Closed Loop Cooling & Monitoring System - Opengate Data Systems

Part of attending the http://www.energetics.com/datacenters08/ conference is I am able to catch up with a lot of people in a good networking environment. Jason Dudek from APC is one of person who I have worked with off and on for over 10 years.  Jason wanted to make the point that he has a lower carbon footprint as he drove to a park and ride, took a bus, and he is riding his bike home. So, yes he wins the lowest carbon footprint for the day.

But more important Jason pointed me to http://www.opengatedata.com/ who have a closed loop cooling & monitoring system for legacy equipment in a data center.

They have a Cooling System

SiteX Rack Airflow Control Systems

SiteX Rack Airflow Control Systems provide reliable, redundant and easily scaleable closed-loop pressure control to eliminate bypass and maintain proper rack to IT equipment airflow in a variety of rack fan configurations.

Network versions allow remote management and alarm notification for superior intelligence and visibility of critical information over Ethernet. Integrate with SiteX HD Fan Trays or use with your existing rack fans.

Deliver predictable temperatures to your IT equipment and achieve the highest degree of data center cooling efficiency with SiteX from Opengate Data Systems.

  • Effective Rack Airflow Control for the highest degree of efficiency and availability
  • Reduce or eliminate CRAC over-provisioning by providing physical separation of cool supply and hot return airstreams
  • Report rack airflow capacity for future application deployment
  • Couple CRAC cooling to IT equipment heat loads, raising return air temperatures and improving CRAC and chiller efficiency
  • Report rack airflow rate for CRAC/H airflow delivery control

and, a Monitoring system

SiteX HD Environment Monitoring

Effective environment monitoring with up to 16 temperature and humidity sensors plus three additional I/O ports for water, smoke, and door open position sensors in a 1U rack-mount device.

Communicate using HTTP, HTTPS, XML or SNMP and use DHCP to automatically connect to your network.

Customize alarm limits and behavior using the configuration page. Visibility of critical information with the highest degree of reliability can be ensured with SiteX HD from Opengate Data Systems.

Environment Monitoring Series PDF

SiteX HD™ Environment Monitoring
Effective environment monitoring and visibility of critical alarms over Ethernet.

EM160
Environment Monitoring with 16 sensor ports + 3 I/O ports

EM160D
Environment Monitoring with 5 sensor ports + 3 I/O ports

EMS10
Temperature Sensor, 12 Foot* Cord

EMS20
Temperature / Airflow Sensor, 12 Foot* Cord

EMS30
Temperature / Airflow / Humidity Sensor, 12 Foot* Cord

EMS40
Water Sensor, connect to I/O port

EMS50
Smoke Sensor Kit, connect to I/O port, requires 120V source

EMS60
Door Open Sensor, connect to I/O port

EMS700
Camera, Web Enabled

Read more

Enterprise-Rent-A-Car uses Green Computing to Attract IT Job Prospects

ComputerWorld has an article where Enterprise-Rent-A-Car is the main reference customer and discusses the deployment of thin-clients.

Environmental impact: IT sees the light on green computing

For some organizations, reducing the energy consumed by IT equipment is becoming a selling point with customers and even potential new hires.

By Patrick Thibodeau

June 30, 2008 (Computerworld) This version of the story originally appeared in Computerworld's print edition.

When Enterprise Rent-A-Car Co. completes a move from PCs to thin clients this summer, as part of the rollout of a new rental transaction system, it expects to cut internal energy consumption by 5 million kilowatt-hours.

That will save about $500,000 annually while reducing the company's carbon dioxide emissions by 6.5 million pounds each year, according to Enterprise officials. Energy costs and consumption were considerations when planning for the new system began earlier in the decade. But they have grown in importance for Enterprise -- so much so that the company issued a press release in April to let the public know how its shift to thin clients would help the environment.

Here is the enterprise press release, and an environment web site for Enterprise, National and alamo rental cars.

