Can billions of search results in milliseconds, 6 billion hours of YouTube video per month & storage for 425 million Gmail users be wrong?
As if businesses weren't relying on Google enough, the company took a major step toward gaining even more dependence from businesses this week with the launch of general availability of Google Compute Engine. It's been called part of Google's "quest to dominate the world and its entry into a "heavyweight competition in cloud computing". One thing's for sure. Google is courting businesses like never before.
Are you interested in running your operations on Google Compute Engine? Let us know in the comments.
Compute Engine is part of Google Cloud Platform. It enables businesses to run large-scale workloads on virtual machines utilizing Google's own infrastructure. And that's a powerful infrastructure.
“You now have virtual machines that have the performance, reliability, security and scale of Google’s own infrastructure,” as Greg DeMichillie, director of product management, puts it. It includes thousands of miles of fiber optic cable. Data is automatically mirrored across storage devices in multiple locations.
That's the same infrastructure that lets Google return billions of search results in milliseconds, serve 6 billion hours of YouTube video per month and provide storage for 425 million Gmail users.
Google is now competing directly with Amazon Web Services and Microsoft Windows Azure, among others.
Google introduced Compute Engine at I/O last year, showing off how an app being used to research the genome that's helping to find potential cancer cures. Our own Zach Walton recapped the presentation:
Under current computational standards, Google pointed out that the research on the Genome Explorer app would take about 10 minutes to find each one match. To show off Compute Engine, they showed the same app being powered by 10,000 processor cores being powered by Google. This allows a match to be made every second.
Compute Engine shows off the potential of cloud computing for research. A match a second wasn’t good enough for Google though and they showed a ticker that revealed there were now over 700,000 cores in Compute Engine. From there, they allotted 600,000 cores to the same genome app. Using that many cores, the app was able to discover multiple matches on a constant basis.
Imagine how that kind of power can help a business scale. It's been working well for companies like Snapchat, Cooladata, Mendelics, Evite and Wix.
Thanks to the efforts of Google and its rivals, businesses of all sizes can get access to this kind of computing power relatively inexpensively. And with Google's announcement this week, its prices just got cheaper. Google has lowered prices for standard instances by 10% in all regions.
All machines types are charged for 10 minutes minimum, and then in 1 minute increments (rounded up to the nearest minute). Here's a look the full pricing chart:
The general availability also comes with a 99.95% monthly SLA, 24/7 support, and support for all out-of-the-box Linux distributions (including SELinux and CoreOS) with any kernel or software (including Docker, FOG, xfs and aufs). Google also added support for SUSE and Red Hat Enterprise Linux (in Limited Preview) and FreeBSD.
“At Google, we have found that regular maintenance of hardware and software infrastructure is critical to operating with a high level of reliability, security and performance,” says VP, Cloud Platform, Ari Balogh. “We’re introducing transparent maintenance that combines software and data center innovations with live migration technology to perform proactive maintenance while your virtual machines keep running. You now get all the benefits of regular updates and proactive maintenance without the downtime and reboots typically required. Furthermore, 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, with others to follow in the coming months.”
“Building highly scalable and reliable applications starts with using the right storage,” he says. “Our Persistent Disk service offers you strong, consistent performance along with much higher durability than local disks. 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. I/O available to a volume scales linearly with size, and the largest Persistent Disk volumes have up to 700% higher peak I/O capability.”
As if the competition for hosting your businesses data wasn't hot enough already, Google with all of its data center might (which spans across the Americas, Asia and Europe) is now making its presence known, and will no doubt take advantage of the fact that businesses are already relying on Google for numerous other components of their operations.
Google has a couple of case studies from customers using Compute Engine here.
What do you think? Sold on Compute Engine? Prefer another provider? Let us know in the comments.