Ah, but many other "true Grid" projects do. There really is not any distiction between Distributed Computing and the classic definition and implementation of a Grid. Un-federated wide-area computers collaborating to work on many different tasks, scheduled to run efficiently and effectivley.
There are many other Grid projects that don't have the long waiting line of TeraGrid. Much designed with the Globus toolkit allows anybody to contribute CPU power and just about anyone to run jobs. It's less powerfull and is more usefull for highly-parallel computations, but it is more true to the definition of what a Grid really is.
As noted in earlier comments, the TeraGrid's individual nodes _are_ NUMA clusters. This allows large, non-parallel computations to be run without individual service level agreements, login coordination and scheduling issues gumming up the process. The TeraGrid is an effort to remove the administrative nightmare's keeping most clusters from being fully utilized and most small-time scientists work from being completed.
They are in beta stages of a massive computation cycle for hire program that will allow organizations without the funding for an entire cluster to purchase cycles provided by a large IBM Power cluster.
It will allow for a computation cycle market to eventually arise, much like the wheat, corn or gold markets. Companies will compete to provide cheaper cycles, small-time scientists around the world will be able to have thier computation intensive problems solved at a fraction of the current cost possible today.
Hmmm. I think you are confusing the distributed OS additions they've made to Linux for their own clusters with the idle process harvesting of thier Google Toolbar.
The distributed OS and Filesystem in thier own clusters is far more advanced than a SETI@Home parallel work distribution algorithm. This OS/FS and projects like it are where the grid's heritige lies. There are many problems unique to the grid, but none of it could exist without the distributed system problems first solved in local area clusters.
There are some really cool innovations here that were a long time coming. One thing about Linux being slightly aged, large and now widley adopted, is that there is too much inertia against major change. DragonFly targets some serios kernel inefficiencies that will probably never get addressed in Linux. Rock on DragonFly!
This is good insight. I figured that going to graduate school would be a way to differentiate myself from the masses of B.S. and B.Eng students climbing on top of one another fro programming jobs. I realize now that expeience is what is valued in Sillicon Valley above all else. If I could not go out and get experience after my B.S., I should have been reconsidering my choice of field at that point.
Graduating Ph.D.'s have to accept whatever opportunities come their way. Teaming together with other graduates to start a company or consulting group is a risky and poor decision for most. The venture capital does'nt exist for that type of company any more. The reality is that graduates are at the whim of big companies with big R&D budgets. It's a matter of supply and demand. Right now there is a strong supply of graduates and a weak demand. I will be willing to bet that you graduated with a Computer related MS during the Internet boom and are able to charge high "freelance" rates becuase at that time, supply and demand were reversed.
This is other people's problem with Ph.D's but does not generally impact their employment opportunities or job performance. They are paid to be good with the formulas, et cetera. They never fully adapt to the working life becuase their knowledge is deep not broad.
The employment opportunities for U.S. Ph.D's are bleak becuase the field is competitive. There aren't that many positions outside of academia that require that specialized knoledge and there are plenty of talented people from other countries itching to plant themselves in the U.S. to get away from less than perfect conditions in their own country.
I also the situation as a factor of the US being the richest nation with a strong corporate culture and influence. We are the upper management of the rest of the world. Of course our bright all american kids are going to be interested in bossing around other people rather than pursuing advanced knoledge through the study of science.
This is not to say that there are'nt plenty of people who break stride from the norm, but our countries place in the world is a factor for those not influenced by any other.
I know that the bleak employment opportunities for a Computer Science Ph.D. in a 50th ranked school were the main reason I left my program and finished with a Masters instead.
Now I'm employed doing the same work I did while interning as an undergraduate 4 years ago. If I'm not able to move my way up through the ranks and get to some real development, going back for an MBA is a real possibility.
There is a good reason for this lag. Most Linux users (1% of all PC users) don't require slick UI to get their tasks done. Linux users primarily still use the terminal window and considering that KDE is ahead of Windows, I'd say they are doing pretty well.
I can't wait until this {Meta|Virtual|Smart} Folder technology makes it's way into mainstream OS's. So far, the only beta work I've seen in this area has come from the Gnome desktop group. Anyone here of Redmond catching onto this "everything is a database" thing?
Thanks for that link. Most of us who weren't at WWDC completly missed that on the Apple homepage, countless news stories and web logs covering the preview release. It's a good thing you are there to provide the link.
It looks like Apple caught on quickly to the Gmail label paradigm shift away from folders and has put "smart folders" into Mail 2.0 for 10.4.
IMHO labels and smart folders are long overdue for mail. They've been usefull in iTunes for months and just make good sense data that does not belong in only one bin.
That's easy. Name your files "Hot Lesbian 4-some[0000-1024].mpg" and make it available to the P2P sharing community. Every horny male in the country will then help to download and distribute your data across the country and when you want file number 5 back, just search for Hot Lesbian 4-some0005.mpg and bingo, 10 available good copys ready for re-use...
A Masters is just a blip in the keyword scanning tools used by most HR monkeys. I just graduated from a top 40 CS school with an Masters and my experience parallels most B.S. graduates in most all aspects.
