When Sun Microsystems announced its Cloud Strategy, I was one of the few who got really excited about it. One of the reasons was that the soundbites from the Sun had the term “openness” used again and again. It was my strong belief that such an approach could lead to what is known as open federated clouds and play a critical role in keeping the playing field leveled without any monopoly player.
With their emphasis on openness and interoperability, they are helping the idea of Federated Clouds. If we do not push for this idea of Open Federated Clouds, we will end up with a monopoly of one or two providers in the infrastructure space. Such a monopoly goes against the open federated structure of the internet. The very foundation of Cloud Computing is on top of the internet and it is only natural to take the same open federated structure to Cloud Computing also. In this sense, the announcement of Sun Microsystems is exciting and I hope they follow through on their promise.
In their initial announcement, they promised to release public cloud offerings that will compete against players like Amazon, GoGrid, Rackspace, etc.. In fact, the industry wanted a strong player who could offer a stiff competition to Amazon Web Services and help drive the prices down a bit. Sun’s announcement was sweet in this respect.
Yes, Sun is going to release products that competes directly with Amazon S3 and Amazon EC2 services. Their public cloud offering will initially start off with a storage and compute service and they may add more in the future. This service will have general availability sometime in the summer. On the compute service, similar to Amazon EC2, users can run various operating systems like Solaris, Open Solaris, Linux, Windows, etc.. They are touting openness here and any application or service that is set up on top of Sun’s public cloud can be moved to any other provider or, even, private clouds.
Then Oracle acquired Sun Microsystems and it not only marked a sad day for Open Source enthusiasts like me, it also raised questions about Sun’s Cloud strategy. But, while talking to our own Paul Miller, Sun promised that they are committed to their Cloud Computing plans. There was not much movement on this front after that. In fact, I once received an email from my contact at Sun Microsystems asking me if I would like to get a briefing on the up coming Cloud announcement but they went silent immediately after that. When I was in Structure ’09 conference last week, I had a chance to talk to some folks from Sun and I asked them about what happened to their Cloud Computing plans. Their response got me intrigued. They told me that Sun is rethinking their Cloud strategy. They told me that Sun is now discussing whether they should continue with their plans for public cloud offering or focus only on what is known as Private Clouds. However, they put in lot of efforts to emphasize that their engineers are still continuing with their work on their promised public cloud offerings but there is no final decision on the path Sun will take in this regard. I asked them if it has anything to do with Oracle and they replied in negative.
Well, this says nothing about the Sun’s plans but it does offer some insight into their thinking. I could see why Sun is interested in the Private Cloud market but I feel that any such move has the potential to hit them bad on a long term. A public Cloud, as envisioned by Sun, will be good for the marketplace and this dilly dallying doesn’t bode well. Even though they deny that this shift has anything to do with Oracle, it is hard to believe it knowing what Larry thinks about the idea of Cloud Computing. I just hope that Sun will come forward and talk to the community about their Cloud Computing strategy.










Insightful article… the thing is Sun beat Amazon to market with its March 2006 launch of network.com – Amazon EC2’s private beta started in August 2006 but the product wasn’t generally available until October 2008, some 2.5 years after the Sun launch. Unfortunately [for sun] they got the pricing wrong by an order of magnitude ($1/hr rather than 10c/hr) and eventually pulled the service, presumably to revamp it on a more competitive platform (their costs were likely substantially more than Amazon’s 10c/hr).
I agree that it would be a shame if they pulled out of offering a public cloud service – even if it were (slightly) more expensive and targeting enterprise customers I think they’d do OK, but a lot of that depends on the adoption of VMware by hosters… they may well have lost their first mover advantage.
Sam