Blog Archives

Yesterday Microsoft CEO Steve Ballmer gave a talk on Cloud Computing at the University of Washington, Seattle. Check out the video of the entire talk here. Let me briefly highlight the key points from his talk and then offer my perspectives on it. He was highlighting five key dimensions of the cloud

• Cloud creates opportunities and responsibilities: He was highlighting how cloud computing is opening up various business opportunities for people to take advantage without any huge investment. He was talking about new kinds of businesses that are possible only due to clouds. He then went on to highlight the responsibility of cloud providers in protecting the privacy of user information and data.
• The cloud learns and helps you learn, decide and take action: Here, he was clearly coupling Semantic Web technologies with the cloud computing and argued how the feedback loop from the data could help us do better things. He brought in Blaise Aguera Y Arcas from the Bing Map team to show the recently announced new features.
• The cloud enhances your social and professional interactions: He was highlighting how cloud computing has drastically changed the way we interact with our professional connections, family and friends. He brought in Simon Atwell from XBox team to demo how they have partnered with Sky Network in UK to bring in a social approach to watching television. The point was cloud computing makes it very easy to socialize without moving the butt from the couch.
• The cloud wants smarter devices: This topic was meant to show how cloud computing has enabled mobile devices to do things that were not possible in the past. He was showing how smarter devices of all form factors has made the traditional PC somewhat redundant. He then spoke about the capabilities of upcoming Windows Mobile 7.
• Cloud drives server advances that in turn drive the cloud: Here he was talking about the advances made in hardware and software that makes cloud possible for the kind of scale and reliability needed to manage the amounts of data processing we do. He talked about how Windows Azure has made software development easy for developers. He also harped a bit on Private Clouds and used the idea of federated models to justify their Private Cloud push.

Basically, it was a great sales talk on cloud computing and on Microsoft’s plans to take over the cloud world. I will add some of my thoughts watching this talk and what it means to cloud computing from my perspective.

• The most important takeaway from his talk is that Microsoft is damn serious about cloud computing and they are going to go all out to capture the market. In the past, I have doubted Microsoft’s chances in the cloud and was even critical of their perceived lack of seriousness. With this talk, Microsoft has put an end to any such doubts. Steve Ballmer claimed that already 75% of their workforce are working directly on Cloud Computing related projects or on projects inspired by the Cloud. He said this number will raise to 90% in 2011. It definitely underscores their seriousness.
• Essentially, the talk is a continuation of Microsoft’s (well, Ray Ozzie’s) push for three screens strategy. They want to be at the mobile, PC and the big television screens. Their push is more about their existing OSes and applications with cloud providing the necessary backend to store data. In fact, I also care only about my data in the cloud. Ideally, I would want my data sitting in the cloud and I should be able to access it from any device I want and from anywhere I want. However, the rich interface approach of Microsoft and Adobe poses some problems compared to the browser based approach of Google and other SaaS players. With Microsoft’s approach the traditional applications will have more features than their browser versions. This restricts users to only those operating systems and devices supported by Microsoft. Just take the example of the video of Steve Ballmer’s talk. You need silverlight to watch the video and it doesn’t work well with Google Chrome browser. Microsoft’s point of view of Chrome is that it doesn’t merit their attention due to the lack of marketshare at this time. The rich interface approach of Microsoft and Adobe using proprietary technologies are very restrictive and it doesn’t augur well for either cloud computing or the tech market. With an open standards based browser app strategy, users are not restricted to any specific device or operating system. Just take a look at how Google is overcoming the restrictions of Apple using mobile webapps. It is just a matter of time before these browser based apps match with the traditional applications on the richness of the interface.
• Looks like Microsoft is going gaga over their Powerset acquisition and the push for Semantic Web ideas by Ballmer is a clear indication of where Microsoft wants to go in the future. They want to make it easy for users by using these technologies to make decisions on their behalf. I am really impressed with the new features released for Bing Maps. I really want a strong competition between Microsoft and Google on this front (Google services vs Live services) so that we, the users, gain ultimately. This competition is definitely going to get interesting.
• Microsoft is embracing cloud computing deeply as a part of their mobile strategy. Even though I agree with them philosophically, I am still not convinced that they can gain big marketshare with Apple and Google having a strong foothold there.

