Citizens of Holyoke, Mass., had reason to celebrate this week as the governor of Massachusetts and other dignitaries--including Cisco CEO John Chambers, EMC CEO Joe Tucci, and the presidents of Boston University, MIT, and the University of Massachusetts--announced plans to build a $100 million data center in their town. It's an ambitious proposal, but is it necessary?
Citizens of Holyoke, Mass., had reason to celebrate this week as the governor of Massachusetts and other dignitaries--including Cisco CEO John Chambers, EMC CEO Joe Tucci, and the presidents of Boston University, MIT, and the University of Massachusetts--announced plans to build a $100 million data center in their town. It's an ambitious proposal, but is it necessary?The Holyoke high-performance computer center--at this point, it's a mere glean in their eye--would support academic and corporate research, create local jobs, and serve as a state-of-the-art facility for green computing. Advocates enthusiastically describe it as a "cutting edge concept" and "absolute game changer" for the region. Governor Deval Patrick said the center would "lift up" Holyoke and surrounding economies.
Yet, it's hard to fathom why such a tech-smart group would sink $100 million into a data center construction project at the same time that many others see the emergence of on-demand cloud computing services as a faster, better, cheaper alternative, one not requiring capital investment, bulldozers, and hydroelectric gear. This is the stomping grounds of "Big Switch" author Nicholas Carr, after all.
As I see it, the folks behind the proposed Holyoke data center really have two scenarios to consider: One in which they, as proposed, build and operate a new data center, in essence becoming a cloud services provider; and a second in which they leave that work to Amazon, IBM-Google, or another experienced data center operator and pay for IT services only as needed.
Another idea: Let Cisco or EMC operate the data center, thrusting themselves directly into the cloud services market.
The partners in this venture have 120 days to come up with a plan, during which they need to pick a site, figure out how they will pay for the project, and all the other details that go into building a data center from scratch. Their evaluation should include an assessment of whether third-party cloud services would support the research needs of the universities and others involved.
The drawback to a cloud services approach is that Holyoke wouldn't get the new construction and other related jobs that local politicians and residents understandably hope for. But job creation is a bad rationale for building a data center, anyway. Better to focus on research, innovation, and the myriad business opportunities created by on-demand IT resources regardless of how they're delivered.
2014 Next-Gen WAN SurveyWhile 68% say demand for WAN bandwidth will increase, just 15% are in the process of bringing new services or more capacity online now. For 26%, cost is the problem. Enter vendors from Aryaka to Cisco to Pertino, all looking to use cloud to transform how IT delivers wide-area connectivity.
Server Market SplitsvilleJust because the server market's in the doldrums doesn't mean innovation has ceased. Far from it -- server technology is enjoying the biggest renaissance since the dawn of x86 systems. But the primary driver is now service providers, not enterprises.