IT@Intel Data Center Blog

RE: Intel IT Deploys Virtualization - How we did it!

BillSunderland — Thu, 19 Jun 2008 22:01:06 GMT

With regards to differing views on RAID: Yes, IT technical folks do differ in their views on which types of RAID to use, but they have differing requirements behind their decisions. Here's a good website that explains some of the different use cases for both software and hardware RAID.
http://en.wikipedia.org/wiki/Redundant_array_of_independent_disks

"RAID — which stands for Redundant Array of Inexpensive Drives (as named by the inventor) or Redundant Array of Independent Disks (a name which later developed within the computing industry) — is a technology that employs the simultaneous use of two or more hard disk drives to achieve greater levels of performance, reliability, and/or larger data volume sizes.

The phrase "RAID" is an umbrella term for computer data storage schemes that can divide and replicate data among multiple hard disk drives. RAID's various designs all involve two key design goals: increased data reliability and increased input/output performance. When several physical disks are set up to use RAID technology, they are said to be in a RAID array. This array distributes data across several disks, but the array is seen by the computer user and operating system as one single disk. RAID can be set up to serve several different purposes, the most common of which are outlined below.

Operating system based ("software RAID")
Software implementations are now provided by many operating systems. A software layer sits above the (generally block-based) disk device drivers and provides an abstraction layer between the logical drives (RAIDs) and physical drives. Most common levels are RAID 0 (striping across multiple drives for increased space and performance) and RAID 1 (mirroring two drives), followed by RAID 1+0, RAID 0+1, and RAID 5 (data striping with parity) are supported.

Microsoft's server operating systems support 3 RAID levels; RAID 0, RAID 1, and RAID 5. Some of the Microsoft desktop operating systems support RAID such as Windows XP Professional which supports RAID level 0 in addition to spanning multiple disks but only if using dynamic disks and volumes.

Apple's Mac OS X Server supports RAID 0, RAID 1, and RAID 1+0.[2]

The software must run on a host server attached to storage, and server's processor must dedicate processing time to run the RAID software. This is negligible for RAID 0 and RAID 1, but may be significant for more complex parity-based schemes. Furthermore all the busses between the processor and the disk controller must carry the extra data required by RAID which may cause congestion.

Another concern with operating system-based RAID is the boot process, it can be difficult or impossible to set up the boot process such that it can failover to another drive if the usual boot drive fails and therefore such systems can require manual intervention to make the machine bootable again after a failure. Finally operating system-based RAID usually uses formats specific to the operating system in question so it cannot generally be used for partitions that are shared between operating systems as part of a multi-boot setup.

Most operating system-based implementations allow RAIDs to be created from partitions rather than entire physical drives. For instance, an administrator could divide an odd number of disks into two partitions per disk, mirror partitions across disks and stripe a volume across the mirrored partitions to emulate a RAID 1E configuration. Using partitions in this way also allows mixing reliability levels on the same set of disks. For example, one could have a very robust RAID-1 partition for important files, and a less robust RAID-5 or RAID-0 partition for less important data. (Some high-end hardware controllers offer similar features, e.g. Intel Matrix RAID.) Using two partitions on the same drive in the same RAID is, however, dangerous. If, for example, a RAID 5 array is composed of four drives 250 + 250 + 250 + 500 GB, with the 500-GB drive split into two 250 GB partitions, a failure of this drive will remove two partitions from the array, causing all of the data held on it to be lost.

edit Hardware-based
Hardware RAID controllers use different, proprietary disk layouts, so it is not usually possible to span controllers from different manufacturers. They do not require processor resources, the BIOS can boot from them, and tighter integration with the device driver may offer better error handling.

A hardware implementation of RAID requires at least a special-purpose RAID controller. On a desktop system this may be a PCI expansion card, PCI-Express Expansion Card or built into the motherboard. Controllers supporting most types of drive may be used - IDE/ATA, SATA, SCSI, SSA, Fibre Channel, sometimes even a combination. The controller and disks may be in a stand-alone disk enclosure, rather than inside a computer. The enclosure may be directly attached to a computer, or connected via SAN. The controller hardware handles the management of the drives, and performs any parity calculations required by the chosen RAID level.

Most hardware implementations provide a read/write cache which, depending on the I/O workload, will improve performance. In most systems write cache may be non-volatile (i.e. battery-protected), so pending writes are not lost on a power failure.

Hardware implementations provide guaranteed performance, add no overhead to the local CPU complex and can support many operating systems, as the controller simply presents a logical disk to the operating system.

Hardware implementations also typically support hot swapping, allowing failed drives to be replaced while the system is running."

With regards to core segmentation, i'm not sure I follow the intent of the question. If you can describe some more, perhaps I can help answer.

Thanks,
Bill Sunderland

RE: Intel IT Deploys Virtualization - How we did it!

