Olivier Sanche Memorial Service Jan 28, 2011 in Los Gatos, CA

It’s been almost 2 months since Olivier Sanche voice went silent and we’ll never hear him again, speaking with passion on data centers and the environment. 

A Memorial Service will be held on Jan 28, 2011 at 4p in Los Gatos, CA.

Here are details.

Dear all,


Some of you might not be aware of this terrible news so it is with great sorrow that I must inform you that Olivier passed away on November 26, 2010 in Europe from a sudden heart attack.


The funeral took place on December 3rd in Pignan ( his hometown), France.
There will be a memorial in his honor on Friday, January 28, 2011 at St. Mary's Catholic Church, 219 Bean Avenue in Los Gatos ( California) at 4PM.
Here is the link for directions.

The memorial will be held after school hours so there should be some parking available in the church's parking lot, otherwise the town of Los Gatos has several free public parking lots (along University Street ) as well as street parking.

Sincerely,
Karine Sanche

Please feel free to send me stories you have about Olivier as I’ll be helping to pull together a perspective on his awesome past. 

-Dave Ohara

dave@greenm.com

Read more

Increasing Energy Efficient Server Competition, ARM efforts ramps up with Virtualization Support

James Hamilton has a good post on the state of ARM powered servers.  Sometimes when I read other people’s work I think what they say in the end should be moved to the beginning of the conversation.  Here is the last paragraph from James’s post.

We are on track for renewed competition in the server-side computing market segment and intense competition on power efficiency at the same time as internet-scale service operators are willing to run whatever processor is least expensive and most power efficient. With competition comes innovation and I see a good year coming.

James points out the ARM instruction set as an advantage.

ARM has become an incredibly important instruction set architecture powering smartphones, low-end network routers, printers, copiers, tablets, and other embedded applications. But things are changing, arm is now producing designs appropriate for server-side computing at the same time that power consumption is becoming a key measure of server-side computing cost. The ARM design team are masters of low power designs and generations of ARMs have focused on power management. ARM has an impressively efficiently design.

Here is another little known fact is Virtualization Extensions in the ARM architecture.

Virtualization Extensions

  • The ARM Architecture Virtualization Extension and Large Physical Address Extension (LPAE) enable the efficient implementation of virtual machine hypervisors for ARM architecture compliant processors.
  • Connected consumer devices and cloud computing demand energy efficient, high performance systems to handle complex software with potentially large amounts of data.
  • The ARM Architecture Virtualization Extensions provides the basis for ARM Architecture compliant processors to address the needs of both client and server devices for the partitioning and management of complex software environments into virtual machines.
  • The ARM Architecture Large Physical Address Extension provides the means for each of the software environments to efficiently utilize the available physical memory when handling large amounts of data.

Part of the Virtualization extensions is more than 32 bit virtual memory addressing.

  • As the complexity of software increases the requirement for multiple software environments to be available on the same physical processor increases simultaneously. Software applications that require separation for reasons of isolation, robustness or differing real-time characteristics need a virtual processor exhibiting the required functionality.
  • To provide virtual processors in an energy-efficient manner requires a combination of hardware acceleration and efficient software hypervisors. The ARM Architecture Virtualization Extension standardizes the architecture for implementation of the hardware acceleration in ARM application processor cores, while high performance hypervisors from the world’s leading virtualization companies provide the software component upon which to build effective software combinations.
  • Cloud computing and other data or content oriented solutions increase the demands on the physical memory system from each virtual machine. The large physical address extensions provide a second level of MMU translation table so that each 32-bit virtual memory address can be mapped within a 40-bit physical memory range. This allows systems to allocate sufficient physical memory to each virtual machine for efficient throughput to be maintained when total demands on memory exceed the range of 32-bit addressing.

People will laugh at ARM servers, but when you can get a dozen or more for the price of one Xeon, there are scenarios where ARM servers will work.

Read more

China Telcos lead countries data center industry, partners with LBNL to green its data centers

Web Host Industry Review has a post on US DOE LBNL partnering with China to green data center.  I have had the pleasure of visiting LBNL labs data center efficiency group, and it is a good group to work with for this type of effort for China.  There are also some interesting facts in the post you should think about to understand the China Data Center industry.

First the role of Telcos in data center development.

China differs from the US in that the telecommunications industry has taken the lead over the IT industry in dominating the data center market.

10% of the USA data center capacity is for the Federal government, and there is a huge market for data centers that have nothing to do with the gov’t.  China’s government is the #1 player in data center capacity build out, and you will need to fit in China government’s plans.

Another major difference between China's data center market compared to the US is that most data centers in China are owned and run by the government.

LBNL points out the government influence has an advantage to green the data center in that they only need to convince the government to make the changes.

Sector[sic] Sarton says that this difference could be advantageous if they can "convince the government to make the changes... they can make the changes quicker."

Berkeley Lab is also hoping to work with China to install technologies such as warm-water liquid cooling and DC power through its partner agency.

So what is the project?

The project is being run by Dale Sartor, head of the Building Technology's Applications Team, and Bo Shen of the China Energy Group.

The Berkeley Lab scientists will work with the China Institute of Electronics and the China Electronics Standardization Institute to "share best practices, case studies and other material" to help form standards and training programs for China's data center industry.

Sarton says that China has a wide range of energy efficiency within its data centers but ultimately shows "a lot of room for improvement."

The original article referenced is here in Physorg.com.

