But *no one* else does. Contrary to your beliefs, Apple does not define standards, FireWire is dead except in the minds of a few obscure fanboys. eSata has almost completely replaced Firewire in every capacity that USB could not.
Doesn't every digital cable box sold in the US have to have firewire, by law? My FiOS (motorola) boxes certainly do...
Heh, I've been on Stony's team since '07, captained it last year. I probably know you...
That power outage was *awful* for us, we were going with the strategy we used last again of a large ram based communal scratch.. the outage *really* didnt help there
Thatd somewhat be true, which is why there is a restriction on commercially available hardware. The hardware is loaned, and comes from the engineering labs of the vendors. There are no vendors big enough to be approached for this that are small enough not to have top of the line hardware in their labs. Sometimes it's not one vendor either, nor are all vendors exclusive. The rule of thumb has been the OEMs are exclusive, but chips and networking are not always.
Last year for stony brook we had Dell, AMD, and Mellanox as sponsors. AMD provided 6-core low power chips, dell provided the blade center and blades, Mellanox provided us with QDR 40GB infiniband for our interconnect. AMD also helped out Purdue.
I can't speak to this year publicly yet, but I know that some of our hardware this year is non-exclusive as well.
Basically the answer is that practically speaking money isnt a factor. In particular since the systems are loaned the vendors can put a lot of hardware in without worrying too much about the cost. The commercially available restriction *is* there to keep money from being a factor (ie having a heavyweight with deep pockets game the competition by designing a machine *just* for the competition to win some press)
This year it looks to be shaping up to be interesting. I can't speak to other teams, but we're working with Cray and have some nice hardware to use:-). We don't have to pay for the hardware BTW, the vendors provide it (usually out of their engineering labs).
The problem is top-of the line xeons come with power penalties, and a lot of the competition comes in the form of the apps too, and how well you configured your os, and how well you managed the queue, and... The raw power team does not always win, not even close.
The competition is harder than it sounds, you have to build a cluster from the ground up, fit it into the power requirement (which means stripping out redundancies among other things), strip down a distro (we used Debian as a starting point), get the apps optimized, and then run through the data sets. Your team needs to *understand* the apps, the OS, and the hardware in order to win. There are several people from various teams from past years who have moved on to doing their PhDs in comp sci based on work from this competition (At Carnegie Mellon, MIT, and UMich off the top of my head).
It's important too, in a few ways. For one I know I learned more about clusters the first day I started working on the team for this competition back in 2007 than I ever knew before. That knowledge has led to research fellowships, jobs, and knowledge of what I want to (biochemical modeling). It's an experiance that very few undergrads get, and I think that's a shame.
For the industry it's an important highlight of what can be done with a lot of dedication and a focus on wringing the most from your hardware and software. and in doing that we showcase a lot of work that people dont think about. For example our cluster last year ran off a single disk, plus a large ramdisk as scratch exported over QDR infiniband to the compute nodes. No, it's not new, but it was novel to a lot of people who dropped by our booth.
For another, the ASU team was the first time *I* and many others ever saw a windows cluster in the wild.
Competitions like this are important, they showcase technology and introduce it to undergrads early, with positive benefit!
I had images in my brain of an article about getting torque and mpi (something that, to my knowledge, there's no reason outside practicality to stop from working) plus some sort of auto-meshing running on android phones or some such (though using wifi as an interconnect makes me shudder), *that* would be phone supercomputing, this is *not*.
My driving is a mix of highway and city (NYC--> LI). I think the new round of hybrids *might* be economical for me, I've been looking at numbers (not ready to replace my current car yet), but up until recently the higher *maintenance costs (of which the battery is part) was what prevented me from getting one.
I dont know about vancouver, but her in NY, Taxi drivers have been *required* to get hybrids (a court case which the city just lost and Bloomberg was rather irritable about in the press). Just sayin'
You know, there are those of us who *do* drive 35k miles/yr, or close to it. My *normal* daily commute at the moment is 130miles/day, 5-6 days/week. One of the reasons I've held off on a hybrid is the battery issue with my driving. So yes, there are those of us to whom that issue matters
While I usually buy books I want to read, sometimes it's a book I don't want to own, and the library doesn't have, so I head to B&N and read it there... BUT, I (and lots of other people) usually sit in the in-store starbucks, drink coffee... have a snack... they most likely make more money on my food and drink (the coffee at least has much higher margins) than they would have if I had bought the book and had coffee at home:-p
Heh, trust a/. thread to put me on the side of defending MS::rolls eyes::
I think the point all the way up above that Double Drop was trying to make, the point that I was trying to make as well later on, is that having everything easily centrally configurable makes things *easier*. Why else do you think PITA software like Centrify is so popular (which, btw, uses AD as it's backend, and exports to everything else from that AD database)?
Yes, you can deal with web traffic at the proxy/firewall/gateway, and you should since in a large environment your network *is* going to be to some extent heterogeneous, but configuring it in AD is another tool in the arsenal, and being able to do *scriptable policies about specific users* without playing annoying games with a proxy box is useful too, as is customizing look and feel, enforcing search engine pref, etc. All from one centralized location, conveniently the same one that most of the rest of the network is probably configured from.
