Developments in integrated computing
ESG recently published a Research Brief focusing on integrated computing trends which reveals that 2011 will likely see an uptick in the number of organizations committing IT budget to the purchase of integrated computing solutions. Why?
- Simplified management
- Reduced deployment times
- Better economic model
(The link to the brief goes into further details on each) Driven by customer demand, Dell has built its own integrated computing stack named vStart. You should follow the vStart link if you are interested in learning more, but it is basically an integrated solution that includes servers, storage, networking, hypervisors, and management software delivered as a single unit. Initially, Dell will offer a vStart 100v and 200v that supports 100 and 200 virtual machines, respectively. The initial configuration will support VMware with plans to support additional hypervisors in the coming quarters (note: interesting not to see initial support for Microsoft Hyper-V based on how Microsoft pumped its Dell relationship at MMS 2011.) Dell’s potential with vStart will have to align with what we hear from customers:
- Hand me the keys. Don’t give me a bunch of boxes delivered within weeks of each other and ten different engineers to set it up. I’m not interested in building the car–I want to drive it.
- Please make it simple. I don’t have enough time in the day as it is. Please deliver something that my kids can assemble and that I can quickly start to use. If components start to fail (and they will), I don’t want to spend hours troubleshooting the problem and risking application downtime. Please tell me what’s wrong, send me the new part, and make it stupid simple for me to replace it.
- Give me scalability options. Don’t sell me a Dell vStart 100v and when I need to deploy my 101st virtual machine tell me I need to by another six-figure stack of infrastructure. Please provide me the option of building with modular components. If I need more disk capacity, please give me the option to add additional “plug and play” capacity. If I need more CPU or memory, please give the option to scale a server or two before I have to wait to go through another budgeting cycle.
The last thing I would note is: how are customers going to properly size these solutions? I know Dell has sized them based on the number of virtual machines, but virtual machine workloads differ from one IT shop to the next. It’s also important to note that the majority, with VDI (virtual desktop infrastructure) as the possible exception, of the environments running these configurations will be mixed workload. My point is that capacity planning and application expertise are essential to the success of sizing these units right the first time and ultimately leading to long term, high customer satisfaction.
