2009 was a great year for the virtualization industry. We saw milestones such as the releases of VMware vSphere4 as well as more palatable virtual desktop infrastructure (VDI) solutions emerge (conveniently timed around a Windows 7 launch, no less). My predictions for 2010 include a mix of both virtualization-focused topics as well as predictions for the cloud computing market.
1. VMware will add one of its ancillary products (e.g. ConfigControl, AppSpeed, etc.) to the Enterprise Plus SKU at no additional cost.
VMware has a stable of complimentary vCenter products, some arguably better than others. With Chargeback, CapacityIQ, AppSpeed, et al., VMware can provide a fairly robust framework for the enterprise.
There was a lot of grumbling when VMware announced the full gamut of supporting vCenter products that were previously supplied by the VMware partner ecosystem; some partners saw their business potentially waning. Also surprising is that a good portion of the products were made in-house without spending acquisition dollars. My pick is that either Lifecycle Manager (LCM) or AppSpeed get rolled into the Enterprise Plus licensing. This may be a bit of Christmas wishing, but I do think it makes sense for VMware to either start attacking LCM or AppSpeed from a sales perspective, or give the goods away. Both products are fantastic when implemented correctly.
2. Dell to further its relationship Xsigo and simultaneously push agnostic 10 Gigabit Ethernet (10 GbE).
It appears as though the ability to provide a building-block platform for future virtual infrastructures is an important capability (enter Cisco Unified Computing System, or UCS). In order for Dell to build a Cisco UCS rival, it will need to equip its PowerEdge servers with a converged network adapter for connection to a unified fabric.
Dell partnered with Xsigo back in February and is also able to sell Xsigo off the Dell General Services Administration schedule to the public sector. A furthering of this partnership would also require a management and automation component (similar to a BMC offering). Hewlett-Packard's (HP's) pending acquisition of 3Com will give HP some serious switching firepower. Dell will need to ensure that its well-loved PowerEdge servers can play in the unified fabrics of the near future. In addition to pushing the Xsigo solution, Dell will continue the agnostic, "we don't force vendor lock-in," line as they continue to push 10 GbE as the storage platform of choice. Still, without the automation secret-sauce, Dell's stack is still just a collection of gear.
3. VMware Cloud Express will finally leave the station, but there won't be many on the train.
There has been much talk about VMware's vCloud Express since the launch at VMworld 2009. If you go to the official webpage for vCloud Express, you will see that all of the listed vendors are showing, "this service is currently in beta."
I think it's a minor travesty that Microsoft's cloud will be in production likely before a VMware vCloud Express cloud, as VMware is clearly the leader in server virtualization (a component responsible for ~70% of what cloud computing entails). Granted, this has little to do with VMware, as VMware is merely providing the platform and the advanced programming interface (API) set. Cloud service providers such as Terremark and Hosting.com need to up their game, and probably should have quarters ago.
VMware vCloud Express will be a success when:
- I can swipe a credit card and be online in a reasonable amount of minutes
- I can VMotion between a vCloud Express provider and my own data center
- I can pair up Site Recovery Manager with a VDC living in a vCloud Express environment
4. Microsoft Azure will debut an Infrastructure-as-a-Service (IAAS) offering, impacting Amazon Elastic Compute Cloud (EC2) sales.
Amazon EC2 is regarded by most to be the current cloud king. Amazon is responsible for the eye-opening of corporate America to the feasibility of the cloud technology, a way it can be offered, a price structure (to pilfer) and some verticals in which to sell. The juggernaut that is Microsoft has deep pockets, however, and perhaps more importantly, a potent salesforce. With Office 2010 offering a click-to-run delivery mechanism (provided by App-V) and Business Productivity Online Suite (BPOS) gaining more traction, Microsoft is starting to fully embrace (and market) the cloud.
While its Platform-as-a-Service (PaaS) offering may not be for many because of a lack of control or a learning curve creating a barrier to entry, Azure Infrastructure-as-a-Service could be readily tested and adopted, especially if the price is right.
5. Zero clients gain traction.
If the hardware drag around VDI is often focused on storage, servers and thin clients, the focus in 2010 will trend towards zero clients. Zero clients have the potential to offer greater security, a similar end-user experience, and in the case of some vendors (e.g. VMware), a good bit of flexibility. Not to mention, with thin clients currently available like the Samsung NC190 (a 19" monitor with a Teradici card inside), no longer does an organization have to price out a thin client and a monitor; today you can get them all-in-one. Pano Logic had a noteworthy part in opening up people's eyes to zero clients. In 2010, everyone will be in that game (although some will be a quarter late to the game).
6. There will be no VDI winner in the VMware versus Citrix battle.
Despite the catchy banner ad wars, there will not be a clear VDI winner in 2010, the reason being that the market is simply too large for there to be only one winner next year. While Citrix may have more years of selling VDI, VMware is the incumbent in most enterprise data centers today from a server virtualization standpoint (of those that have virtualized). There's plenty of market share to go around.
HDX and PCoIP are both good protocols and both companies make good brokers. Neither solution is perfect and they both have areas in which one wins over the other, but at the end of the day the VDI market is too large to have only one winner next year. I think it's more of a question of which company can build a more lethal sales organization around VDI. Both companies need to build a stellar VDI sales force and get their client-side hypervisors to market.
7. People still won't know how to use VMware Lifecycle Manager (which is a shame).
VMware Lifecycle Manager (LCM) is a great product from VMware that is still too complicated to get up and running for most people. Not only does VMware need to make this product a bit easier to install and use (or offer an installation service with the SKU), integration into the vCenter user interface would also be ideal.
I think VMware should also offer free online training around this product, because once you get the hang of it, you never look back. This is a product a lot of organizations sleep on but once they understand and see its potential, they don't know how they lived without it. Additionally, easier integration into ticketing solutions like Remedy would also increase LCM's adoption rate.
8. Amazon will remain the leader in cloud computing, despite what marketing regimes may tell
This isn't too surprising, but when you think about the fact that Microsoft's Azure will be launching next year along with (assumedly) some major players with VMware-based clouds, 2010 could finally see Amazon EC2 face some real competition. Out of all the cloud service providers next year, I think Microsoft Azure IaaS (not to be confused with the current PaaS offering) will see a good deal of traction. Nevertheless, Amazon has more enterprise features, more enterprise customers and more enterprise success than any public cloud service provider to date.
9. Another protocol will become more readily available.
Now that Red Hat's Simple Protocol for Independent Computing Environments (SPICE) remote desktop protocol is open source, I think we may see another vendor lesson its grip on its proprietary protocol (Citrix with HDX or VMware with soft PC-over-IP, for example) so that the protocol is more readily adopted. While the protocol may not have a place in other VDI solutions, it may have a place in an externally hosted desktop solution (e.g. from a cloud service provider).
10. Client-side hypervisor stays behind the curtains.
See you in 2011.
Jason Langone is vice president of Virtualization Services at Infinite Group, Inc. He has spoken at VMworld, Green Computing Summit and Virtualization Congress. Langone won the VMware Vanguard Award in 2007 and has architected some of the largest virtualization and cloud computing implementations to date. His solutions have been primarily implemented at Fortune Global 500 and public sector organizations and have received various accolades. Langone's focus remains on designing virtualization and cloud computing solutions in large-scale environments.