Slashdot Mirror

Bingo on the graph theory. My gut feeling (read: logical extrapolation of probabilities) leaves me pretty certain that over the next 20 years it will become a core factor in understanding, designing and perceiving most application programs - driven in equal measure by increased use of the Resource Description Framework (RDF) which is core to semantic web markup, increased use of aggregated computer systems based on web services, and the eventual commoditization of Grid services which are still very much at the research stage.

The combination of all of the above will result in the a norm where services can be perceived in the way (code) libraries are today, so designing, building, visualising and proving the completeness of a program becomes "just" an application of graph theory on a very large graph.

The beauty of a graph is that a novice can begin to understand it very quickly, but to know it you need the math. So just as today, people will be able to create simple programs & scripts without requiring too much investment of time, but ultimately, systems architects, designers, programmers and administrators will find their jobs easier if they have the mathematical background to be able to communicate how their systems work.

In that case, it's your lucky day. Check out these sites:

The Dream Project
Global Grid Forum

Some companies provide loaner equipment to universities. It would be interesting to find something that is of interest/use on your campus and then get some hardware to play with. Videoconferencing equipment (end point stations and MCUs) might be a good start.

SIP and VOIP. This is something that is cool, fun and fairly accessible to folks with a decent network connection. And, if you have the energy, the administration on your campus might be interested to hear that it could save them money.

Games. Games are an interesting and fun topic. With a bit of thought, a talk could be constructed that mixed fun with learning (grin).

Internet2 Days. Advanced Networking. If you are at a school that is involved in Internet2, there are resources that are available.

Microsoft Resrach Group. Evil blah blah yes, whatever. Say what you will, but they are still smart people. Your login indicates that you might be interested in HCI, so Dumais's site might be a good start. Other companies are also listed in various places.

GRID. buzz buzz buzz. You say that no marketing hype is allowed. Well, for all the buzz about GRIDs, the folks that can give you the skinny are prob to be found here. Depending on your location, asking one of them might just get a visit.

As you are at a university (and I am assuming you are in the U.S.), you might find that someone from the NSF would be able to give you some interesting insights into what they consider to be important, interesting and fundable. Again, not sure how successful you would be at getting anyone to visit, but it is a a starter for ideas.

Lastly, many of the folks that you might want to have present will be busy. It might be worth considering an interactive presentation over video. Sounds cheezy, but many folks are used to this type of presentation.

Some companies provide loaner equipment to universities. It would be interesting to find something that is of interest/use on your campus and then get some hardware to play with. Videoconferencing equipment (end point stations and MCUs) might be a good start.

SIP and VOIP. This is something that is cool, fun and fairly accessible to folks with a decent network connection. And, if you have the energy, the administration on your campus might be interested to hear that it could save them money.

Games. Games are an interesting and fun topic. With a bit of thought, a talk could be constructed that mixed fun with learning (grin).

Internet2 Days. Advanced Networking. If you are at a school that is involved in Internet2, there are resources that are available.

Microsoft Resrach Group. Evil blah blah yes, whatever. Say what you will, but they are still smart people. Your login indicates that you might be interested in HCI, so Dumais's site might be a good start. Other companies are also listed in various places.

GRID. buzz buzz buzz. You say that no marketing hype is allowed. Well, for all the buzz about GRIDs, the folks that can give you the skinny are prob to be found here. Depending on your location, asking one of them might just get a visit.

As you are at a university (and I am assuming you are in the U.S.), you might find that someone from the NSF would be able to give you some interesting insights into what they consider to be important, interesting and fundable. Again, not sure how successful you would be at getting anyone to visit, but it is a a starter for ideas.

Lastly, many of the folks that you might want to have present will be busy. It might be worth considering an interactive presentation over video. Sounds cheezy, but many folks are used to this type of presentation.

Yeah. There's actually quite a lot of research going into this currently. It's called the Grid (think "power grid", ubiquitous, simple to use), and I predict it will be the next big buzzword.

See Global Grid Forum, Grid Today and the Globus project for starters.

The problem of buying and selling computation power on some sort of broker basis is a quite interesting problem in itself. Exactly what are you selling? Hardly CPU hours, since the value of those depends on the hardware.

• The amount of data is fairly trivial compared to the Particle Physics data (AstroGrid, EU DataGrid, Grid Forum
• Trust. As in lack there of. SETI had problems with people returning bogus results in the past and had to spend time (and several articles) on improving security and getting the same unit processed by several seperate users.
• Lack of man power / time.
• Dependability - e.g. an astronomer wants to run an analysis now! Is there any certainty that there will be x amount of processing power available? That your computer will still be on the network in 5 minutes time? That the unit will get processed in a reasonable real time?

DISCLAIMER: I work for one of the centres involved in the DataGrid project.

One of the things DataGrid is designed to do is to give researchers easy access to the data they need.

It's kind of like a distributed data store with a tree like structure. The collider feeds data to national centres, they feed data to regional centres, regional centres feed data to local research groups, the researchers analyse the data.

What's more interesting, is what happens when these researchers start to exchange their results... terabytes of data flying around in all directions, not just downstream.

As for Grid Computing, yes - most of the technology isn't new, but then again neither was the World Wide Web. The Web was successful because it took existing good ideas, added a killer application (Mosaic) and proved to be useful to other fields than the one it was developed for.

The problem is that "grid" computing is being used to describe a number of distinctly different things: distributed data stores, clustered supercomputers, run-anywhere computing resources, commodity computing...

See the GlobalGridForum pages at: http://www.gridforum.org
for more details about Grid research and projects across the world.

I agree as long as you talking about Linux. Nonehteless, apparently, this article was designed to project it on truly exciting field of cluster computing, where it, IMHO, fails.

As for Linux it's simply a tool of choice being among a small number of operating systems able to facilitate such an extensive research that clustering (especially the branch of global computing) necessitates.

Another project that is interested in performance prediction is the Network Weather Service (NWS). An important issue in systems such as RPS (my system) and NWS is accurate and scalable measurement of hosts and networks. Remos is able to do this.

A lot of work in this area is taking place in the context of Computational Grids. The Grid Forum is an IETF-like body that is trying to standardize Grid middleware systems. Globus and Legion are examples of Grid middleware systems.