Why VMware Needs A New Direction
In my last article, I analyzed the real motivation behind the VMware’s recent intention to acquire Nicira. In this article, I am going to review VMware’s past strategies and predict its long term strategies. In short, VMware’s past growth strategy is “vertical,” and its future growth strategy should be “horizontal.”
Past Strategy Review
Since 2008 VMware has been aggressively seeking to grow up the IT stack into middleware and application space (in cloud term, PaaS and SaaS). That is basically Microsoft strategy which had been extremely successful for Microsoft to grow from Windows operating system to middleware/tooling, and Office suite. As today, each of these businesses represents 1/3 of Microsoft’s overall revenue. Given top exectutives’ long time at Microsoft, it’s no surprise to anyone that similar strategy was adopted at VMware.
As part of the vertical growth strategy, VMware acquired SpringSource (the company behind the popular Spring Framework in Java developer community) in 2009, followed by a series of smaller acquisitions like GemStone, RabbitMQ, etc. Mostly open sourced, these products and the in-house developed database based on Postgres consist of a complete offering of middleware, even though lots of works are still needed to unify these products for better integration and consistencies. If you’ve worked in IT industry long enough, you know it takes time and many of the key talents may be gone within years. That is what’s happening with SpringSource including its founder Rod Johnson, COO Rob Bearden who is now CEO of Hortonworks, VP Engineering Peter Cooper-Ellis who is in same role at Cloudera.
On the application side, VMware bought Zimbra from Yahoo for a discounted price, and then a few startups companies like SlideRocket, Socialcast, Digital Fuel. The level of the products is at application level but really delivered as cloud services via Web. At one point, I was wondering VMware would build a SaaS version of Microsoft Office, therefore Web based spreadsheet, word processor would be its next acquisition target. Somehow it didn’t happen, at least not yet.
Three years after the SpringSource deal, we haven’t seen the magic happened in terms of rapid revenue growth. VMware’s huge success on virtualization did not warrant its success on the middleware and application space. Although VMware has won the trust of system administrators, these system administrators are not really the target audiences for the middleware and application products. VMware definitely knew about it, and that is exactly why it paid good money ($362 million in cash and equity and $58 million of unvested stock and options) for SpringSource in hope to boost its mindshare with developer community. The obstacles seem just too big to overcome in short time.
To be fair, both the middleware and application spaces are tough businesses competing with not only Microsoft but also other big names like IBM, Oracle, Redhat, not to mention many open source options out there. Beyond that, technically speaking the virtualization technology does not give VMware much differentiation in supporting its middleware and application because the hypervisor interfaces are simply not as sticky as the Windows APIs. In other words, VMware does not have much leverage with its leading virtualization products.
As of today, if you ask around about VMware, most people still equal it with virtualization, not much association with other products in its rich portfolio.
Future Strategy Prediction
While the vertical growth strategy has not yet succeeded, VMware’s core server virtualization is increasingly challenged by Microsoft’s Hyper-V and open source XEN and KVM. Any mistake on the virtualization could be disastrous for VMware as it’s still the only cash cow in house. So VMware must solidify and enhance its core virtualization.
After the “software driven data center” last year, VMware finally got it right with “software defined data center” this year. The one word difference led to the recent acquisition of Nicira, which is a leader in the “software defined networking” space. Maybe the SDD was inspired by SDN?
Anyway, the acquisition is the first step for VMware to grow horizontally in the data center infrastructure software. In some way, you can call server virtualization “software defined compute.” After the Nicira deal, you can naturally guess the next big acquisition should be “software defined storage,” whatever that is. I will explore the SDS later on.
That is not the end of the story. With the building blocks of SDC, SDN, and SDS, VMware should then move up the management stack (not the runtime stack as before. Difference? see my previous article on two stacks) to build a suite of management tools that covers different aspects of the data center management and operations including automation, security, chargeback, DR, backup, etc.
With the horizontal growth, VMware will have better integrated infrastructure software that covers all the aspects of a data center – compute, network, storage, and management. This new growth strategy will have far better chance to succeed because it can leverage not only VMware’s existing user base, but also its strong engineering expertise in infrastructure software. More importantly, these are relatively new areas without established players yet.
Again, nothing is easy to grow a technology company especially in size like VMware. More important than the strategy (do the right things) is the execution (do things right) which ultimately determines whether acquisitions worth big money. With a right strategy in place, same investment can yeild multipled result.
There were rumors around whether VMware would spin off a new company focusing on cloud. I will discuss it in a separate article later. Stay tuned.