EMC's Innovation group builds thermal profile data center robot for $200

One of my readers, Vivek sent a link to a cool robot used in the data center to collect thermal profile data.  You could have permanent thermal sensors in your data center or have someone wander around to collect data or send a robot around.  Having the robot go around 24x7x365 a year seems like a good choice, and given it is built on iRobot you have a clean floor too. :-)

This idea is to build a low cost platform to monitor environmental parameters in a data center. We initially  planned to take an arduino with DS18B20 temp sensors around & build a temperature map of the data center. But we need to take care of the indoor location information as well with this method. That looked tedious & error prone. It is a good thermal detector but not good to build a thermal map. So we brainstormed with our team and some one joked about putting it on a Roomba & driving it around. The idea looked frugal because either you can put hundreds of sensors in your data center or take few sensors & walk around. Both are different in technical perspective but the later approach which is very low cost & good enough for quick data center cooling fixes.

I had a chance to have an e-mail discussion with Vivek and one of questions I had is how he knows where the robot is in the data center.  The answer.  They know the start point, and they know the wheel movement which then creates a path of where the robot is.  But, if the robot is kicked, then the location is unknown.  

 

Seems like this is a good project for a summer intern to try at your data center.

 

Here is a video of the robot.

SW = HW Google's Jeff Dean, employee #20

Google has some really smart people and there is an inner circle of smart rich people who keep Google infrastructure going.  Silicon Valley has an article on Google's Jeff Dean.

How Google's Jeff Dean became the Chuck Norris of the Internet

By Will Oremus, Slate

Posted:   01/24/2013 12:46:39 PM PST

 

 

"The speed of light in a vacuum used to be about 35 mph. Then Jeff Dean spent a weekend optimizing physics." — Jeff Dean Facts

Jeff Dean facts aren't, well, true. But the fact that someone went to the trouble to make up Chuck Norris-esque exploits about Dean is remarkable. That's because Jeff Dean is a software engineer, and software engineers are not like Chuck Norris. For one thing, they're not lone rangers - software development is an inherently collaborative enterprise. For another, they rarely shoot cowboys with an Uzi.

Jeff is a low level guy who gets to the bits running on the HW.  What really smart people like this get is it is all about processing bits.  Processing bits is done in SW and HW, and sometimes SW changes the HW. 

Nevertheless, on April Fool's Day 2007, some admiring young Google engineers saw fit to bestow upon Jeff Dean the honor of a website extolling his programming achievements. For instance:

Jeff Dean (Courtesy: Google)

Compilers don't warn Jeff Dean. Jeff Dean warns compilers.

Jeff Dean writes directly in binary. He then writes the source code as documentation for other developers.

When Jeff Dean has an ergonomic evaluation, it is for the protection of his keyboard. Jeff Dean was forced to invent asynchronous APIs one day when he optimized a function so that it returned before it was invoked.

Here is a bit of a peak into the inner circle fueled by cappuccinos

Almost every morning, he comes into work at the GooglePlex in Mountain View, Calif., and sits down for coffee with the same core group of people. "We've made 20,000 cappuccinos together" over the years, he estimates. These people don't all work together. In fact, as Google has grown, some have moved to different buildings on opposite sides of the campus. But when they get together to dish about what they're doing, their problems spark ideas in one another, Dean says. These coffee talks are what has enabled Dean to put his expertise in optimization, parallelization and software infrastructure to work on such a wide array of projects. That and healthy doses of ambition and confidence. "He's always very enthusiastic and optimistic about how much we can get done," says Ghemawat, his longtime collaborator. "It's hard to discourage him."

Here is a presentation that Jeff did on Google in 2009, it gives you some history and pictures of what Google Data Centers used to look like.

Calling Data Center Hardware Hacks, Open Compute Project event on Jan 16-17

Open Compute Project is hosting for the first time a hardware hack event.

OCP Hardware Hack!

Wednesday, December 5, 2012· Posted by John Kenevey at 16:07 PM

OCP is hosting its first hardware hackathon at the upcoming Open Compute Summit, January 16-17, 2013 in Santa Clara, California. Starting today, you can register for the hack. We are limiting attendance to 100 people. Registering for the hack also registers you for the entire OCP summit, so you can register for both events at once. The summit and the hack are both being held at the Santa Clara Convention Center.

We ask that once you register for the hack, you participate in the entire hack, which will last 6-10 hours over the course of the two day summit.

An example hack project is.

Example hack project: Use low-power sensors for temperature information across a data center. Use the Zigbee wireless protocol and aggregate the heat data across the data center. This has the benefit of not requiring any additional wiring or interfaces.

Here is what you can expect.

Goals, Tools and an Example

Goal: Design a set of “Lego” blocks that can be applied to the scale compute data center space with a focus on improving energy efficiency, operational efficiency and cost reduction.

