Light Reading
Google's Andromeda infrastructure is designed to provide bare-metal performance and low latency to customers worldwide.

Google's Andromeda Relieves Cloud Strain

Mitch Wagner
8/26/2014
50%
50%

MOUNTAIN VIEW, Calif. -- Hot Interconnects -- Google relies on its Andromeda networking platform to deliver a global cloud infrastructure that gives customers the security and performance benefits of local private networks.

"We want bare-metal performance and low latency for the services we deliver," said Google (Nasdaq: GOOG) Distinguished Engineer Amin Vahdat, delivering a keynote at the conference here today.

SDN is key to delivering the needed performance and security, he said.

SDN at its most fundamental involves separating the control plane from the data plane, Vahdat noted. "A logically centralized hierarchical control plane beats peer-to-peer every time," he said. The data plane can run at network speed, while the control plane can run on commodity hardware, scaling as needed. The control plane requires 1% of the overhead of the entire network, Vahdat said.

But managing that infrastructure requires new tools and skills, he said.

"It turns out that running a hundred or a thousand servers is a very difficult operation. You can't hire people out of college who know how to operate a hundred or a thousand servers," Vahdat said. Tools are often designed for homogeneous environments and individual systems. Human reaction time is too slow to deliver "five nines" of uptime, maintenance outages are unacceptable, and the network becomes a bottleneck and source of outages.

Google looks to SDN and network functions virtualization (NFV) to orchestrate provisioning, high availability, and meet application performance requirements, Vahdat said. The technology must be distributed throughout the network, which is only as strong as its weakest link.

Andromeda is Google's code-name for its network virtualization platform. It's designed to provide each external user with the illusion that they're on a dedicated network with dedicated performance and its own IP address space. Applications require real-time high performance and low-latency communications to virtual machines. Users also require service chaining to tools such as load-balancing, and the ability to grow and shrink the number of servers available to applications as demand requires. (See Google, Microsoft Challenge Service Providers and Google's Andromeda Strain Is Spreading.)

Security is a huge requirement. "Large companies are constantly under attack. It's not a question of whether you're under attack but how big is the attack," Vahdat said.

Power and cooling are the major costs of a global infrastructure like Google's. "That's true of even your laptop at home if you're running it 24/7. At Google scale, that's very apparent," Vahdat said.

Google has a global infrastructure, with data centers and points of presence worldwide to provide low-latency access to services locally, rather than requiring customers to access a single point.

The company runs two networks. Its private, server-to-server network is bigger than its public network, and one of the world's largest SDN deployments. Connectivity between data centers is comparable to within data centers.

Andromeda provides significant performance improvements over a state-of-the-art baseline, as seen in Vahdat's slides:

The promise of cloud computing is just beginning.


Find out more about key developments related to the systems and technologies deployed in data centers on Light Reading's data center infrastructure channel.


"Many people think about cloud computing as being able to get on-demand access to computing. I don't have to go buy servers; I can rent them for a minute, or an hour, or a day. I can get burst capacity of as many servers as I like, whatever memory, configuration or disk, etc., that I like. I think actually yes, this is powerful, but this is really just the beginning," Vahdat said. "The really exciting parts of cloud computing are on the verge of happening."

These include a fundamentally easier operational model; higher uptime; state-of-the-art infrastructure services such as denial-of-service protection, load balancing, and storage; and new programming models for low latency and massive input-output performance.

What cloud doesn't do is take away the challenges of running an IT infrastructure. "Most cloud customers, if you poll them, say the operational overhead of running on the cloud is as hard or harder today than running on your own infrastructure," Vahdat said.

Click the photo below for a selection of Vahdat's slides -- and more.

He's Hydrated
Google Distinguished Engineer Amin Vahdat
Google Distinguished Engineer Amin Vahdat

— Mitch Wagner, Circle me on Google+Follow me on TwitterVisit my LinkedIn profileFollow me on Facebook, West Coast Bureau Chief, Light Reading. Got a tip about SDN or NFV? Send it to wagner@lightreading.com.

(8)  | 
Comment  | 
Print  | 
Newest First  |  Oldest First  |  Threaded View
pcharles09
50%
50%
pcharles09,
User Rank: Light Beer
8/31/2014 | 10:40:01 PM
Re: What's the protocol?
@jabailo,

Wouldn't the packet alterations get messy though? In the TCP case , if somethings off just a little bit. Also, there'd have to be extra overhead for error checking right?
jabailo
50%
50%
jabailo,
User Rank: Light Sabre
8/27/2014 | 10:52:17 PM
Re: What's the protocol?
Like in a private network you can make assumptions that you can't make when shipping it out across the public Internet.

For example, I'm looking at this diagram:

http://www.freesoft.org/CIE/Course/Section4/8.htm

What about all that space for "source port" and "destination port".   Inside your own network, do you need to allocate that many bits?

Seems like for every bit you can reduce in a packet you get that much greater throughput.

 
pcharles09
50%
50%
pcharles09,
User Rank: Light Beer
8/27/2014 | 5:58:28 PM
Re: What's the protocol?
@jabailo,

Ahh ok I see. That's a good point.
kq4ym
50%
50%
kq4ym,
User Rank: Light Sabre
8/27/2014 | 9:45:56 AM
Androworld
Google certainly is going to lead the way in NFV/SDN services and of course the free PR they get for all announcements is not a bad thing for them either. Security is still going to be an ongoing problem and probably others are going to see their way to NFV and the cloud just to help solve that issue.
jabailo
50%
50%
jabailo,
User Rank: Light Sabre
8/27/2014 | 12:56:38 AM
Re: What's the protocol?
Sure, I was just thinking they could reduce the packet sizes by removing some of the headers, since it all runs "in-house" on their cloud.
pcharles09
50%
50%
pcharles09,
User Rank: Light Beer
8/27/2014 | 12:29:06 AM
Re: What's the protocol?
@jabailo,