Overall, it was good to see Microsoft pushing cloud computing even if it was skewed to fit their strategy. I have told many times that Microsoft is an important player in the cloud space. It will be interesting to watch if they can really catch up on the consumer side even though they have a better chance on the enterprise side.

 

 

 

 

 

Image via Wikipedia

Yesterday Dries Buytaert, original creator of open source content management software Drupal, wrote a blog post highlighting how open source can play a major role in enterprise SaaS. In his post, he points out to SaaS being dominated by proprietary players and talks about how open source can play a significant role in avoiding vendor lock-in.

At the same time, I think we have an opportunity to go beyond that, and to redefine the Software as a Service model based on Open Source values, almost exactly like we started doing 10+ years ago with off-the-shelf software. Almost all Software as a Service providers employ a proprietary model — they might allow you to export your data, but they usually don’t allow you to export their underlying code. While a lot of these services might be built on Open Source components, they have a lot more in common with proprietary software vendors than Open Source projects or companies.

There is room for Open Source companies to disrupt this model, and it is probably not something that can be done without the help of Open Source companies.

This is clearly a call for open source vendors to wake up and dominate the SaaS space. However, some of the bloggers have interpreted it as a criticism of SaaS by Mr. Buytaert. Part of the reason could be due to the binary approach taken by many in the tech field in the open source vs proprietary debate. They see it as open source or proprietary but in a truly open society, it should be open source and proprietary giving the customers choice to decide. 

With this in mind, let us try to understand the dynamics at the interface of open source and cloud computing worlds. There are two schools of thought regarding the role of open source in the cloud based era. One school of thought lead by Tim O’ Reilly calls for open standards, open protocols, open architecture, etc. with the claim that architecture trumps licensing anytime. However, the other school of thought, to which I belong and promoted by many like Simon Wardley of Canonical and Sam Ramji of Sonoa Systems, puts emphasis on open source along with open standards and protocols. I have emphasized this point many times here in this blog including this post and my talk at Cloudslam ’09. I really don’t see a diminished role for open source in the cloud based world. From Mr. Buytaert’s post, it is amply clear that he is firmly in the latter camp. He also believes in the SaaS as the future of how applications are consumed and he strongly advocates a role for Open Source in the SaaS world.  

Let us face it folks. Even in the traditional desktop computing world, open source didn’t gain much traction on the consumer side except for some of us who used Linux and other open source software religiously. Ultimately, open source attained its nirvana in the enterprise. Apart from open source fanatics like us, whatever little traction open source got among the consumers and small businesses were due to the lower cost or, rather, no cost of many of these products. It is only the enterprise that found the real value (both economic and the importance of openness) of open source. 

In the SaaS world, it doesn’t make sense for consumers to expect the source code (not including the philosophical considerations regarding openness by some of us) because it will cost them more to host the code themselves than use the services of proprietary SaaS vendors. It is only in the enterprise SaaS space, we can expect the users to derive the full benefits of the openness. This opens up a great monetizing opportunity for enterprise open source vendors. By releasing a SaaS version of their products, they can gain traction among the enterprise customers. As this trend progresses further, we may even see the adoption of open source SaaS trickling down to small businesses first and, eventually, to consumers. But the focus should be on the enterprise SaaS space.

In short, instead of debating whether Mr. Buytaert was critical of SaaS or not, let us get the open source vendors embrace SaaS overwhelmingly and offer open source alternatives to the proprietary SaaS offerings. A real free market approach is one where customers have the choice to decide on what they want.

