Cloud // Infrastructure as a Service
News
12/3/2013
04:10 PM
Connect Directly
Twitter
RSS
E-Mail
50%
50%

Google Compute Cloud Challenges Amazon

Google's Compute Engine comes out of preview, with Google touting lower pricing and "transparent maintenance."

8 Great Cloud Storage Services
8 Great Cloud Storage Services
(click image for larger view and for slideshow)

Google has entered the highly competitive cloud services marketplace, where it will face off against major tech vendors like Amazon and Microsoft. Google first unveiled its Compute Engine June 2012 as a beta service and Tuesday announced its infrastructure-as-a-service is now generally available. Google also lowered prices on some compute instances 10% and on disk storage by 60%.

Google's IaaS will appeal to those who want infrastructure geared to a high level of performance. The company's emphasis on high-performance infrastructure is evident from the speed of its famous Google Search engine. Compute Engine public cloud infrastructure is based on the same architecture. At the same time, it's offering a service level agreement that offers 99.95% uptime. That's a match for Amazon Web Services, which changed its SLA on June 1 to 99.95% from its previous 99.9% uptime.

Google will charge by minutes of use, rather than rounding up to the nearest hour after a customer has ventured 15 minutes into the hour, as Amazon does. Google does charge for a minimum of 10 minutes use, after which it charges in one-minute increments. It will round up to a full minute of charge after 15 seconds, even if the virtual machine is shut down at the end of 15 seconds.

"Today we’re lowering the price of Persistent Disk by 60% per Gigabyte and dropping I/O charges so that you get a predictable, low price for your block storage device," wrote Google VP of Engineering Ari Balogh in a Compute Engine blog post. Block storage is used with running applications and remains a highly competitive front in service offerings from Amazon, Microsoft, and Google.

[Want to learn more about why cloud pricing is so hard to compare? See Why Cloud Pricing Comparison Are So Hard.]

Balogh said Google is also reducing compute instance prices by 10%. The announcement didn't specify which instance types other than to refer to pricing on its "most popular, standard Compute Engine instances."

At the same time it's introducing in "limited preview" three more power instance types: 16-core virtual machines coming in standard, high memory (up to 104 GB of RAM), and high CPU configurations. Google's limited preview means a customer must fill out a form and be approved to test drive the new types. They are not generally available.

During its preview phase, Compute Engine supported Debian Linux or Centos, the Red Hat knock-off; each was given a customized, Google-built kernel. The custom kernel wouldn't run software that wasn't designed for those. Now Compute Engine will support all Linuxes out of the box, including Ubuntu, Suse, Red Hat Enterprise Linux (but only in limited preview), CoreOS, FreeBSD, and SELinux, used with various distributions as a security-enhanced kernel.

It will also support the use of open-source Docker containers with Linux applications running inside.

Google will offer a "transparent maintenance service" that uses live migration to move virtual machines off a host that's undergoing operating system patches or other maintenance, with no disruption to end-users. That is, Google customers will get the protections of frequent, proactive maintenance without downtime and rebooting, Balogh wrote.

The transparent maintenance feature is supplemented by another to offer more of a worry-free operations environment. "In the event of a failure, we automatically restart your VMs and get them back online in minutes. We’ve already rolled out this feature to our US zones," he wrote in the blog. Amazon customers are responsible for monitoring and restarting their own workloads, in the event of failure.

"Google also had the vision to buy up dark fiber nearly ten years ago to create low-latency connections between its data centers," wrote RightScale's VP of products Rishi Vaish on Tuesday in welcoming Compute Engine to the set of clouds that RightScale supports with its cloud management service. Its high-speed connections help power Google Search and Google Maps. "Now this same advanced infrastructure is available to Compute Engine users," he noted.

The use of cloud technology is booming, often offering the only way to meet customers', employees', and partners' rapidly rising requirements. But IT pros are rightly nervous about a lack of visibility into the security of data in the cloud. In this Dark Reading report, Integrating Vulnerability Management Into The Application Development Process, we put the risk in context and offer recommendations for products and practices that can increase insight -- and enterprise security. (Free registration required.)

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
<<   <   Page 2 / 2
cbabcock
50%
50%
cbabcock,
User Rank: Strategist
12/3/2013 | 6:49:31 PM
RightScale ran a performance test
Any support for that claim of Google performance? Well, RightScale ran a test that built up to 200,000 simulated, concurrent users making 23,000 requests a second against a content serving application running on 42 Compute Engine servers. It served up 330,000 page views a minute, or 20 million in the course of building up to peak use over an hour. The content sought was designed to vary from request to request and issued from a MySQL database. Rightscale spokesmen said Compute Engine's pace of operation was double the rate experienced by a widely used online application, Evernote, which allows individuals to list, record and recall their events and documents.

 

 

 
<<   <   Page 2 / 2
Multicloud Infrastructure & Application Management
Multicloud Infrastructure & Application Management
Enterprise cloud adoption has evolved to the point where hybrid public/private cloud designs and use of multiple providers is common. Who among us has mastered provisioning resources in different clouds; allocating the right resources to each application; assigning applications to the "best" cloud provider based on performance or reliability requirements.
Register for InformationWeek Newsletters
White Papers
Current Issue
Flash Poll
Video
Slideshows
Twitter Feed
InformationWeek Radio
Archived InformationWeek Radio
Join InformationWeek’s Lorna Garey and Mike Healey, president of Yeoman Technology Group, an engineering and research firm focused on maximizing technology investments, to discuss the right way to go digital.
Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.