Archive for June, 2008

Green Storage and The Idle Time Secret

June 30, 2008

Recently, Steve Lohr of the New York Times wrote an interesting article titled, “Demand for Data Puts Engineers in Spotlight. A paragraph of particular interest states, “The problem is that most computers in data centers run at 15 percent or less of capacity on average, loafing the rest of the time, though consuming electricity all the while. (In the old days, when they housed a few large computers, data centers were far more efficient. Mainframe computers run at 80 percent of capacity or more).”

This corroborates my almost two decades of experience in disk storage. While many administrators believe their storage systems operate 24×7, most do not, and in fact few storage systems do. At Nexsan’s suggestion, one customer turned on AutoMAID (Nexsan’s proprietary energy-saving technology that reduces disk drive energy when disks are idle) to Level 1, the first of three levels of energy savings. After a short period of time, he was surprised to find that his storage system supporting an important and frequently used application was idle more often than he thought. In fact, he was able to save 20% on energy without ever experiencing an application delay or receiving a complaint from a user, because the disks sprang back to full power without a drop in performance every time an I/O request interrupted their idle state.

Like computers, if storage systems are accessed only 15% of the time, they spin disks 100% of the time. I like car analogies, so here goes another one: this reminds me of sitting at a red light and revving your car engine to high RPMs while you wait for the light to turn green.

No one would do that for long, so why do that with your storage systems?

One of the first steps in reducing storage energy consumption is making storage administrators aware that their systems are idle a lot more than they think. Far too many administrators are wasting large amounts of energy (and, of course, money) because they think their storage systems need to run at full power 24×7. Realizing the secret that storage is idle at times (especially on nights and weekends) is the first step towards significant savings.

AutoMAID versus Old MAID

June 23, 2008

With the recent press activity around MAID, I am frequently asked “What is the difference between traditional MAID and AutoMAID?”  In a phrase, AutoMAID provides the benefits of MAID without the limitations.

Traditional MAID, or old MAID, trades application performance for energy savings by turning off power or putting drives into a deep sleep.  This is an “on/off” approach to energy savings.  While this method certainly saves energy, turning drives completely “off” dramatically reduces application performance.  In fact, when activated, old MAID only spins a maximum of one-fourth of its drives at any given time.  Storage systems that employ old MAID technology can take several minutes to spin up a group of drives, controllers, and other related components before they are ready to transmit and receive data. This is unacceptable for most datacenter applications.

To make a car analogy, when is your car most energy efficient? When it’s parked in your garage.  While this method certainly saves energy it’s not very practical for getting around.

AutoMAID, in contrast to old MAID, is a next-generation technology that reduces energy while removing the application performance penalty.  In other words, you get energy savings and application performance.  AutoMAID significantly reduces energy usage while maintaining the ability to respond to application requests with nearly instantaneous response times.  

Continuing with the car analogy, AutoMAID can be thought of as a car at an intersection.  It saves energy while idling but accelerates quickly when the light turns green. 

While old MAID was exciting in its age, AutoMAID is part of a new era of MAID that provides energy savings without compromise.

Nexsan’s Green History

June 12, 2008

When Nexsan introduced the concept of Enterprise ATA in early 2001, we were thinking primarily about how much money could be saved in the data center by using low-cost desktop hard drives to store Tier 2 data and D2D images.  Soon, however, we started hearing about how ATA storage was saving power.  Raw electricity wasn’t particularly expensive, but once you factor in the costs of UPS protection and redundant air conditioning, the picture starts to change, especially if you are operating near the limits of the existing power and cooling infrastructure.  I was recently told by a data center manager in midtown Manhattan that if he wanted more power, there was a three-year waiting list and he had to pay to rip up the streets.  Not good.

About five years ago, we started to specifically engineer storage for power efficiency, which led to our revolutionary 42-drive ATABeast (and its descendant, the SATABeast).  We basically shared two power supplies and two RAID controllers across what is essentially three shelves-worth of disk drives.  Saves money and increases density of course, but also saves a lot of power.  We continued to improve efficiency in later releases because disk drives were improving in density and efficiency, but also by working on various power optimizations, which led us to develop AutoMAID.