I would'nt go so far as to say it was as waste of time, but it definatly did not yield the results I had hoped for when I entered the program two years ago.
And yes, a Ph.D. puts you strictly in the research areana and classes you out of most all jobs. I Quit the Ph.D. program in part becuase most who graduate from schools out of the top 20 almost never go on to teach/research at a University and have a much harder time finding a job than a M.S. degree holder does.
Uhm, the Internet is fault tolerant by design. Theoretically, it would take knowledge of the DNS root servers or the toasting of _many_ key locations to significantly reduce the ability to use the rest of the Internet.
Think about the large Niagra/New York power outage. During the whole ordeal, I sat in my California place of work and never noticed a single interuption.
Re:They got something like that already...
on
Mobile Wifi Backpack
·
· Score: 4, Insightful
What is possible using the existing protocol included in 802.11b devices (ad-hoc) is limited at best. There is no routing capability in these protocols and this prevents the networks from stringing out beyond the range of a single node. Considering a single node can only reach a little over a km given average xmit power and antenna gain, this is a problem for anything mobile and ad-hoc. Enter AODV, used by many "mesh" networks to extend the reach of a single wired gateway to much greater distances.
There are a _ton_ of well qualified people out of work as a result of nothing more than circumstance. Those people could easily be doing your job much better than you and your co-conspirator. I hope you sweat bullets and feel sick about what you are doing. Your going to get so fired when (not if) your employer finds out about this. Have fun bullshitting your way through meetings, wondering how your job gets done. You should save all the money you make from this job, so that when you're canned and out of work for more than 5 years; you might have a shot at being able to feed yourself.
I for one feel uneasy about my potential new Microsoft overlords.
I work as a student researcher under the Chief Scientist of Ask Jeeves and the Co-creator of Teoma. My work does not focus on the search algorithms, but the server management systems. I am not sure what MS would have planned in the way of integration, but I can tell you that everything we have written is heavily hooked to the Linux kernel and will be difficult to port into the Windows kernel.
That aside, I do feel that all that is required to challenge Google is a lot of motivated talent and a bit of extra funds. Google did well at writing an excellent "Version 2.0" of the search egine intially attempted by Yahoo! et. al. It will be interesting to see what "Version 3.0" will be like.
Slashdot Mirror
Ah, but many other "true Grid" projects do. There really is not any distiction between Distributed Computing and the classic definition and implementation of a Grid. Un-federated wide-area computers collaborating to work on many different tasks, scheduled to run efficiently and effectivley.
There are many other Grid projects that don't have the long waiting line of TeraGrid. Much designed with the Globus toolkit allows anybody to contribute CPU power and just about anyone to run jobs. It's less powerfull and is more usefull for highly-parallel computations, but it is more true to the definition of what a Grid really is.
As noted in earlier comments, the TeraGrid's individual nodes _are_ NUMA clusters. This allows large, non-parallel computations to be run without individual service level agreements, login coordination and scheduling issues gumming up the process. The TeraGrid is an effort to remove the administrative nightmare's keeping most clusters from being fully utilized and most small-time scientists work from being completed.
IBM is already making that vision a realization.
They are in beta stages of a massive computation cycle for hire program that will allow organizations without the funding for an entire cluster to purchase cycles provided by a large IBM Power cluster.
It will allow for a computation cycle market to eventually arise, much like the wheat, corn or gold markets. Companies will compete to provide cheaper cycles, small-time scientists around the world will be able to have thier computation intensive problems solved at a fraction of the current cost possible today.
Hmmm. I think you are confusing the distributed OS additions they've made to Linux for their own clusters with the idle process harvesting of thier Google Toolbar.
The distributed OS and Filesystem in thier own clusters is far more advanced than a SETI@Home parallel work distribution algorithm. This OS/FS and projects like it are where the grid's heritige lies. There are many problems unique to the grid, but none of it could exist without the distributed system problems first solved in local area clusters.
There are some really cool innovations here that were a long time coming. One thing about Linux being slightly aged, large and now widley adopted, is that there is too much inertia against major change. DragonFly targets some serios kernel inefficiencies that will probably never get addressed in Linux. Rock on DragonFly!
This is good insight. I figured that going to graduate school would be a way to differentiate myself from the masses of B.S. and B.Eng students climbing on top of one another fro programming jobs. I realize now that expeience is what is valued in Sillicon Valley above all else. If I could not go out and get experience after my B.S., I should have been reconsidering my choice of field at that point.
Graduating Ph.D.'s have to accept whatever opportunities come their way. Teaming together with other graduates to start a company or consulting group is a risky and poor decision for most. The venture capital does'nt exist for that type of company any more. The reality is that graduates are at the whim of big companies with big R&D budgets. It's a matter of supply and demand. Right now there is a strong supply of graduates and a weak demand. I will be willing to bet that you graduated with a Computer related MS during the Internet boom and are able to charge high "freelance" rates becuase at that time, supply and demand were reversed.
