 delivers 35 percent better memory management, 22 percent better integer arithmetics and 18 percent better multi-thread scheduler management when compared to the latest Linux releases.(1)

For detailed information and a complete list of all benchmarks on Sun systems visit: http://www.sun.com/benchmarks and http://blogs.sun/weber for more information on OpenSolaris performance.

Share and Enjoy

Check out this article which is talking about how the Japanese Aersospace Exploration Agency is using a NetApp FAS6080 storage solution as part of its infrastructure to improve the performance and I/O, in the article it notes a doubling of the I/O on the new system. It’s always great to read how people are using the technology and what benefits they are managing to realize.

Share and Enjoy

There have been a number of posts/articles talking about a new iPhone in a few days time (8th of June), how cool would that be? I wonder if this might be the device that encourages me to switch? It would be wonderful if it was released on more networks though, I still love being on Vodafone, we’ll have to wait and see.

Share and Enjoy

Check out this article talking about IBM offering solid state drives for its server and storage offerings, bringing solid state drives should bring improved energy efficiency, availability and performance. It’s interesting to see where IBM see solid state drives fitting into their storage offerings, and at the same time how they can continue to make the provisioning and management of customer storage between the different forms of storage we have in the enterprise, an interesting read, do check it out.

Share and Enjoy

Super Micro’s innovation of their storage chassis is great news, anything we can do to make storage more energy efficient, more accessible and scalable for the end user has to be a good thing for the platform and the industry. Improving the energy efficiency of the power supply, the storage chassis is the first step, we need to look at the way we manage and scale or tier the storage. Can we use data de-duplication? Do we need a data strategy?  Do we need the volume of data we have available and online all the time? Can we look at technologies where we put legacy data on less immediate or expensive storage, where can make it available on demand, but not store is unnecessarily. In the meantime, the more the vendors can do with software and hardware to make it more compatible, energy efficient, accessible and affordable, the better for the end user community.

Share and Enjoy

Blade.org are having a Virtual Technology Symposium on the 11th of June (next week), it is meant to cover key trends in the data center in a vendor neutral format. I look forward to attending and seeing what the trends and topics are, virtualization is set to be one, I wonder if there will be anything about Green IT, coupled with 10GB Ethernet or Fibre Channel over Ethernet? We’ll have to see, for more information go to the blade.org site or check out this press release.

Share and Enjoy

Recognizing the energy footprint of the IT not only in terms of ‘Green IT’, but in terms of your operating costs remains an important issue. Looking at the pc estate might be the first step, simply switching the pc off overnight can reduce your energy bill as well as your support costs, fewer patch update failures, fewer pcs being out of memory or unstable. An interesting read, illustrating come examples of power utilization, do chcek it out:

Engineering News

Information Technology (IT) was said to account for 2% of global energy demand, and because a single Personal Computer (PC) running continuously could use in excess of 1 000 kWh/y, South African business could cut costs and associated carbon-dioxide emissions by implementing efficiencies.

Of that 2% of global energy used to power IT, 39% was attributed to PCs and monitors, 23% servers – including cooling, 15% fixed-line telecoms, 9% mobile telecoms, 7% land and office telecom, and 6% printers.

While significant time and resources are dedicated to accelerating ‘green computing’, one of the most effective solutions is to switch off a PC when not in use.

Studies have shown that there are many reasons why people don’t do this, ranging from security, to laziness.

Share and Enjoy

Check out this article talking about Apple reported to be looking at building a data center, I wonder if this is to power further online services as part of their mobile me service, backup/online storage? Even web based Apple desktop hosting? Could I virtualize my mac and carry it around the web with me? We’ll have to wait and see, in the meantime, I’m off to read up more.

Tuaw

As Mike mentioned in the news roundup yesterday, word is going around that the corporate overlords at Apple are planning to build a massive server farm in the state of North Carolina within the next decade or so. How do we know? Because the company is working on getting a few nice tax breaks to go along with the deal, eventually worth about $46 million if Apple fulfills their end of the agreement. Their end of the agreement being a huge investment of a whopping $1 billion (that’s over 333,000 Xserves, if that is indeed what they’re installing — and we bet not) into an economy that could probably use it. What exactly will go in there? We aren’t sure — lawmakers say the investment is worth it, and obviously Apple isn’t showing their cards.

[Note: the $1B investment figure is not an upfront number, but rather a total investment over the first 9 years of operation. –Ed.]

But the facility, once up and running, will start by employing at least 100 people, so that’s a nice start. Google has apparently gone through the same deal with North Carolina recently, though their deal was a little smaller: $600 million investment for a server farm that opened last year.

Share and Enjoy

Check out this article talking about a study illustrating the performance benefits of their LightPulse 8GB adapters when used with Windows Server 2008 Hyper-V, the performance results from your fibre card will be dependent on the driver, the server and operating system configuration as well as the SAN storage and switch. Regardless, check the article below:

Emulex

Study Shows Emulex 8Gb/s HBAs Consistently Deliver Better CPU Effectiveness in Windows Server 2008 Hyper-V Virtualized Server Environments

COSTA MESA, Calif., May 26, 2009 -

* Emulex Corporation (NYSE:ELX) today announced that its LightPulse 8Gb/s Fibre Channel Host Bus Adapters (HBAs) consistently deliver better CPU effectiveness in Windows Server 2008 Hyper-V virtualized server environments over comparable solutions from its key competitors.

* Overall, Emulex LightPulse LPe12002 8Gb/s HBAs showed 37 to 78 percent higher CPU effectiveness when compared to dual-channel 8Gb/s HBAs from its nearest competitors, QLogic and Brocade, in an evaluation report from Demartek, a firm specializing in independent analysis and lab validation testing of server, storage and security technologies.

* Greater CPU efficiency enables higher consolidation ratios and provides greater cost savings in virtualized server deployments.

* Emulex 8Gb/s HBAs deliver improved enterprise scalability and flexibility through an advanced driver, unique dynamic interrupt coalescing and support for Message-Signaled Interrupts eXtended (MSI-X), offering end users greater CPU effectiveness, reduced system latency and improved I/O throughput – all of which contribute to increased server virtualization performance and reduced power consumption.

Share and Enjoy