I also got rid of my TiVo after this was not resolved or even addressed properly by TiVo. Too many bad features lately, not enough consumer advocacy. I am now TiVo-less for the first time in 4+ years. That says a lot.
I couldn't agree more with you. No real HD; no AAC -- despite a partnership with Apple for, what, two years?; no TiVo 2 Go yet (I've been on the "priority list" for over a month); charging for HMO then giving it away free to those who chose not to support TiVo by paying for it -- without refunds to those that did; and now from what I hear, DirecTV is pulling out, which I expect to be a death blow.
As soon as I buy an HD set, my TiVo is likely to be ancient history. Too bad, since I've had them since the original HDR-112 and have spoken nothing but praise to everyone I've met until now. It's disappointing.
Nope- that still does not tell me what "grid computing" is. This vague, loosely defined definition can describe just about every "next big thing" since the mainframe.
I like to think of "grid computing" as the idea of building technology with the global use in mind: pluggable security models, standard protocols for job management/data transfer/etc., and so on. how to build services with the future (where the future is sharing those services) in mind.
grid computing MAY take the form of: shared supercomputers; grid services; cycles for sale; cluster-of-workstations; distributed focused problem solving; etc... but these are all just applications of the above.
yes, I know this is still a very abstract idea, but I never promised it would be as good as the glossies promised.:)
First and foremost, grids are designed to run in a distributed environment which makes security design and administration that much more complex.
I'd have to say that in its essence, grid security is just traditional security on a larger scale with heavy X.509 use. sure, there's federation ideas and such, but there's nothing nothing "new" and unique to grid computing. we already have tons of users using public key authentication (agent systems, like "proxies" in grid-speak) from a distributed set of sites (think: computational centers).
just my.02. I don't buy the idea that a hacker can (today, anyway) cause more damage in a quicker fashion in a grid environment.
Sandia contracted out Red Storm development, hence RS is based on Sandia technologies towards petascale computing: Puma, Catamount, Portals, etc. No UNICOS, no IRIX, no Linux (except on the service nodes).
This is a quintessential kernel architecture without support for threads, VM, IPC, etc. The interconnect is also asynchronous (5ns end-end speculated), point-to-point, and uses Portals.
Jack Bernard is a good guy. The University believes, strongly, in privacy, and I can assure you that their counsel is approaching this from that point of view.
That being said, they aren't out to allow students to do illegal things and misappropriate bandwidth, either. It's RIAA strong-arm tactics that won't fly with the University.
If proper procedures are followed, documented evidence exists, and adequate warning happened (I believe the U still gives violators a warning and a chance to cease and desist), well.. sorry, Mr. Student, you may have to own up here.
On the bright side, I think Student Legal Service is still free.:)
Do NOT Read The Difference Engine
on
The Zenith Angle
·
· Score: 1
_The Difference Engine_ is quite possibly the most disappointing book I've ever slagged through. I honestly only spent the last month with it to say I read it, and to never have to think about it again. I had a lot of excitment for a "steampunk" (uh, ok) book from two great SF writers, but it rivals Hardy Boys #102 for "worst book ever."
I like Sterling most of the time, and love Gibson (huge plug for _Pattern Recognition_), but that book was just f'in horribly long and dichotically written. It's pretty much _Quicksilver_ without the action (heh).. ya dig?
(ObBookPlug for Peter F. Hamilton too.. with a negative for John C Wright to balance it out in the space opera subgenre.)
Anyway.. no OpenMosix here, this is using MPI. Specifically, on top of DK Panda's MPI libraries, they brought Kazushige Goto in to optimize the BLAS libraries in order to obtain the Top500 ranking of 10+ TF.
Incidentally, the Top500 rankings are based on a standardized LINPACK benchmark and formula, not "raw" processor rankings. I saw another comment that implied the latter.
Other interesting notes:
With conventional air cooling, the airspeed throughout the facility would have been 60-70MPH+. Try working on a console in a hurricane.
Dr. Varadarajan is a very very cool guy. He absolutely knows every detail that is going on in this machine, and knows how to make a good story out of it.
The facility this is in was upgraded to handle 3MW. The current cluster takes around 1.5MW. And you thought your Athlon was hot.:)
The #1 Top500 machine, the Earth Simulator, not only runs on custom Hitachi vector hardware, but required an entire new building to be built. The facility is a feat in itself, and is a big portion of the cost (for those of you extrapolating cost/performance if it was built at the same time as System X).
