Dan, the Digipede Network Product Manager, has just announced an MSDN Webcast he’ll be doing: Introducing Object Oriented Programming for Grid (OOP-G). The event is Tuesday, April 11th @ 11:00 PDT.
This is the first of two MSDN Webcasts he has scheduled. I’m sure he’ll make them both informative and entertaining.
Go to his post (MSDN Webcast: Object Oriented Programming for Grid) for more information and a registration link.
Yesterday Dan and I discussed a pet peeve of mine: compute cycles being likened to electricity. This comes up nearly every time someone talks or blogs about utility computing. The catalyst this time was Nicholas Carr and his piece, Is the Server Industry Doomed? While he doesn’t even mention electricity here, it doesn’t take much to get us back on this topic. BTW, I agree that much of IT is heading toward utility computing; . . . but, the electricity metaphor doesn’t hold up because the cycles are not electricity.
I told Dan I might blog on this topic, but he beat me to it in this post: Because 1.21 gigaflops just aren’t 1.21 gigawatts (and then has apparently continued arguing with Mr. Carr since then). Even better than me having to write a thing! As Dan says:
Partly, it’s just physics. All electrons look alike (let’s not get into electron spin here: as far as my appliances are concerned, every electron looks the same). It doesn’t matter to me if the power that’s lighting up my life, running my refrigerator, and powering my PC came from a wind farm, a hydroelectric plant, or a diesel turbine. Well, for environmental reasons, I might prefer the former two, but the point is that when an electron gets to me, I can’t tell where it came from.
Computes just aren’t the same. Computes look different on different operating systems. Not all software runs on all operating systems. Different people prefer different toolsets, and they always will. Some OSs are better for some things than others, and people choose the appropriate OSs for them. Yes, we’ve all read about “write once, run everywhere” software–but a small minority of software actually runs that way. OSs are different, and they will continue to be different. People will continue to write software that takes advantage of particular OSs.
But, that wasn’t really my point either.
Lets allow that utility computing is the centralized generator (analogous to electric utilities), but what is being generated? Certainly not compute cycles. The generator burns compute cycles (and electricity and data) to produce . . . data.
So, is the commodity the data? Of course not. Data are almost always proprietary. Is moving the data the commodity? Bandwidth is a commodity, but it only scarcely resembles electricity transmission and distribution.
So, the commodity in utility computing is the compute cycle, but utility computing doesn’t resemble an electric utility. The metaphor just doesn’t make sense. I guess people like it because there was a time when electricity generation moved from a local to a centralized model.
How about a moratorium on the electric utility metaphor? Utility computing doesn’t need a metaphor to make sense.
Tags: Digipede, grid, Utility-Computing
Dan, the product manager for the Digipede Network will be giving a talk tomorrow night at SDForum on object-oriented programming for Grid (or OOP-G, as I call it).
He blogs about it here: West Coast Grid: Windows Grid: SDForum Talk Wednesday Night.
He will be talking about using the Digipede Network behind Web Services — check it out.
A paper, entitled, Evolutionary Approach to Realizing the Grid Vision has been published by Marvin Theimer, Savas Parastatidis, and Tony Hey of Microsoft, Marty Humphrey of University of Virginia, and Geoffrey Fox of Indiana University. Presentation from the authors and a link to the document itself can be found here.
This paper suggests several areas of focus to bring the grand vision of the greater Grid down to a practical level that can be accomplished in smaller steps. Not surprisingly, the authors suggest starting first with computational grids. The paper focuses on the basics of this area where the problem set is fairly well understood and puts off those areas that are strictly outside the requirements of a computational grid.
It is partially the “overarching intellectual context” with which the OGSA Working Group is charged that keeps the greater Grid outside the interest of much of industry. This kind of focus articulated in the paper, properly executed, really helps progress towards the greater Grid.
BTW: I still think we don’t know what Savas is actually working on at Microsoft 😉
The Digipede Network has been chosen as a finalist for the 2006 Codie Awards.
Thanks, SIIA judges!
Update: Dan has a more complete description of the awards, here.
On Thursday, January 12th, Kim Greenlee will be presenting at eBig. She’ll be talking about OOP-G and introduce the Digipede Network. I understand she’ll be giving away Developer Editions too!
Details on eBig’s site, here.
Nathan just updated the Digipede News and Announcements feed to use FeedBurner. If you are interested in getting news about our product and company, subscribe to: http://feeds.feedburner.com/DigipedeNewsAndAnnouncements.
This is not news to those who subscribe to Software Development Magazine, but they reviewed the Digipede Network in their January, 2006 issue. You normally would need to sign up for their online content to read it, but we purchased a reprint link which you can find here. Two things I’m happy about:
The reviewer, Mr. Rick Wayne, pointed out a couple of issues. Some of these we have already addressed and others are in our plans.
Thanks, Mr. Wayne — yes, we are listening).
Today Dan’l Lewin, Microsoft Corporate VP of .NET Business Development, highlights some startups using .NET (see Checked My List Twice). Of course, he included Digipede! About his list, he says:
I believe they are among the most creative, inventive (and yes, ‘innovative’ startups) that we follow. They are doing some amazing things — creating solutions and products that would definitely make it to Santa’s list (not that any of us actually make lists anymore.)
Thanks Dan’l!
Postscript: Robert Scoble and Don Dodge blogged about Dan’l’s list too.
In Kill all screensavers, Nick Carr writes about a problem facing CIOs wanting to harness idle compute resources: graphic and compute-intensive corporate screensavers. We have all seen these screensavers either in our own companies or while visiting others. Depending on who you ask, these either encroach on the employee environment or they are a valuable conduit for employee education. They are also mistakingly considered a green alternative to employee communications versus paper-based communications as screensavers can cause enormous power consumption in aggregate. Nick Carr has some numbers supporting this view in his article.
Delivering employee messaging through screensavers is incredibly inefficient: aren’t employees either using their computers (i.e., there is no screensaver running), or doing something else (i.e., not looking at their screensavers)? But when the employee is out of the office, it runs 100% of the time. 16+ hours of employee communications directed at an empty chair.
Want a green alternative to employee messaging? Use e-mail.
Save your idle compute resources for applications that deliver real business value. That is one of the reasons we developed the Digipede Network: to deliver real business value on idle resources.
Aside from grid computing, the CIOs should be successful fighting the screensaver with the power savings argument.
Tags: .NET, green, grid, screensavers
FriendFeed Status
About me
Main feed
Email subscriptions