For a long time, I am debunking the idea that the cloud infrastructure market will consolidate and we will end up with a handful of monopoly players. In my opinion this is short sighted idea and it is due to the lack of understanding of the diverse needs of users around the world. One way to counter such a consolidation is by supporting open federated clouds, an idea advocated by Tim O’ Reilly. Off late, we are seeing the idea of regional clouds gaining traction in the market. The adoption of regional clouds is driven by the fact that, in some cases, customers need more than just the cloud economics. They want to know where their data is located and where their compute power is coming from. They also want to have a direct interaction with their cloud provider, something that is not available with the giants like AWS or Google. In US, Minnesota based ReliaCloud is the poster child for the so called regional clouds. There are many such regional players, mostly from the traditional hosting world, who are repositioning themselves as cloud providers. One such player from Germany is ScaleUp Technologies. Recently, I had a chance to talk with their CTO, Christoph Streit, and Kevin Dykes, their Director of Business Development.

ScaleUp Technologies is based on the premise of Open Federated Cloud ecosystem. They are a spin off from the German webhost called Internet4you. They have built their cloud architecture on top of 3Tera’s Applogic product (see our coverage of CA’s acquisition of 3Tera). As it is the case with smaller cloud providers, they have two datacenters in Germany catering to their cloud business. However, the fact that their cloud is built on top of Applogic allows for federation and they have partnered with XSeed Co. Ltd., a Japan based cloud provider also built on top of 3tera’s Applogic platform. This partnership allows ScaleUp to let their customers tap into XSeed’s infrastructure (and vice versa) right from the their UI. This is a perfect example of an open federated cloud ecosystem in action, albeit a smaller one.

ScaleUp’s cloud offerings comes in three different flavors

• Self Service Model: This is similar to Amazon Web Services and other cloud providers. However, ScaleUp’s offerings differ from AWS’ offerings in terms of certain additional features like availability of templates based on software stacks (for example, LAMP stack). These readymade templates makes it easy for the customers to get their infrastructure running quickly. Similar to AWS and other public cloud providers, load balancing and auto scaling are part of their cloud offerings
• Virtual Server Infrastructure:  This is similar to Virtual datacenters with a collection of large number of servers like web servers, database, servers, etc.
• Private Cloud: Single-tenant cloud infrastructure for customers who want complete control over their perimeter

I asked them about the European cloud market and what kind of traction they are seeing in the market. Christoph told me that they are finding it difficult to convince customers to move their assets to cloud. He told me that they have to put considerable efforts to educate customers about the benefits of cloud computing. In his opinion, European market is lagging behind US market by at least a year on cloud adoption. I also asked him about their support for open standards and he told me that he checked OCCI initiatives sometime back but they are not ready enough for them to embrace just yet. He said they are not participating in any standards initiatives at present but it may change in the future. In my opinion, it is important for such small cloud providers to be part of the initiatives around open standards. I feel that these smaller players have to adopt open standards to fend off competition from bigger players and, also, for the very success of the open federated ecosystems.

The idea of regional federated clouds is slowly taking off. With more and more traditional web hosts repositioning their strategy towards clouds, we are going to see further increase in the number of regional cloud players using the federated model. More importantly, regional cloud providers like ScaleUp Technologies are crucial to raising the awareness about cloud computing in different parts of the world and in ensuring more widespread adoption.

And all this leads us to today’s exciting news: we’ve just been acquired by Google! What does this mean for Picnik? It means we can think BIG. Google processes petabytes of data every day, and with their worldwide infrastructure and world-class team, it is truly the best home we could have found. Under the Google roof we’ll reach more people than ever before, impacting more lives and making more photos more awesome.

• This is an awesome news for the Picnik team who recently announced the one billionth edited photo and this acquisition only confirms the awesomeness of their product
• Picnik already has a deal with Yahoo’s Flickr. I am sure this move will push Flickr away from Picnik and, possibly, more closer to Microsoft
• Even though Google is a major SaaS player, they don’t have a browser based photo editing software. Their existing desktop application goes against the idea of relying only on SaaS. With Google’s push for Chrome OS, it is only natural that they have a browser based photo editorOtherwise, it will be difficult to gain traction for ChromeOS with consumers without relying on a third party vendor. With Google push against Flash, there is no way they will want to partner with Adobe to offer an online photo editor
• One of the biggest attractions for me was their browser plugin which lets anyone take any image from any website, edit and save it on their computer or one of the online photo sharing sites. I hope Google will keep this extension as a part of their plans
• I am hoping that Google will not charge for the pro features and let paid customers of Picasaweb use it for free