My guess is because that's what they've always used. No reason to change unless there's a problem/vulnerability.
jabailo
50%
50%
jabailo,
User Rank: Light Sabre
8/26/2014 | 10:51:07 PM
What's the protocol?
Just how raw is the data layer protocol?  Is there any reason for it to be tcp/ip? 
Atlantis-dude
50%
50%
Atlantis-dude,
User Rank: Light Sabre
8/26/2014 | 7:42:42 PM
Autopilot
Is Andromeda the same as Azure's Autopilot? And what is the baseline?
Flash Poll
LRTV Huawei Video Resource Center
Network Solutions Help the Philippines Jump Ahead

9|17|14   |   2:59   |   (0) comments


In the past, the Philippines has under-invested in technology. Now, the CEO of Softshell talks about how Huawei products help the Philippines jump ahead as the economy improves.
LRTV Huawei Video Resource Center
VCS Observation for Safer Cities in the Netherlands

9|17|14   |   5:20   |   (0) comments


Holland's VCS Observation has been operating for 22 years. Its main goal is to get cities safer. CEO Wim van Deijzen tells us some of the challenges his company faces and how Huawei is helping to overcome these challenges.
LRTV Huawei Video Resource Center
A Conversation With Serbia's Ministry of Interior

9|17|14   |   4:38   |   (0) comments


At HCC 2014, the Assistant Minister of the Ministry of Interior of the Republic of Serbia talks to us about his projects and corporation with Huawei. Solutions like Safe City and E-Government and services like cloud computing are just some of the areas his department is interested in.
LRTV Huawei Video Resource Center
IHS Analyst Discusses eLTE at CCW 2014

9|10|14   |   7:09   |   (0) comments


Thomas Lynch, associate director of critical communications at IHS Technology, talks about broadband in critical communications.
LRTV Huawei Video Resource Center
TCAA on Huawei eLTE: A Broadband Solution for Mission-Critical Communications

9|10|14   |   2:29   |   (0) comments


At CCW2014 in Singapore, the TCCA's Phil Kidner talks about the importance of broadband data for critical communications.
LRTV Custom TV
Spotlight on Cisco: SDN for Optical Networks

9|8|14   |   9:27   |   (0) comments


Cisco's Greg Nehib talks OpenFlow and more on the 'Software-Defined Networking for Optical Networks' panel at the Big Telecom Event in June 2014.
LRTV Custom TV
Cisco's Evolved Programmable Network (EPN)

9|8|14   |   4:05   |   (0) comments


A look at the various demos Cisco showed at Light Reading's Big Telecom Event highlighting Cisco's EPN innovation and how SDN and NFV technologies are enabling a variety of new services.
LRTV Huawei Video Resource Center
The Future of Ultra-Broadband, With Kevin Kelly (UBBF2014)

9|5|14   |   1:13   |   (1) comment


If you think the technological changes we've seen up to now are astounding, just wait until you see what the future has in store. Discuss upcoming breakthroughs with Kevin Kelly, Founding Executive Editor of Wired Magazine, at the Huawei Ultra-Broadband Forum on September 24.
LRTV Huawei Video Resource Center
The Inaugural Optical Innovation Forum in Nice

9|2|14   |     |   (0) comments


More than 170 attendees from network operators, service providers, analyst firms, and component companies from around the world convened in Nice in June for the inaugural Optical Innovation Forum, co-produced by Huawei and Light Reading.
Wagner’s Ring
Data Centers Drive Telcos Into the Future

8|28|14   |   2:20   |   (2) comments


Data centers are at the heart of key trends driving telecom -- network virtualization, the drive for increased agility, and the need to compete with OTT providers.
LRTV Custom TV
Why SPs Should Consider Cisco's EPN

8|27|14   |   5:40   |   (0) comments


Sultan Dawood from Cisco discusses Cisco's EPN, which enables SPs to build agile and programmable networks delivering new network virtualized services using Cisco's Evolved Services Platform (ESP).
LRTV Huawei Video Resource Center
Huawei’s Showcase @ Big Telecom Event 2014

8|26|14   |   2.56   |   (0) comments


SoftCOM is Huawei's framework for telecom business and network transformation. Haofei Liu, Solution Marketing Manager, Carrier Business Group, Huawei, showcases Huawei's SoftCOM architecture in this video.
Upcoming Live Events!!
September 23, 2014, Denver, CO
October 29, 2014, New York City
November 6, 2014, Santa Clara
November 11, 2014, Atlanta, GA
December 2, 2014, New York City
December 3, 2014, New York City
December 9-10, 2014, Reykjavik, Iceland
June 9-10, 2015, Chicago, IL
Infographics
A survey conducted by Vasona Networks suggests that 72% of mobile users expect good performance all the time, and they'll blame the network operator when it's not up to par.
Today's Cartoon
Vacation Special Caption Competition Click Here
Latest Comment
Hot Topics
Glimpsing the Self-Driving Car
Dan Jones, Mobile Editor, 9/12/2014
AT&T to Launch WiFi Calling in 2015
Sarah Reedy, Senior Editor, 9/12/2014
AT&T: We'll Bundle Fixed Wireless & DirecTV
Mari Silbey, Independent Technology Editor, 9/15/2014
New NFV Forum Focused on Interoperability
Carol Wilson, Editor-at-large, 9/16/2014
Connecticut Cities Crowdsource Gigabit Nets
Jason Meyers, Senior Editor, Utility Communications/IoT, 9/15/2014
Like Us on Facebook
Twitter Feed