11 Comments

Summary:

This week’s announcement by Google of its new Compute Engine cloud offering is a big deal, but most commentators are missing the real reason Google will get some stalwart Amazon customers to give Compute Engine a try. Performance, not scale, could be Google’s real differentiator.

speedometer

This week’s announcement by Google of its new Google Compute Engine cloud offering is a big deal, and GigaOM’s coverage to date has been pretty spot on. However, having read the excellent coverage by Om Malik and Derrick Harris, as well as some interesting analysis on other sites (like here and here), I’m stuck with the feeling that most are missing the real reason Google will get some stalwart Amazon Web Services customers to give Compute Engine a try. Google’s quest to win over users will be all about performance.

The Google Developers Blog post announcing the service broke down three key “offers” of GCE,  which I interpret as the three key differentiators from Google’s perspective of its service over the competition (not necessarily just AWS):

  • Scale. At Google we tackle huge computing tasks all the time, like indexing the web, or handling billions of search queries a day. Using Google’s data centers, Google Compute Engine reduces the time to scale up for tasks that require large amounts of computing power. You can launch enormous compute clusters – tens of thousands of cores or more.
  • Performance. Many of you have learned to live with erratic performance in the cloud. We have built our systems to offer strong and consistent performance even at massive scale. For example, we have sophisticated network connections that ensure consistency. Even in a shared cloud you don’t see interruptions; you can tune your app and rely on it not degrading.
  • Value. Computing in the cloud is getting even more appealing from a cost perspective. The economy of scale and efficiency of our data centers allows Google Compute Engine to give you 50% more compute for your money than with other leading cloud providers. You can see pricing details here.

Scale and price matter, but …

A demo of genome analysis on Compute Engine.

All of the coverage I’ve read to date has focused on the scale and value elements of Google’s story. And these are critically important. When it comes to scale, few can match a launch that includes roughly 100,000 servers and 770,000 cores of available capacity (though I doubt you’ll be able to grab half that for yourself in a few weeks). Google does scale for a living, and with a reported 1 million servers in operation across the company, nobody comes close — except maybe Amazon.

When it comes to value, Google fired a shot across AWS’s bow with roughly equivalent pricing, though Google argues that app-for-app, their service will be cheaper—or, as it says, “50% more compute for your money than with other leading cloud providers.” Other analysis questions this, and certainly any advantage Google has today will be challenged quickly by Amazon, given its own history of reducing prices.

Performance is Google’s ace in the hole

However, I think the key potential differentiator for Google, if it wants to take market share from AWS, is the performance of the cloud infrastructure itself. If Google can deliver a service that eliminates most of the I/O and network performance inconsistencies that AWS customers currently experience, I can guarantee you there are many major compute customers of AWS that will want to give Compute Engine a test run.

The I/O experience alone has been a real thorn in the side of many technologists, who — despite having designed applications to account for performance inconsistency — have real concerns about whether their applications can run as efficiently as possible in such an environment. If Google’s performance claims are confirmed, you will see one or two large-scale AWS customers begin to spread their compute loads between the two services by the middle of 2013. Heck, at least one may actually move off of EC2 altogether.

Still, if performance turns out to be not significantly better than AWS, or if there are other major limitations to the Compute Engine service, those customers will quietly put to rest their experiments and return to the “tried and true” AWS service that they know. And others will no doubt use Compute Engine, even if it’s not the best thing since sliced bread, turning it into a real contender in the cloud computing space but nothing near an AWS killer. The scale and value stories are legitimate, they’re just not compelling enough by themselves to drive too many customers to change how they do cloud computing.

Feature image courtesy of Shutterstock user David Castillo Dominici

You’re subscribed! If you like, you can update your settings

  1. Daniel Lopez Sunday, July 1, 2012

    It is smart for Google to aim for the cream of the crop of AWS customers first. It both makes an statement and has a chance to hurt them in a high-value customer segment. However, long-term success will require Google building a full-fledged ecosystem which is much harder to do. We live in interesting times :)

  2. I agree. Amazon has built a suite of products, reducing complexity and lowering the barrier to rapid development and deployment of services in the cloud. E.g. RDS, SQS, Route53, SimpleDB, DynamoDB, Elastic Map Reduce, to name a few. Google has an opportunity in front of them, if they are able to provide complementary services and features, similar to what Amazon has done. EBS is Amazon’s Achille’s heel and will be for some time, given the insidious dependency so many AWS customers have on it. Since the courts have ruled in Google’s favor that the ‘name’ of an API, in and of itself, is not a violation of copyright infringement, providing a set of compatible libraries and tools to aid in migrating existing AWS customers would be a big win for those who are to deeply dependent on the AWS platform to move.