And, it was nice to see Enterprise IT staff has seen Green Computing as a way to make its company stand out in recruiting IT staff. This point is missed by many who don't understand how much pride people take in working on Green solutions for the data center and IT.

Meanwhile, Enterprise isn't just touting its green computing exploits to the general public. The committee that's evaluating the environmental impact of new technologies includes a representative from the human resources department. Miller said the company has found that green computing is a strong selling point in attracting IT job prospects, and it wants HR to know what it's doing in that area.

Read more

Virident Announces Flash Memory platform, an Enterprise Server Version of Microsoft Windows Vista ReadyBoost feature

Virident announced the GreenGateway product to enable integration of Flash Memory into servers. I had the opportunity to interview Virident CEO Raj Parekh to get a better understanding of the Green Gateway and how the product enables a Green Data Center.

The WSJ has coverage as well.

What is the Virident Green Gateway? One simple way to look at this is this is the Microsoft Windows Vista ReadyBoost feature for Servers, but with a lot more.

Windows Vista introduces Windows ReadyBoost, a new concept in adding memory to a system. You can use non-volatile flash memory, such as that on a universal serial bus (USB) flash drive, to improve performance without having to add additional memory "under the hood."

The flash memory device serves as an additional memory cache—that is, memory that the computer can access much more quickly than it can access data on the hard drive. Windows ReadyBoost relies on the intelligent memory management of Windows SuperFetch and can significantly improve system responsiveness.

A bunch of companies have been evaluating flash memory for servers, but no one has a product meant to work with most existing servers.

For more about the GreenGateway here is an excerpt from Virident's press release.

Virident GreenGateway turns Flash into DRAM-class memory enabling high-capacity, industry standard data-centric servers

SAN FRANCISCO – June 24, 2008 – Virident Systems announced today that it has developed the first data-centric technology platform to deliver a new class of green memory, servers and applications for the data center. Virident’s high-performance, low-power platform solution enables Internet data centers to cost-effectively keep pace with the rising performance and energy demands of Internet-scale.

GreenGateway™, conceived and developed by Virident, turns today’s compute-centric servers into large memory data-centric servers. This new class of servers offers random access, low-power, non-volatile, high-performance flash memory to applications such as internet search, social networking, data analytics and content distribution. The new memory solution provides DRAM-class read performance with larger memories, improving server efficiency and application throughput immediately. Ultimately, this sets the stage for several revolutionary changes in Internet server design and power efficiency to be announced in the future by Virident. Spansion’s (NASDAQ: SPSN) EcoRAM™ is a MirrorBit Flash-based product built on the Virident platform, utilizing GreenGateway™ technology.

image

Back to the interview here are some details from notes my interview with Raj.

  • The Software and Hardware Virident developed has multiple functions acting like a flash memory controller to address Flash Memories weaknesses and maximize the strengths, making GreenGateway perform like RAM, but with lower power consumption for a greener solution.
  • GreenGateway loads custom drivers in Linux (their target market is Linux first, and other OS's later), creating expanded memory that provides up to 1/2 TB of low power flash memory for a server.
  • Data Redundancy and Data Failure are examples of issues Virident have addressed in their software given issues like Flash Memory Wear.
  • GreenGateway handles the movement of data between Flash and RAM, but they do have the ability for customers to handle this movement themselves. This is a good customer feature as it reminds me when working on OS's Virtual Memory functionality is designed for the masses, but you have a customer like Adobe Photoshop on the desktop, saying get out of way, we want to handle memory mapping between RAM and HD ourselves as we know how our app performs. Virident has the ability to ship custom SW/Drivers which tunes the performance for specific user scenarios. I think this is one of the smartest things they have.
  • Given the persistence of Flash, it is possible to create a diskless server, working completely from GreenGateway memory.  It is rumored Google does this on some of their servers, loading from Linux from PROM.
  • At this time there is no compression in memory system
  • The first steps of Virident is to make Flash perform like DRAM. Next step is total performance of system and optimization analytics.
  • Virtualization is of course a scenario for GreenGateway, but it is very dependent on what users are virtualizing to know how well GreenGateway will work.