I'm very tempted to pre-order it along with Richard Morgan's new books (Broken Angels and Market Force) from amazon.co.uk, if shipping wasn't so much.
Peter F Hamilton
on
Coalescent
·
· Score: 2, Informative
It was accompanied by an another long short story in the Futures compendium by Peter F Hamilton called 'Watching Trees Grow'.
I've said it (here) before, and I'll say it again. If you're looking for excellent authors in the space-opera subgenre of SF, read all of Peter F Hamilton's stuff first.
I just finished his Second Chance at Eden, a collection of short-stories loosely related to his other novels and works. Very good stuff. But, the Reality Dysfunction/Neutronium Alchemist/Naked God trilogy (6 700+ pg paperbacks in the US) is just plain excellently engrossing, in my opinion. Check it out.
ask these guys for tips, pointers, and experiences:
Traverse Community WiFi
It sounds very identical in city-size and aim to what you want to do, and they seem to have made it work.
I knew I was getting old when I appreciated the four tires I got and wanted *more* socks and underwear. I actually need them and don't want to buy $10 pairs of boxer shorts for myself.
Although, they could have told me before I got an oil change and tire rotation last week.
Not all the Timbuk2 bags are made in SF. In fact, it's the laptop ones that are made in China. From this page:
This brings us to our new Laptop bags. These bags are much more complex to build, and require substantially more labor, and a variety of very expensive machines we don't have here in our factory. Unfortunately, it is not practical for us to make these new products in our San Francisco factory - the labor cost alone would make the retail price absurdly high. We designed our new computer bags to provide the best possible features, quality, and value at reasonable price points - and we designed them right here in San Francisco. We specified all of the materials and construction details to assure they meet the superior quality standards our customers have come to expect; and, we researched a dozen factories in China before we found one we thought was up to the task. Our factory in China employs a team of hardworking craftspeople earning good wages and an honest living, just like our factory in San Francisco. We visit our China factory on a regular basis every 4 to 8 weeks, to ensure superior quality standards and acceptable working conditions.
So, we're still the same hard working group of bag fanatics - designing and making great bags, and supporting our local community - in an increasingly competitive global market. If "made in China" offends your sensibilities, we're sorry, and we encourage you to purchase one of our messenger bags made in San Francisco. Meanwhile, we are pleased to say our new laptop bags are receiving rave reviews, and the additional business is allowing us to hire more people in all departments at our San Francisco headquarters - creating even more jobs locally, and paying even more taxes to support our wonderful city.
So, I completely agree that they're EVERYWHERE now, and pretty passe at this point. Well made, good bags, but about as hip as a Jansport from Meijer. That is, if you care. Plus -- I'm not sure their corporate philosophy is as strong as I would hope; see above.
I'd put in my vote for Chrome Bags. Also from SF, pretty cool stuff. Bailey Works bags are also bomber. They're a NH-based company, and their bags suit a snobby elitist pragmatic bastard like I, who prefers function over form but likes both.
The beginner's camera is (and has been for years) the Pentax K-1000. It's pretty much the defacto standard for students and beginning photographers.
I cannot agree more. I've been seriosuly into film and digital photography for almost 11 years, now. I would suggest nothing less than a study, manual, easy-to-use SLR like the K-1000. Get a 50mm lens to go with it, as that's approximately the human eye focal length at 35mm.
You will learn what apertures mean, what shutter speeds do, how to control depth-of-field, and how to work with light. It's manual, yes, but that's a GOOD thing in your case. Learn first, appreciate features afterwards. And you'll learn to conserve film before you learn to waste it (i.e. digital):).
I'd even go so far as to suggest a simple B&W developing setup. It is dirt-cheap, saves a little on processing costs, and allows a LOT more flexibility when learning.
You'll be able to sell the body and lens(es) for what you paid for them, roughly, if you buy them used. If you don't like the hobby, you've lost little and hopefully gained some new favorite photographs. Much better than if you discover you like taking pictures, not photography, and have spend $2k on a digital SLR that has since rapidly depreciated.
But then again, I drive a standard transmission. What do I know.