Image via Wikipedia

Yesterday CA Inc., the enterprise IT company formerly known as Computer Associates, announced its intent to acquire the cloud computing provider 3Tera. Even though the details of the deal are not available, there are some unconfirmed reports going around the blogosphere and twitterverse saying it could be 30x the current 3Tera revenue. This acquisition follows CA’s other acquisitions relevant to cloud space like Cassatt, NetQoS and Oblicore. This news has the potential to shake up the enterprise cloud computing market.

Traditionally, CA is a leading player in Enterprise IT Management software for mainframes and distributed systems. After the controversy surrounding the indictment of their then CEO, Sanjay Kumar, in the mid part of last decade, they put their house in order and started pushing products around “Lean IT” campaign. As cloud computing gained steam towards the end of the last decade and the recession forced the enterprises to take a second look at the cloud for adoption, CA was busy acquiring companies like NetQOS, Oblicore and Cassatt. But there were no telltale signs of any immediate cloud play. It appears, with all these acquisitions, CA was putting the necessary pieces of puzzle in place to take off as a strong and credible IaaS player. 3Tera’s acquisition gives us some hints about what is going to come out of CA in the coming years. It also sends a unambiguous signal that they are in for a long haul.

CA’s path to cloud nirvana through acquisitions is hardly surprising because CA is, in fact, built by a series of acquisitions from 1977 onwards. Already the blogosphere was buzzing around with the news that CA plans to spend $300 million buying cloud startups. So, their strategy to buy 3Tera is not at all surprising. I will briefly analyze the impact of this acquisition.

• First and foremost, this deal firmly puts an end to the spin that cloud computing is all hype and a marketing term. This move by a traditional IT vendor is a clear message that cloud computing is the future for the enterprises. We are soon going to see more and more enterprise customers jumping into the cloud bandwagon.
• This could also be considered as a beginning of consolidation in the infrastructure space. Even though I don’t believe in the idea of monopoly of handful of infrastructure players, a shakeup through consolidation will always be there in a maturing market. This just shows the maturation of the cloud computing market leading to more and more enterprise adoption.
• This also puts CA as one of stronger contenders in the Cloud infrastructure marketplace. It is like a parachute for CA to jump right into the middle of this competitive marketplace without any hard work. Well, we cannot consider CA to be an innovator like other cloud startups in the market and the only way for a titan from the old world to get into the mix of the new generation is by acquiring a hot company like 3Tera.
• This is also a good news to other small players in the industry. It legitimizes their approach and offers them hope that someone else might come knocking on their door pretty soon. Overall such a consolidation advances the cloud marketplace to the next level.

Even though it is exciting from many angles, I do have some concerns about this merger.

• When I think of this acquisition of 3tera, the first thing that comes to my mind is “why CA?”. Personally, I strongly believe that a titan from a previous generation cannot innovate in the next generation like some of their newer counterparts. I am sure some of the fellow members of Clouderati will vehemently criticize this belief but I think of it as IBM in the desktop world dominated by Microsoft. They can do some interesting things but the innovation vigor will be lacking. Frankly, I just don’t like a old world enterprise guy messing around with clouds.
• Another important concern is what will happen to Applogic? Will it continue to be an independent product? From all the indications in the press, I get a feeling that it is going to be tightly integrated with CA’s existing product lines. If it happens, some of the existing 3Tera customers, especially the small and medium sized ones, will be left in a lurch. A move away from the current standalone version of Applogic software could be disastrous for these 3Tera’s clients. I spoke with one of their clients today and asked them about their strategy in the aftermath of this news. They told me that they are cautiously optimistic and their sources in 3tera has given them assurances about long term support for Applogic. Realizing this predicament of 3Tera’s customers, their competitors have already started to fish in the troubled waters.

It will be interesting to see how it plays out in the long term. It will also be interesting to see who is the next target for CA with the rest of $300 Million they have have in the pockets.