Berkeley lab to help China improve energy efficiency of data centers

December 21, 2010

The amount of energy consumed by data centers is increasing rapidly around the world, and China is no exception. With its growing information technology and telecom industries and its emerging status as a supercomputer power, China continues to expand its data center capacity. In an effort to help reduce carbon emissions, the U.S. Department of Energy's Lawrence Berkeley National Laboratory, a world leader in high-performance buildings, has started working with China to improve the energy performance of its data centers.

Read more

China need to decentralize data centers to accommodate demand outstripping supply

Marbridge Consulting references www.cctime.com article on Jan 10, 2011 that China’s data center demand is outstripping supply.

China's Data Center Demand Outstripping Supply

CCTIME, 1/10/11

Last week, standing deputy director Zhou Hongren of the Advisory Committee for State Informatization (ACSI) said that China's data centers were facing a fundamental problem: the speed with which new data centers are being built is being outpaced by growing demand. Furthermore, China's data centers are primarily centralized in Beijing, Shanghai, Guangzhou, and other regions along the country's east coast, resulting in high power consumption and a heavy concentration of waste heat. Because of this, Zhou said, the migration of data centers northward will be a major trend over time. Zhou added that cloud storage could offer a solution to the issue.

China Unicom has a strategy to create greener data center strategy as data centers move away from the current data center hubs.

Tong Xiaoyu, deputy director of the China Unicom (NYSE: CHU; 0762.HK; 600050.SH) Research Institute, said that owing to energy-saving and cost considerations, Unicom will establish data center industry bases in the energy resource-rich provinces in central and western China.

Keywords: Zhou Hongren, cloud computing, Tong Xiaoyu, China Unicom Research Institute, China Unicom, telecom, Advisory Committee for State Informatization, data center, 0762.HK, 600050.SH, CHU, energy efficiency

Read more

Google Mail/Apps ups SLA, removes schedule downtime allowance

Data Centers and uptime is assumed.  Service Level Agreements (SLA) are made between groups.  But, many times there are exceptions for planned maintenance/downtime vs. unplanned downtime when calculating SLA.

InformationWeek reports on Google Apps/Gmail's change to this common practice.

Google Promises No Planned Downtime

A new service level agreement (SLA) for Google Apps customers strives to make Google's cloud as reliable as dial tone.

By Thomas Claburn ,  InformationWeek
January 14, 2011 02:42 PM

Google has changed its service level agreement for paid versions of Google Apps, its suite of online applications. The goal, says Google Enterprise product management director Matt Glotzbach, is to deliver service that's as reliable as telephone dial tone.

For today's mobile generation, who may lack experience with landlines, let it suffice to say that dial tone under Ma Bell was very, very reliable. Not sunrise reliable but chances were if you didn't hear a dial tone when you picked up a handset, the phone was disconnected from the wall.

Google is taking a leadership position.

But with millions of enterprise customers, Google aims to become more reliable. As a sign of its commitment, the company has disavowed planned downtime. "Unlike most providers, we don't plan for our users to be down, even when we're upgrading our services or maintaining our systems," wrote Glotzbach in a blog post. "For that reason, we're removing the SLA clause that allows for scheduled downtime."

Glotzbach says Google is the first major cloud service provider to make that pledge.

In Google's blog post they call out the competition.

Gmail: 99.984%
In 2010, Gmail was available 99.984 percent of the time, for both business and consumer users. 99.984 percent translates to seven minutes of downtime per month over the last year. That seven-minute average represents the accumulation of small delays of a few seconds, and most people experienced no issues at all. For those few who were disrupted for a longer period of time, we're very sorry, and Google Apps for Business customers received compensation where appropriate. We're particularly pleased with this level of reliability since it was accomplished without any planned downtime while launching 30 new features and adding tens of millions of active users.
Seven minutes of downtime compares very favorably with on-premises email, which is subject to much higher rates of interruption that hurt employee productivity. The latest research from the Radicati Group found that on-premises email averaged 3.8 hours of downtime per month. In comparison to Radicati's metrics for on-premises email, our calculations suggest that Gmail is 32 times more reliable than the average email system, and 46 times more available than Microsoft Exchange®.1

Fortunately Microsoft Exchange® customers can still benefit from the reliability of Gmail withGoogle Message Continuity. Comparable data for Microsoft BPOS® is unavailable, thoughtheir service notifications show 113 incidents in 2010: 74 unplanned outages, and 33 days with planned downtime.

You may be thinking I can't do this in my data center.  And you are right you can't.  This solution requires geo redundancy between data centers.  For a bit on some of Google's approach check out this Google presentation at Stanford University.

Google – A study in Scalability and A little systems horse sense

By ksankar

16 Votes

Google’s Jeff Dean did an excellent talk at Stanford as part of EE380 – it is worth one’s time to listen. Very informative, instructive and innovative. As I listened, I jotted a few quick notes.

  • Interesting comparison of the scale in search from 1999 to 2010
    • Docs and queries are up 1000X, while the query latency has decreased 5X
    • Interesting to hear that in 1999 they used to update a web page store in a month or two, but now it is reduced 50000X to seconds!
  • They have had 7 significant revisions in 11 years
  • Trivia : They encounter very expensive queries for example “circle of death” requires ~30GB of I/O
  • Trivia : In 2004, they did a rethink and refreshed the systems infrastructure from scratch
  • He discussed a little about encodings – informative discussion on Byte aligned variable length & group encoding schemes << I have to try it out …
  • Trivia : They have had long distance links failure by wild dogs, sharks, dead horses and (in Oregon) drunken hunters !

The presentation referenced is by Jeff Dean.

image

Read more