AWS worry misplaced: Welcome to the maturing cloud

AWS worry misplaced: Welcome to the maturing cloud

Summary: Amazon's earnings spurred a flurry of worry about Amazon Web Services' growth. The larger takeaway that the cloud service game is maturing.


The angst over Amazon Web Services and the effects of its price war with Google and Microsoft Azure to offer cloud infrastructure as a service may be a bit overdone. AWS isn't in a fight for its life as much as the company is navigating the maturation of the cloud services market.

Here's the storyline from last week's Amazon earnings conference call. Second quarter "other" revenue---mostly Amazon Web Services (AWS)---only posted growth of 38 percent from a year ago in North America. Sequentially revenue for AWS fell. Other revenue fell well short of expectations.



Why? AWS was cutting prices and for once it wasn't driving the race to the bottom. Google was. Microsoft Azure was in the mix too. IBM Softlayer may have also played a role. Think of the hyperscale infrastructure as a service providers as the airline industry. The players will cut prices to put fannies in the seats and eat up capacity. Ultimately, money will be made with additional services.

Special Feature

The Art Of The Hybrid Cloud

The Art Of The Hybrid Cloud

Cloud computing is insatiably gobbling up more of the backend services that power businesses. But, some companies have apps with privacy, security, and regulatory demands that preclude the cloud. Here's how to find the right mix of public cloud and private cloud.

Of course, this hyperscale game isn't going to be supremely profitable when there's a land grab going on. Luckily for Amazon it has put off really being profitable since inception and Wall Street has largely played along.

The other wrinkle in the "AWS is in a big battle" storyline was the hybrid cloud. As noted by GigaOm, companies like Dropbox and many enterprises aren't solely going to depend on the public cloud. Given the percentage of companies going solely public cloud is low it's a bit of a stretch to think that hybrid clouds are hurting Amazon. Meanwhile, companies like NetApp are plugging their infrastructure into AWS. Co-location providers such as Equinix are also connecting to AWS. These moves only expand AWS' market, but the maturation of the cloud also means that NetApp will have more connections to Microsoft Azure by the end of the year.

So yes, the cloud price wars have hurt AWS, but the dynamics are a bit more complicated.

Amazon CFO Tom Szkutak said:

We had very substantial price reductions for customers starting in second quarter. They range from 28% to 51% depending on the service. But AWS continues to grow very strongly. In Q2, we had usage growth close to 90% year-over-year for the quarter. So the team's doing a fantastic job, and we put in some price reductions that were very substantial that are saving customers hundreds of millions of dollars over the next several months as we mentioned 90 days ago. But again, we love that business.

And guess what? The price cuts have probably hurt Google and Microsoft too, but their cloud businesses aren't material enough to move the margin and revenue needle much---yet.

Should we worry about AWS? Probably not. AWS is the big dog, had and open field and doesn't anymore. The laws of large numbers are catching up to AWS somewhat. What's really happening is that the cloud market is maturing.

Consider the following moving parts:

The hybrid cloud is the way in for the big IaaS providers. Enterprises are all building out various use cases for public cloud providers, but aren't going to toss out their existing infrastructure. Even Amazon isn't completely on AWS as CTO Werner Vogels noted in a recent interview. Given that reality, the game is to forge partnerships and be a player in the data center ecosystem. AWS and Microsoft are ahead of the pack here. IBM can tether Softlayer to its infrastructure stack. Google is playing from behind on the partnership front.

There will only be a few hyperscale players. It's worth noting that NetApp has segmented cloud players in buckets. Hyperscale includes AWS, Google and Azure as three big horses. Cloud service providers include vendors like Verizon Terremark and Rackspace. All of the cloud service providers are important in the stack, but the price wars will largely be in the hyperscale market.

netapp azure1

No cloud buyer will be on one service. Sure companies have grown up on AWS, but at some point the economics favor owning infrastructure. Or even if an enterprise is totally public cloud happy, it will ultimately use a series of service providers. AWS was the only game in town for years. Now there are a few other options. Buyers will diversify workloads because it's just good business.

Instagram finally picked up and moved from AWS to Facebook

Bottom line: The reality that price wars hurt isn't news. And most of the tech industry knows that cloud infrastructure pricing is going to near zero anyway.

But AWS' ability to roll out new features and services (250 this year to date) means it will be able to move up the cloud stack, find new revenue streams, play with the hybrid cloud and adapt with its installed base. Microsoft also has that installed base advantage as does IBM. Google needs to poach reference accounts and forge a lot more partnerships with hardware vendors like NetApp to really threaten AWS.


Topics: Cloud, Amazon, Data Centers

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Wall Street Losing Game

    Face it, ya'll are trying to apply Wall Street & IT analytics to a firm that doesn't operate that way and doesn't answer to investors and operators with those expectations. After getting it wrong for twenty some odd years, ya'd think you woulda larned by now. Guess not. I don't expect the learned consultancies to get it right. The blinders are surgically implanted on hiring. Bezos only needs to grow clients while not losing too much across the firm, everywhere. Actually, so long as everyone else is keeping that IT plant running for the other 47 weeks of the year he doesn't need it going full-tilt, he's got it made. ANY increases in capacity and mild profitability is all net. On the other hand, being an economist and engineer here, IBM, Google, and most especially Microsoft answer to a much less discerning crowd of investors. [I need to lay in some more popcorn. This'll be fun. Amazon subscription, natch.]
    Brian J. Bartlett
  • "Any Cloud" is the strategy of the future

    Innovative companies increasingly rely on multiple clouds and on-premise infrastructure to guarantee the highest performance and availability for their solutions. GoGrid (disclaimer: I work at GoGrid) saw the trend toward any-cloud flexibility early on across its customer base, and realized that today’s internet-of-things world, infinite application combinations are a foregone conclusion so the concept of one cloud fits all would quickly become a thing of the past. Now, fluid movement across numerous partner technologies are a must to create the utmost flexibility for enterprises to port not just across clouds, but seamlessly move across in house infrastructures, and spin up or down within a matter of minutes on any number of technologies, big data or otherwise.

    Light Reading explained it well in a recent article, saying this holds true for telco providers in particular: “Telecom cloud service providers have been trying to develop this capability themselves so they can offer business customers the ability to stand up a whole range of applications in the cloud without individually engineering each one through a complex process.”

    Enterprises are no longer constrained. Our customers have the option to keep some portion of their apps on AWS, Google, In-House, wherever they already sit – then move them around as needed. This ensures reliability and performance using multiple cloud platforms, including large commodity clouds and private clouds in on-premises data centers. More importantly, it makes high-availability environments a reality for enterprises, not depending on a single provider, so companies can focus on their core businesses.
  • AWS Worries

    I think that the IaaS market is very similar to the current hosting market. There are a number of larger players that will continue to draw a huge amount of clients, but they are always going to be playing defense against companies with better service and more tailored hosting solutions. Right now AWS obviously has Google and Microsoft to think about, but the smaller providers are also impacting their business. I think when you look at providers like DigitalOcean, Linode, and, we’re definitely seeing companies that can impact the bottom line of places like AWS over the long haul.