Skip to main content

Google reveals secret server hardware

Google for the first time on Wednesday revealed the hardware at the core of its Internet operations at a conference about the increasingly prominent issue of data center efficiency, reports CNET’s Stephen Shankland.

Instead of buying hardware from companies such as Dell, Hewlett-Packard, IBM, or Sun Microsystems, Google designs and builds its own. (The company has hundreds of thousands of servers.)

Ben Jai, who designed many of Google’s servers, unveiled the server hardware. The first surprise: each server has its own 12-volt battery to supply power if there’s a problem with the main source of electricity.

Shankland writes:

Why is the battery approach significant? Money.

Typical data centers rely on large, centralized machines called uninterruptible power supplies (UPS)–essentially giant batteries that kick in when the main supply fails and before generators have time to kick in. Building the power supply into the server is cheaper and means costs are matched directly to the number of servers, Jai said.

“This is much cheaper than huge centralized UPS,” he said. “Therefore no wasted capacity.”

Efficiency is another financial factor. Large UPSs can reach 92 to 95 percent efficiency, meaning that a large amount of power is squandered. The server-mounted batteries do better, Jai said: “We were able to measure our actual usage to greater than 99.9 percent efficiency.”

Since 2005, Google’s data centers have been composed of standard shipping containers — each with 1,160 servers and a power consumption that can reach 250 kilowatts, the company said.

Google has been using the design since 2005 and now is in its sixth or seventh generation of design.

“It was our Manhattan Project,” Jai said of the design.

Energy efficiency, power distribution, cooling, and ensuring hot and cool air don’t intermingle are all on the top of Google’s list, the company said.

As for the actual unit, the server was 3.5 inches thick (2U) and had two processors, two hard drives, and eight memory slots mounted on a Gigabyte motherboard. Google uses x86 processors from both AMD and Intel. The battery design is used on its network equipment as well, Jai said in Shankland’s article.

What’s fascinating about all this is that Google operates servers on such an immense scale that every decision it must make in turn has a large affect (and potential cost/savings).

Take the power supply design, for example: Google’s designs supply only 12-volt power, with the necessary conversions taking place on the motherboard. That adds $1 or $2 to the cost of the motherboard, Shankland writes, “but it’s worth it not just because the power supply is cheaper, but because the power supply can be run closer to its peak capacity, which means it runs much more efficiently.” Google even pays attention to the greater efficiency of transmitting power over copper wires at 12 volts compared to 5 volts, Shankland writes.

That kind of attention can translate to big savings in power or cost — or both.

 

Comments

Popular Posts

DOT NET Interview Questions with Answers

When not to use Design Patterns? Do not use design patterns in any of the following situations. • When the software being designed would not change with time. • When the requirements of the source code of the application are unique. If any of the above applies in the current software design, there is no need to apply design patterns in the current design and increase unnecessary complexity in the design. When to use Design Patterns? Design Patterns are particularly useful in one of the following scenarios. • When the software application would change in due course of time. • When the application contains source code that involves object creation and event notification. Benefits of Design Patterns: The following are some of the major advantages of using Design Patterns in software development. • Flexibility • Adaptability to change • Reusability What are Design Patterns? A Design Pattern essentially consists of a p...

PDF Converter Ultimate - Awesome Converter

PDF Converter Ultimate  Convert your documents to PDF, and PDF documents to Word, Excel, JPG, and more, right on your Android device. Conversions are accurate and very fast, layout is preserved, and best-on-market OCR engines will take care of your scanned files. Easy to use app, and amazing conversion quality makes this app the ultimate converter for your phone. You can convert files from your phone, or files from your favorite cloud services directly within our app. Files are being converted by our powerful servers, and after conversion is done, result is downloaded to your device and files are being deleted from our servers. This also means the app won't burden your processor or RAM, nor it will drain your battery. • Conversions are fast and accurate • Scanned and complex documents are converted as well  • Immediate conversions for free users • Several conversion types available depending on the file format needed • No limit on file size or number of files you c...

DesignEvo - A free and easy online logo designer

DesignEvo is  easy-to-use online logo design tool with millions of icons, 100+ stylish  fonts, 3,000+ practical logo templates and handy editing tools. It can help  people easily create appealing and effective logos for their brands,  websites, blogs, social media channels, etc.  It is really awesome and intuitive interface. Easy to design and customize. A person can easily design by selecting a pre-defined logo design and modify it on the fly. Hardly you need 1 minute to design your new logo with minor modifications. Its really awesome site. I suggest you to try it out.   You can discover more about DesignEvo here: https://www.designevo.com Learn more about Designevo  @ YouTube designevo, free logo design, logo maker, create a logo, easy logo maker, design logo free, free logo, logo designs, logo images

Java J2EE Interview Questions and Answers - Part 1

What is the difference between procedural and object-oriented programs? - a) In procedural program, programming logic follows certain procedures and the instructions are executed one after another. In OOP program, unit of program is object, which is nothing but combination of data and code. b) In procedural program, data is exposed to the whole program whereas in OOPs program, it is accessible within the object and which in turn assures the security of the code. What are Encapsulation, Inheritance and Polymorphism? - Encapsulation is the mechanism that binds together code and data it manipulates and keeps both safe from outside interference and misuse. Inheritance is the process by which one object acquires the properties of another object. Polymorphism is the feature that allows one interface to be used for general class actions. What is the difference between Assignment and Initialization? - Assignment can be do...

Tavant Technologies Interview Questions

Hi Friends, This interview is for 3 years experienced Java/J2EE guy. First Round: 1) About project 2) All basics of Java – Encapsulation , Polymorphism, Over riding, inheritance, over loading 3) Exception handling – Checked and unchecked exceptions 4) Example for runtime exceptions 5) What is abstraction and difference between interface and abstract class 6) Basics of threads 7) Which one is preferable thread class or runnable interface 8) Single ton – Other design patterns - Write sample singleton class 9) Collections 10) What is the output: Try{ sysout(“hello”); some code written here which throw an exception } catch(Exception e){ System.exit(0) } finally{ Sysout (“end”); } 11) A class Thread A with overridden run method.. {run(){Sysout("TT"); }} A main program psvm{ a.run(); // call run a.run(); // call run ...