This is other people's problem with Ph.D's but does not generally impact their employment opportunities or job performance. They are paid to be good with the formulas, et cetera. They never fully adapt to the working life becuase their knowledge is deep not broad.
The employment opportunities for U.S. Ph.D's are bleak becuase the field is competitive. There aren't that many positions outside of academia that require that specialized knoledge and there are plenty of talented people from other countries itching to plant themselves in the U.S. to get away from less than perfect conditions in their own country.
I also the situation as a factor of the US being the richest nation with a strong corporate culture and influence. We are the upper management of the rest of the world. Of course our bright all american kids are going to be interested in bossing around other people rather than pursuing advanced knoledge through the study of science. This is not to say that there are'nt plenty of people who break stride from the norm, but our countries place in the world is a factor for those not influenced by any other.
I know that the bleak employment opportunities for a Computer Science Ph.D. in a 50th ranked school were the main reason I left my program and finished with a Masters instead. Now I'm employed doing the same work I did while interning as an undergraduate 4 years ago. If I'm not able to move my way up through the ranks and get to some real development, going back for an MBA is a real possibility.
Maybe that's just me and my lab mates. You kids and your fancy click and point devices.
As soon as someone develops a GUI to allow me to easily command and access all 32 nodes in our Linux cluster, I'll stop using the CLI too.
There is a good reason for this lag. Most Linux users (1% of all PC users) don't require slick UI to get their tasks done. Linux users primarily still use the terminal window and considering that KDE is ahead of Windows, I'd say they are doing pretty well.
I can't wait until this {Meta|Virtual|Smart} Folder technology makes it's way into mainstream OS's. So far, the only beta work I've seen in this area has come from the Gnome desktop group. Anyone here of Redmond catching onto this "everything is a database" thing?
Thanks for that link. Most of us who weren't at WWDC completly missed that on the Apple homepage, countless news stories and web logs covering the preview release. It's a good thing you are there to provide the link.
It looks like Apple caught on quickly to the Gmail label paradigm shift away from folders and has put "smart folders" into Mail 2.0 for 10.4.
IMHO labels and smart folders are long overdue for mail. They've been usefull in iTunes for months and just make good sense data that does not belong in only one bin.
That's easy. Name your files "Hot Lesbian 4-some[0000-1024].mpg" and make it available to the P2P sharing community. Every horny male in the country will then help to download and distribute your data across the country and when you want file number 5 back, just search for Hot Lesbian 4-some0005.mpg and bingo, 10 available good copys ready for re-use. ..
Give an active link to the home page on the home page.
Oh, yeah that always pisses me off; when I can't go to the home page from the home page. Damn, that would definatly cost a company major bucks.
the photos are so secret that only the first 4 people to click on the photos were able to see them before the server crumbled.
A Masters is just a blip in the keyword scanning tools used by most HR monkeys. I just graduated from a top 40 CS school with an Masters and my experience parallels most B.S. graduates in most all aspects.
I would'nt go so far as to say it was as waste of time, but it definatly did not yield the results I had hoped for when I entered the program two years ago.
And yes, a Ph.D. puts you strictly in the research areana and classes you out of most all jobs. I Quit the Ph.D. program in part becuase most who graduate from schools out of the top 20 almost never go on to teach/research at a University and have a much harder time finding a job than a M.S. degree holder does.
Uhm, the Internet is fault tolerant by design. Theoretically, it would take knowledge of the DNS root servers or the toasting of _many_ key locations to significantly reduce the ability to use the rest of the Internet.
Think about the large Niagra/New York power outage. During the whole ordeal, I sat in my California place of work and never noticed a single interuption.
What is possible using the existing protocol included in 802.11b devices (ad-hoc) is limited at best. There is no routing capability in these protocols and this prevents the networks from stringing out beyond the range of a single node. Considering a single node can only reach a little over a km given average xmit power and antenna gain, this is a problem for anything mobile and ad-hoc. Enter AODV, used by many "mesh" networks to extend the reach of a single wired gateway to much greater distances.
I bought an iBook on Friday for around $1300 without tax. That must have been the sale that pushed them over the edge.
There are a _ton_ of well qualified people out of work as a result of nothing more than circumstance. Those people could easily be doing your job much better than you and your co-conspirator. I hope you sweat bullets and feel sick about what you are doing. Your going to get so fired when (not if) your employer finds out about this. Have fun bullshitting your way through meetings, wondering how your job gets done. You should save all the money you make from this job, so that when you're canned and out of work for more than 5 years; you might have a shot at being able to feed yourself.
I for one feel uneasy about my potential new Microsoft overlords.
I work as a student researcher under the Chief Scientist of Ask Jeeves and the Co-creator of Teoma. My work does not focus on the search algorithms, but the server management systems. I am not sure what MS would have planned in the way of integration, but I can tell you that everything we have written is heavily hooked to the Linux kernel and will be difficult to port into the Windows kernel.
That aside, I do feel that all that is required to challenge Google is a lot of motivated talent and a bit of extra funds. Google did well at writing an excellent "Version 2.0" of the search egine intially attempted by Yahoo! et. al. It will be interesting to see what "Version 3.0" will be like.