If you aren’t familiar with AutoMAID, it’s the idea that when a drive is not in use, it is gradually spun down (the heads are unloaded first which reduces air resistance in the HDA, then after a further period of inactivity the drives are reduced to half their nominal RPM, and then if it remains idle we put it to sleep).  AutoMAID takes advantage of the fact that typical Fibre Channel command timeouts are in the realm of 60 to 120 seconds, which is plenty of time to pop the drives back into full speed mode if an I/O request comes down the wire. 

The key point here is that the array still acts like any other disk drive or RAID from the perspective of the host O/S, while opportunistically looking for ways to save power if a drive or a bunch of drives are not needed at a given point in time. This is often the case in D2D or deep archive applications or in applications which are not used 24×7. At the same time, the way we’ve executed AutoMAID means not compromise on the ability to return to maximum performance should I/O recommence.

Green storage is important to Nexsan and is included in all of our products at no extra cost.  We’ve been green since 2001 (long before global warming and green messages began permeating our daily lives) and we know how to provide green storage without compromising performance.

The industry is catching on to the Nexsan AutoMAID idea labeling this class of technology MAID 2.0 or next generation MAID.  MAID 2.0 embraces the idea of intelligently idling disks based on application and need while negating the performance trade off.

A Tale of Two Storries

June 9, 2008

On the evening of Thursday 5th June in the prestigious setting of London’s Tower Hotel Nexsan continued its fine tradition – now five in a row – of scooping top honours in the UK Storage Industry Awards, or “Storries” as they are affectionately known in the trade.

An audience comprising of the great, the good and the downright disreputable (they know who they are!) in the storage industry watched appreciatively as the Nexsan SATABeast walked off with the Green Product of the Year and the Green Project of the Year awards.

The Green Project of the Year award was accepted by the Honda Racing F1 Team in conjunction with our Channel partner NCE Computer Group. The Honda Racing F1 Team has consolidated its storage with virtualisation software operating in tandem with the Nexsan SATABeast and AutoMAID.  The energy savings made possible by Nexsan’s SATABeast with AutoMAID fit perfectly with The Honda Racing F1 Team earthdreams programme of environmental projects. 

In bestowing the SATABeast with these awards the industry has given fitting recognition to the tremendous contribution to energy saving being provided by our very own AutoMAID technology which is a key feature of the SATABeast.  Thanks to AutoMAID no other manufacturer can even come close to the SATABeast in providing high performance and energy saving capabilities in the same solution.

Nexsan Delivers Industry’s First Energy Efficient SAS Solution

June 2, 2008

Nexsan released its first SAS storage solution this week called SASBoy™. This is the first time anyone has put out SAS storage with energy efficiency. It’s designed for fixed content (content that typically changes little over time) storage and offers MAID 2.0 functionality through Nexsan’s AutoMAID™ (Automatic Massive Array of Idle Disks) technology. The great thing about the way we’ve done AutoMAID is that it gives you the energy savings without reducing application performance. 

Some have wondered why you need energy efficiency with a high performance drive. The answer is simple – even high performance applications are not working 24×7, but when they need the data, they don’t want to wait. Turning off a SAS drive could take a minute or more to power back up, versus our approach which is to leave the power on but selectively set the drives to different levels of energy consumption based on idle time or application response time. I think the better question is – why wouldn’t you do this? Application performance is not reduced and drive reliability is maintained. Energy costs aren’t coming down and we don’t charge extra for our AutoMAID. This is a no brainer.

One of our customers said it best in an article on SearchStorage:

Joe Funaro, director of technology for IT at Lenox Hill Radiology, Diagnostic Radiology Associates and Park West Radiology, who will use his first SASBoy for the primary SAN at Park West, attached to a PACS.

“The BladeCenter and our switches in that data center already throttle back power in the evening when I/O isn’t present,” Funaro said. “Adding that kind of power savings was something we considered with all our new data centers.”

For the radiology clinics, business isn’t always coming in at a consistent rate and none of the clinics operate 24 hours a day. “Sometimes they may have no cases, other times they might have an appointment in the morning and the next study isn’t done until three hours later,” Funaro said. “Why have the storage spinning all that time? If there is a busy day, the array will stay spinning automatically.”


Follow

Get every new post delivered to your Inbox.