Brently Davis — Fri, 16 May 2008 20:50:33 GMT

I don't think you're out of place. Hopefully someone here can answer that question for you. I can't, however, I'm in agreement with you that we're not speaking the same speak.

Can anybody at Intel help William with his question? Bill?

RE: Before you throw it away....make sure you know where it is going.

Ilene — Tue, 29 Apr 2008 19:18:56 GMT

Laurie, thank you for your question. We do try to put in place processes to ensure equipment does not end up in the landfill whether here or in other areas. Generally when we donate, we recommend that the recipient use a reputable recycler. In some cases the contract includes a recycler. Having said that, the party receiving the donation can use different options

RE: Before you throw it away....make sure you know where it is going.

Laurie Buczek — Mon, 28 Apr 2008 18:55:38 GMT

Great topic Ilene. I recently read an article that talked about the resale market. Much of the U.S. resale equipment goes overseas. However, a lot of the resales don't get utilized in their original state (i.e. a Server is a server). The PCs/Servers get ripped apart and the metals (copper, ore) gets mined and the rest end up in a landfill. How does Intel ensure that the resale market doesn't transfer U.S. "treasures" overseas just to create another's "trash"?

RE: Is the Data Center Green or is it Efficient?

K_Lloyd — Wed, 12 Mar 2008 18:50:11 GMT

I am a big believer in doing the right thing environmentally, and this includes what we do in the data center. The Green label has been pretty wounded by the massive use and calling almost everything green. http://en.wikipedia.org/wiki/Greenwash To that extent I prefer the term efficient. If the data center is "as efficient as it can be" I am reducing power, cooling, costs, people, etc. The reality is that this efficient data center can be a nasty roaring furnace, and still be very efficient - the key is to maximize your return on the space and equipment. Like a bus, which uses lots of energy, iff you fill it up it is way more efficient than the best "Prius*" on the road. Getting the best efficiency from your data center means squeezing everything you can from every square foot and every watt. I wrote a few entries ( http://communities.intel.com/openport/blogs/server/2008/01/29/almost-free-data-center-capacity )entries on how to use Intel bits to squeze the most from your data center and will be posting some videos on the topic later this month.

RE: The Role of Abstraction in Data Center Transformation

Tom Mant — Tue, 11 Mar 2008 20:33:41 GMT

Alan - interesting dilemma that we have been trying to address for years now. Although I've never really understood why a developer would want to have access to the data center - as a developer for 20+ years, the last thing I ever wanted to do was have access to the hardware - just give me a protected (ah... here is the real issue) environment where my applications can run with enough resources to satisfy my customers and I'm happy. I don't care where it is as long as it meets my performance and reliability concerns.

So the real issue has always been that the windows environment (web or otherwise) has never been secure/protected enough to run with "stacked" applications. Most developers have been okay with being dependent on themselves or their teams to write "well behaved" applications but reluctant (to say the least) to have to depend on others to do so. Thus the "keep your hands off of my server" mentality that is prevelant today (at many companies).

Can't say that I blame them given past history around system reliability and well behaved applications.

However now with the new virtualization technologies it is becoming a more viable option to have multiple applications (from multiple teams which is the real key) running on the same physical device - as long as they are in their own "protected" virtual machine. This is a reasonable near term alternative until we have a much better operating system that provides this type of logical partitioning within a single environment. Although I'm not sure that is really necessary. There is another thought...

Good topic by the way

RE: More on our Data Center Efficiency initiative

mp3knol@gmail.com — Sun, 17 Feb 2008 20:53:27 GMT

This is Interesting Video Clip, thanks travis.

RE: Is the Data Center Green or is it Efficient?

yytzhaki — Sun, 17 Feb 2008 08:38:07 GMT

There is logic in the distiguish between 'Green' and Efficient' regarding data centers, but I still think that they have a certain 'relationship' and the term 'green' may still be used.

In order to make it clearer and not further confuse people, I suggest we bring in another term, which is 'Environmently-friendly'.
As I see it, this term is a comparable one.

What it means is that if a data center is designed to consume less power (by better airflow design, using servers with the newest energy-efficient CPUs etc..) and we utilize the generated heat from the data center to heat the building in the winter, this data center is 'enviroment-friendly'. Is it also 'green'? Again, I think that if you look at it compared to other data centers, you may refer to it as 'green'.
I believe that some industry forums are now in the process of defining specs for a 'Green Data Center', so we will probably hear more on this in the near future.

RE: More on our Data Center Efficiency initiative

dsasser — Sat, 02 Feb 2008 09:16:24 GMT

Blogging - is it hobby or work?

RE: Virtualization and the “Datacenter Operating System” of the Future

RK_Hiremane — Thu, 29 Nov 2007 20:13:21 GMT

I think to many extent it is true that virtualization could be the datacenter OS of the future. With the reduction in software overheads with virtualization over time, virtualization harware assists, increasing capability, and being able to take advantage of multiple cores and flexible infrastructure, there is a good potential.