Limited resources run out, hence four (independent) threads running in parallel cannot write to the RF or fetch from memory concurrently. If your parallelism involves many different types of operations, it's much easier.
I suppose my original comment was worded badly -- being *able to* HARNESS the inherent (independent) parallelism with the resources at hand is the key, you are correct.
In other words, you're laying out the basic problems of:
1) Being able to FIND parallelism 2) Being able to take advantage of it:
a) Issuing multiple instructions (limited fetch bandwidth)
b) Executing them in parallel (limited FUs)
c) Committing them to memory / retiring
20% is generous, but that's a limitation of the simplicity of HT with respect to the EV8 / UltraSparc-V scale of SMT implementation, which leans towards a more full-issue design.
first, you don't just automatically get a linear increase with the width of the multiple-threading capabilities. it's not like it's free to increase the RF size and/or FUs, etc.
you're also confusing contexts with active threads. the Tera^WCray MTA had 128 contexts available -- so that thread switching is more light-weight, more or less -- but only one could be active at one time.
SMT in the various forms have more than one active thread, which introduces the problem(s) of competing for resources in the issue and retire stages, etc et al.
Slashdot Mirror
I also got rid of my TiVo after this was not resolved or even addressed properly by TiVo. Too many bad features lately, not enough consumer advocacy. I am now TiVo-less for the first time in 4+ years. That says a lot.
A lot of current development is going on at the OpenIB project.. in fact, that's where the actual 2.6.11 IB kernel bits came from.
I couldn't agree more with you. No real HD; no AAC -- despite a partnership with Apple for, what, two years?; no TiVo 2 Go yet (I've been on the "priority list" for over a month); charging for HMO then giving it away free to those who chose not to support TiVo by paying for it -- without refunds to those that did; and now from what I hear, DirecTV is pulling out, which I expect to be a death blow.
As soon as I buy an HD set, my TiVo is likely to be ancient history. Too bad, since I've had them since the original HDR-112 and have spoken nothing but praise to everyone I've met until now. It's disappointing.
I like to think of "grid computing" as the idea of building technology with the global use in mind: pluggable security models, standard protocols for job management/data transfer/etc., and so on. how to build services with the future (where the future is sharing those services) in mind.
grid computing MAY take the form of: shared supercomputers; grid services; cycles for sale; cluster-of-workstations; distributed focused problem solving; etc.
yes, I know this is still a very abstract idea, but I never promised it would be as good as the glossies promised.
I'd have to say that in its essence, grid security is just traditional security on a larger scale with heavy X.509 use. sure, there's federation ideas and such, but there's nothing nothing "new" and unique to grid computing. we already have tons of users using public key authentication (agent systems, like "proxies" in grid-speak) from a distributed set of sites (think: computational centers).
just my
Sandia contracted out Red Storm development, hence RS is based on Sandia technologies towards petascale computing: Puma, Catamount, Portals, etc. No UNICOS, no IRIX, no Linux (except on the service nodes).
This is a quintessential kernel architecture without support for threads, VM, IPC, etc. The interconnect is also asynchronous (5ns end-end speculated), point-to-point, and uses Portals.
Peter F. Hamilton's stuff is all also very very worth reading. The Night's Dawn stuff especially. Tons of great pulp reading for a bored geek.
Linux documentation, you say?
I can't say it's 100% geek-approved in terms of full-functionality, but it's a damn good start. Yeah?
He's probably actually pushing the XD1, formerly the OctigaBay 12K. Paul Terry used to be the OctigaBay CTO, remember.
Jack Bernard is a good guy. The University believes, strongly, in privacy, and I can assure you that their counsel is approaching this from that point of view.
:)
That being said, they aren't out to allow students to do illegal things and misappropriate bandwidth, either. It's RIAA strong-arm tactics that won't fly with the University.
If proper procedures are followed, documented evidence exists, and adequate warning happened (I believe the U still gives violators a warning and a chance to cease and desist), well.. sorry, Mr. Student, you may have to own up here.
On the bright side, I think Student Legal Service is still free.
I like Sterling most of the time, and love Gibson (huge plug for _Pattern Recognition_), but that book was just f'in horribly long and dichotically written. It's pretty much _Quicksilver_ without the action (heh).. ya dig?