  3. Abhijeet Kumar Sunday, July 1, 2012

    I thought Compute Engine at its launch caters only to compute-intensive workloads. Where did I/O performance and Google compute engine meet for this article to be written? I have found using AWS, there is plenty of options that I get to store and retrieve my data on-demand. Is there any technical proof reading of articles here? Has anyone seen the page load speeds of Pinterest? Compare that to Facebook or Youtube.

    1. This is an accurate comment. Tuning is important.

  4. This is in private beta- like hiring an amusement park for a party, result: no lines for the best rides. Google still needs to demonstrate they can handle scale as it happens in a true commodity cloud style environment. They haven’t proved that thus far. I bet they can bring some great engineering to this game, but I think its too early to say we’ve seen evidence of that. Also, a lot of performance in cloud is in how you squint at it, is it CPU, IOPS, open pipe that you need? Every provider seems to have an organically enforced niche they excel at, and areas they are complete crap at. I have no doubt GCompute will be the same.

    that said, they are positioned to pull the developers bees out of the AWS cloverfield in ways nobody else is. Be interesting to see if they can deliver when this hits the open market.

  5. Robert Kraig Sunday, July 1, 2012

    I doubt that googles product offering will be environment agnostic. I think they will release some tools right away in mostly java & python oriented and the rest of the environment will be left. You will either use their specific product offerings or you will probably end up having to stay with AWS. Google has a lot of proprietary products which they will probably offer access to, but you will probably have to do it the “google” way. Some will argue that the google way is the right way, and others will say that flexibility in their own view is more important. I think that unless google offers an identical product offering, in terms of alternatives to mysql/postgresql/elastic-search-ebs/load balancers/etc that people won’t have the time or money to translate their existing stack to fit the needs of the compute engine environment. Also my guess is that the compute engine is mostly just a much more scaled out App-Engine environment.

  6. Shouldn’t the title of this article be “Why higher performance could help Google steal cloud customers from Amazon”?

  7. Sebastian Meyen Monday, July 2, 2012

    I agree.

  8. James, as always an interesting and thought-provoking opinion. But I do not see any evidence that Google is going to pull a rabbit out of their hat and – ta-da!! – market-winning performance!

    For a long time I have advised anyone who would listen that you cannot compete in public cloud markets on a par with AWS and win. You need to do more to win. This is borne out by our customers in the cloud provider markets who are differentiating with value-added service through better security, broader compliance, enterprise-style SLAs, greater flexibility, more heterogeneity, and other elements that large enterprises demand. (More on this in my upcoming July vid-cast of CloudViews Unplugged – http://www.youtube.com/user/CloudCommons)

    So on the premise of competing and differentiating against AWS, I agree with you. Google has to do something different – and better – to win against AWS. Offering better performance would be a small step, but perhaps important enough to win share.

    But in Google’s history there is nothing that shows me they will be instantly able to deliver better performance than AWS. This IaaS thing is new territory to them. They bring no apparent experience of running other people’s workloads at massive scale, or of assuring a mission-critical service in a way that is clearly intentional and reliable. This is a huge learning curve, and you don’t get there just by being big; you get there through years of experience in applying performance management discipline to diverse and complex mission-critical workloads. Where is Google getting that from?

    So like many other commodity cloud services, I am more inclined to believe Google will attract great business from SMBs, and perhaps a few growing enterprises (just as I discussed in my recent blog, on why public cloud is a big fat enterprise fail – http://community.ca.com/blogs/perspectives/archive/2012/05/29/why-the-public-cloud-is-a-big-fat-enterprise-fail.aspx).

    But I don’t see better performance as a fait accompli; and even if they do deliver better performance, a truly competitive public cloud will need a whole lot more to win.

  9. I wrote up an analysis of the hardware that is *probably* more accurate than the one from ExtremeTech: http://www.autonomoussystem.net/2012/06/770000-cores-analysis-of-google-compute.html

  10. timfreemandotorg Tuesday, July 3, 2012

    The tech lead for Google Compute Engine comments here about I/O performance: http://www.peakscale.com/noisyneighbors/#comment-575367712

Comments have been disabled for this post