Design tools:

• ECAD, electrical and holistic collaboration: Upverter
• Software collaboration: GitHub
• Mechanical collaboration: GrabCAD

Skill set: electrical engineer, mechanical engineer, software engineer, designer. (Ideally each team has a combination of these skill sets.)

Starting point: We recommend starting with a really simple circuit that can be modified, or even whole-scale deleted, but provides a great base to scaffold onto. It could be a connector, a micro-controller or a power circuit.

APC's UPS achieve ENERGY STAR qualification

An answer in the consumer and small business market to save energy is to disconnect those Vampire Electronics.  Those devices that use power even when you don't need them.  But, one of the devices you can't unplug is your UPS device with the emergency battery back-up for those devices that need to survive power outages.  But, as any of who have focused on a better PUE you know the UPS can be huge user of power and cooling capacity in your data center.

The EPA has recognized this issue with UPS's and has been working on ENERGY STAR program for UPS.  Schneider Electric announced they are one of the first to qualify.

Schneider Electric’s Uninterruptible Power Supply Products First to Achieve ENERGY STAR® Qualification

 Power protection devices achieve inaugural approval in ENERGY STAR program’s new category

 

WEST KINGSTON, R.I. – September 5, 2012 – Schneider Electric, a global specialist in energy management, today announced select Back-UPS™ and Smart-UPS™ (uninterruptible power supply) models have become the first-ever UPSs to earn the ENERGY STAR qualification under the federal program’s new UPS category. Acknowledged for meeting standardized energy efficiency requirements and passing the program’s testing, the selected Smart-UPS and Back-UPS products have been recognized for their abilities to both save customers money while helping to protect the environment.

So, how much energy can you save compared to a device that is not ENERGY STAR?  Here is some info from the EPA site.

Did You Know?

If all Uninterruptible Power Supplies sold in the United States in 2012 meet the ENERGY STAR requirements, the energy cost savings will grow to $471 million and prevent greenhouse gas emissions equivalent to the emissions from more than 636,000 vehicles.

...

ENERGY STAR certified UPS can cut energy losses by 30-55%. A 1000 kVA UPS used in a large data center could save $18,000 annually.

Here are the range of devices that the ENERGY STAR UPS program covers.

ENERGY STAR UPS covers from the small devices beneath your desk protecting your computer to 8-ton versions designed to temporarily provide a megawatt of power to large data centers.

UPS Topology Typically Referred To As:	Referred To In ENERGY STAR Specification As:	Typically Sized Up To:	Typically Used For:
Passive Offline Standby	Voltage and Frequency Dependent (VFD)	1,500 VA	small offices, personal home computers and other less critical applications
Line Interactive	Voltage Independent (VI)	5,000 VA	small business, Web, and departmental servers
Online Continuous Double Conversion	Voltage and Frequency Independent (VFI)	1,000 kVA	data centers

Google hits #5 spot in Server Mfr Ranking, The Big Users defining the market

Wired has a post that has people's attention that HP, Dell, and IBM are no longer the dominant Server force of the past.  Part of the past is the reliance on IDC and Gartner to get a view on the market.

It should be noted, however, that research operations such as IDC and Gartner don’t have the best view into direct sales by the ODMs — let alone Google’s highly secretive hardware operation — and these hidden parts of the market are increasingly important. It’s the big web players that are moving away from the HPs and the Dells, and most of these same companies offer large “cloud” services that let other businesses run their operations without purchasing servers in the first place.

What is catching people's attention is that Google is the #5 manufacturer.

But just four years later, Bryant says, the landscape has completely changed. Today, she explains,eight server makers account for 75 percent of Intel’s server chip revenues, and at least one of those eight doesn’t even sell servers. It only makes servers for itself. “Google is something like number five on that list,” Bryant told us on Monday evening, during a dinner with reporters in downtown San Francisco.

But, the overall pattern that is occurring is Google, Amazon, Facebook, Tencent, and Microsoft are investing and demanding server innovation that the big OEMs (HP, Dell, and IBM) are not necessarily set up to serve.  Which then brings up the ODMs of Quanta, Tyan, Supermicro, ZT Systems and others who are rising up the ranks.

As Bryant points out, other companies are now buying machines directly from “original design manufacturers,” or ODMs, in Asia, working to cut costs in much the same way. This includes Facebook, and according to a former employee of one large ODM, it includes Amazon as well.

These ODMs in general don't have the marketing budgets to have IDC and Gartner analyze and report their sales. 

Would you trust an Intel executive or IDC/Gartner and a vendor to give a report on the state of the server market?

But an HP spokesperson said her comments were inconsistent with the latest server market stats from research outfit IDC, which still put the combined market share of HP, Dell, and IBM at 73.9 percent, down slightly from 78.2 percent in 2008.