(ObBookPlug for Peter F. Hamilton too.. with a negative for John C Wright to balance it out in the space opera subgenre.)
Paragraphs, man. They're useful.
Anyway.. no OpenMosix here, this is using MPI. Specifically, on top of DK Panda's MPI libraries, they brought Kazushige Goto in to optimize the BLAS libraries in order to obtain the Top500 ranking of 10+ TF.
Incidentally, the Top500 rankings are based on a standardized LINPACK benchmark and formula, not "raw" processor rankings. I saw another comment that implied the latter.
Other interesting notes:
More info here.
I'm very tempted to pre-order it along with Richard Morgan's new books (Broken Angels and Market Force) from amazon.co.uk, if shipping wasn't so much.
I've said it (here) before, and I'll say it again. If you're looking for excellent authors in the space-opera subgenre of SF, read all of Peter F Hamilton's stuff first.
I just finished his Second Chance at Eden, a collection of short-stories loosely related to his other novels and works. Very good stuff. But, the Reality Dysfunction/Neutronium Alchemist/Naked God trilogy (6 700+ pg paperbacks in the US) is just plain excellently engrossing, in my opinion. Check it out.
ask these guys for tips, pointers, and experiences: Traverse Community WiFi It sounds very identical in city-size and aim to what you want to do, and they seem to have made it work.
I knew I was getting old when I appreciated the four tires I got and wanted *more* socks and underwear. I actually need them and don't want to buy $10 pairs of boxer shorts for myself.
Although, they could have told me before I got an oil change and tire rotation last week.
So, I completely agree that they're EVERYWHERE now, and pretty passe at this point. Well made, good bags, but about as hip as a Jansport from Meijer. That is, if you care. Plus -- I'm not sure their corporate philosophy is as strong as I would hope; see above.
I'd put in my vote for Chrome Bags. Also from SF, pretty cool stuff. Bailey Works bags are also bomber. They're a NH-based company, and their bags suit a snobby elitist pragmatic bastard like I, who prefers function over form but likes both.
Hate RealOne with a passion hotter than the sun?
Try Real Alternative.
Worked for me.
I cannot agree more. I've been seriosuly into film and digital photography for almost 11 years, now. I would suggest nothing less than a study, manual, easy-to-use SLR like the K-1000. Get a 50mm lens to go with it, as that's approximately the human eye focal length at 35mm.
You will learn what apertures mean, what shutter speeds do, how to control depth-of-field, and how to work with light. It's manual, yes, but that's a GOOD thing in your case. Learn first, appreciate features afterwards. And you'll learn to conserve film before you learn to waste it (i.e. digital) :).
I'd even go so far as to suggest a simple B&W developing setup. It is dirt-cheap, saves a little on processing costs, and allows a LOT more flexibility when learning.
You'll be able to sell the body and lens(es) for what you paid for them, roughly, if you buy them used. If you don't like the hobby, you've lost little and hopefully gained some new favorite photographs. Much better than if you discover you like taking pictures, not photography, and have spend $2k on a digital SLR that has since rapidly depreciated.
But then again, I drive a standard transmission. What do I know.
I wasn't clear.
Limited resources run out, hence four (independent) threads running in parallel cannot write to the RF or fetch from memory concurrently. If your parallelism involves many different types of operations, it's much easier.
I suppose my original comment was worded badly -- being *able to* HARNESS the inherent (independent) parallelism with the resources at hand is the key, you are correct.
In other words, you're laying out the basic problems of:
1) Being able to FIND parallelism
2) Being able to take advantage of it:
a) Issuing multiple instructions (limited fetch bandwidth)
b) Executing them in parallel (limited FUs)
c) Committing them to memory / retiring
20% is generous, but that's a limitation of the simplicity of HT with respect to the EV8 / UltraSparc-V scale of SMT implementation, which leans towards a more full-issue design.
first, you don't just automatically get a linear increase with the width of the multiple-threading capabilities. it's not like it's free to increase the RF size and/or FUs, etc.
you're also confusing contexts with active threads. the Tera^WCray MTA had 128 contexts available -- so that thread switching is more light-weight, more or less -- but only one could be active at one time.
SMT in the various forms have more than one active thread, which introduces the problem(s) of competing for resources in the issue and retire stages, etc et al.
German!