The first version of GreenGateway is meant for existing servers, but what will be interesting is if servers like Dell XS23 and IBM's iDatalplex integrate GreenGateway. It will be interesting to see some performance per watt figures once they are available. Raj gives his view of the potential business value.

“Today’s servers were not built with the data-centric needs of the internet in mind.  As a result, compute-centric servers in internet data centers can be made far more efficient with faster access to larger main memories.  We created GreenGateway to enable alternative memory technologies to replace DRAM and deliver performance and energy efficiency. The GreenGateway platform will enable internet companies to access far larger main memories, achieving growth while living within the pressing power, space and cost constraints of the data center. Ultimately green technology will save the data center.  At Virident, our goal is to make green and growth possible.”

Read more

Is it Time to Turn off your Servers?

My article in TechNet is live, “Is it time to Turn off your Servers?”

One of the simplest ways to achieve energy savings on desktops and laptops is to turn them off (or even use the sleep or hibernation feature). Users have accepted the idea that turning off their devices is a small step that can have significant results. But the idea of turning off your servers—that's a pretty new topic that people

haven't entirely embraced yet. Currently, only small numbers of IT departments shut down servers, but this practice is starting to change. More organizations are taking advantage of the energy savings of server shutdown.

Why would you turn off a server? In general, an idle server consumes about half the amount of power consumed by a server running under high load. Server utilization for most data centers runs in the range of 10 to 20 percent, which means there are a lot of machines idling (systems using energy, but not being used).

Of course, the idea of turning off servers raises some important questions. I'm not suggesting that you shut down Web servers and other critical servers that should be running 24 hours a day, 7 days a week. So what type of machines can you turn off? To help give you an idea, I'd like to share some real-world examples of what some organizations have implemented to cut down on system idle time and reduce energy consumption.

In this article I mention people I know of who are experimenting with turning off servers at Weill Medical College at Cornell University, Citrix, and Microsoft Research.

I’ll have another article in the Aug TechNet, and as soon as it is released and I will blog myself. :-)

Read more

Power Use is Major Factor in Supercomputer Growth

At the International Supercomputing Conference many of the computers are including power consumption along with their performance numbers. ComputerWorld writes the following article.

June 18, 2008 (Computerworld) Proving that size isn't everything, the Top500 List of supercomputers for the first time is looking at power efficiency.

With the IT industry increasingly looking at growing electric bills and calls for greener machines, the world's most powerful and largest systems are under scrutiny. The latest edition of the twice yearly Top500 List of supercomputers was unveiled today at the International Supercomputing Conference in Dresden, Germany. At the same time, the list authors also provided energy-efficiency calculations for many of the computers on the list.

"Power consumption is becoming one of the most important aspects of computing," said Jack Dongarra, a co-creator of the Top 500 List and a distinguished professor at the University of Tennessee. "It will be the most important driving force for supercomputing in the future. Without focusing on that, building bigger machines will be prohibitive. We're trying to understand which machines are more efficient, why they're more efficient, and understand the trends in high-performance computing."

Other interesting parts of the article.

As the industry now moves beyond the petaflop barrier toward the loftier exaflop barrier, Dongarra pointed out that increasing power efficiency becomes even more important.

"The projections for exaflop machines [show that they] will require on the order of 100 megawatts of power," said Dongarra. "It will require very special facilities to run that. The cost of running a petaflop machine at a modest electrical price will almost equal the money you pay for the computer."

He noted that engineers at the Oak Ridge National Laboratory are working with Cray Inc. to build two supercomputers -- Jaguar and Kraken. Both are in the early stages of production but are expected to surpass a petaflop. Dongarra said when they are fully functional, which is expected to happen in 2012, the power costs for both will total an estimated $32 million a year.

The list  is here.

For the first time, the TOP500 list will also provide energy efficiency calculations for many of the computing systems and will continue tracking them in consistent manner.

image

Read more