October 30, 2008
I'm heading to Toronto this weekend to be part of a panel on cloud computing at a conference called "Powering Innovation: A National Summit," co-sponsored by Canadian research networks CANARIE and ORION. Although my presentation will focus on the enterprise, the overall focus of the conference takes me back to the early focus of GRIDtoday -- and I think that's a good thing.
Just as the enterprise is struggling with cloud computing and how to best leverage its myriad benefits, so, too, is the research sector. And just like the research sector drove many advancements in grid computing, it wouldn't be shocking to me to see the same computer scientists driving comparable evolutions in cloud computing. In fact, some might say the TeraGrid, with its Science Gateways and associated Web portal, is a precursor to today's consumer-oriented cloud offerings (albeit a tad more rigid).
And just how, you ask, are a bunch of stodgy, old researchers and academicians going to effect change in a hip, Web 2.0-inspired paradigm like cloud computing? There are many reasons, actually, beginning with three biggies: (1) they’re smart; (2) they’re used to working together; and (3) they’re not as old as you might think.
First, the men and women at our national labs and universities have been responsible for just about every technology underlying cloud computing -- including grid computing, virtualization and the Internet -- so I’m not about to put anything past them. If they are putting their considerable brainpower to work thinking about cloud computing -- and something tells me they are (and here)-- good things are bound to follow. Prior to becoming CTO of Amazon and all but launching the cloud revolution, Werner Vogels was a research scientist at Cornell.
A big area of concern for enterprise users is the prospect of provider lock-in in the cloud, and the research sector should be able to help here, too. For starters, I know that cloud computing is on the Open Grid Forum’s radar, so it’s possible we could see some real work around cloud standards get underway relatively soon from the guys who brought us the HPC Basic Profile Specification. Although it won’t necessarily solve the problems of disparate APIs and databases, any progress on the standardization front is good news. Or maybe the world’s cloud providers could get together to work some of these things out. At any rate, a coming together of researchers, users and vendors under one banner or another could help ensure everybody’s best interests are considered as the cloud moves forward.
Government-funded grid computing initiatives like EGEE also are looking at how to both leverage and learn from cloud computing. This could have a big impact because the key to such projects is interoperability. I think a lot was learned as EGEE worked to get its network of grids to work together, and we all could benefit if they figure out a way to make that international grid network interoperate with large public clouds. The world of scientific grid computing thrives on worldwide collaborative efforts, and I think this experience can play a significant role in cloud interoperability, as well.
Finally, those folks behind the scenes at today’s Web companies -- including many of the founders -- are relatively young and only getting younger. If we look back, we might recall that a couple of the first shots in cloud computing war, if you will, were fired by Google and IBM, then by Yahoo, when they launched their respective cloud efforts within leading computer science universities. Mere teenagers, in some cases, will be exposed to the technologies that make the respective Web giants click, and by the time these 19-year-olds become grad students and Ph.D. candidates, they will be veterans in developing Web-scale applications and management software. It wouldn’t be surprising to see innovation coming directly out of the universities, from either students, researchers or their increasingly knowledgeable instructors. When these kids enter the workforce and have the R&D resources of someone like Google or Amazon at their disposal, forget about it.
Posted by Derrick Harris - October 30, 2008 @ 6:52 PM, Pacific Daylight Time
Derrick Harris is the Editor of On-Demand Enterprise
No Recent Blog Comments
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?