I'm kinda annoyed I ended up in the position of MS shill here, most of my work is on large linux clusters, my workstations are either osx or linux and my laptop is a macbook pro, I avoid using windows as much as possible BUT, they have their strengths, particularly in corporate deployment. And firefox, without AD integration, can't compete there.
I have, albiet a long time ago. The level of fine control you have over the clients, and IE, is fantastic, and far exceeds just "who can view what on the web with ease". Now, I was running the AD server for only ~90 users using 10 workstations for a small student group at college. I am by no means a MS guy normally so I imagine that convenience goes up exponentially in large corporate settings, and with admins who are more well versed in the minutiae of windows server than I was.
Slashdot Mirror
But *no one* else does. Contrary to your beliefs, Apple does not define standards, FireWire is dead except in the minds of a few obscure fanboys. eSata has almost completely replaced Firewire in every capacity that USB could not.
Doesn't every digital cable box sold in the US have to have firewire, by law? My FiOS (motorola) boxes certainly do...
People have tried in the past, announcing the winners, never have managed to be approved. For the record those were
2007: University of Alberta, CA
2008: Indiana University/TU Dresden (Germany) combined team
2009: Stony Brook University (State University of New York)
Heh, I've been on Stony's team since '07, captained it last year. I probably know you...
That power outage was *awful* for us, we were going with the strategy we used last again of a large ram based communal scratch.. the outage *really* didnt help there
Thatd somewhat be true, which is why there is a restriction on commercially available hardware. The hardware is loaned, and comes from the engineering labs of the vendors. There are no vendors big enough to be approached for this that are small enough not to have top of the line hardware in their labs. Sometimes it's not one vendor either, nor are all vendors exclusive. The rule of thumb has been the OEMs are exclusive, but chips and networking are not always.
Last year for stony brook we had Dell, AMD, and Mellanox as sponsors. AMD provided 6-core low power chips, dell provided the blade center and blades, Mellanox provided us with QDR 40GB infiniband for our interconnect. AMD also helped out Purdue.
I can't speak to this year publicly yet, but I know that some of our hardware this year is non-exclusive as well.
Basically the answer is that practically speaking money isnt a factor. In particular since the systems are loaned the vendors can put a lot of hardware in without worrying too much about the cost. The commercially available restriction *is* there to keep money from being a factor (ie having a heavyweight with deep pockets game the competition by designing a machine *just* for the competition to win some press)
Nope, UPSs and such are not allowed (trust me, we've looked into that!)
The drink of choice at the competition tends to be Mountain Dew :-)
This year it looks to be shaping up to be interesting. I can't speak to other teams, but we're working with Cray and have some nice hardware to use :-). We don't have to pay for the hardware BTW, the vendors provide it (usually out of their engineering labs).
The problem is top-of the line xeons come with power penalties, and a lot of the competition comes in the form of the apps too, and how well you configured your os, and how well you managed the queue, and... The raw power team does not always win, not even close.
Hey look, a IU veteran, don't know who you are, so don't know if you're someone I remember from '07, but hey!
...Stony Brook University, and the piece that's missing from the summary is 26 amps@120v, (dual circuits, soft capped at draw of 13 each)
Links to more info from the conference: SC10 CC Page, rules, and app list.
The competition is harder than it sounds, you have to build a cluster from the ground up, fit it into the power requirement (which means stripping out redundancies among other things), strip down a distro (we used Debian as a starting point), get the apps optimized, and then run through the data sets. Your team needs to *understand* the apps, the OS, and the hardware in order to win. There are several people from various teams from past years who have moved on to doing their PhDs in comp sci based on work from this competition (At Carnegie Mellon, MIT, and UMich off the top of my head).
It's important too, in a few ways. For one I know I learned more about clusters the first day I started working on the team for this competition back in 2007 than I ever knew before. That knowledge has led to research fellowships, jobs, and knowledge of what I want to (biochemical modeling). It's an experiance that very few undergrads get, and I think that's a shame.
For the industry it's an important highlight of what can be done with a lot of dedication and a focus on wringing the most from your hardware and software. and in doing that we showcase a lot of work that people dont think about. For example our cluster last year ran off a single disk, plus a large ramdisk as scratch exported over QDR infiniband to the compute nodes. No, it's not new, but it was novel to a lot of people who dropped by our booth.
For another, the ASU team was the first time *I* and many others ever saw a windows cluster in the wild.
Competitions like this are important, they showcase technology and introduce it to undergrads early, with positive benefit!
I had images in my brain of an article about getting torque and mpi (something that, to my knowledge, there's no reason outside practicality to stop from working) plus some sort of auto-meshing running on android phones or some such (though using wifi as an interconnect makes me shudder), *that* would be phone supercomputing, this is *not*.
Doesn't google have dual class stock? I don't know how much of the "A" class Brin and Page have, but I'm sure it's more than 19%
My driving is a mix of highway and city (NYC--> LI). I think the new round of hybrids *might* be economical for me, I've been looking at numbers (not ready to replace my current car yet), but up until recently the higher *maintenance costs (of which the battery is part) was what prevented me from getting one.
I dont know about vancouver, but her in NY, Taxi drivers have been *required* to get hybrids (a court case which the city just lost and Bloomberg was rather irritable about in the press). Just sayin'
You know, there are those of us who *do* drive 35k miles/yr, or close to it. My *normal* daily commute at the moment is 130miles/day, 5-6 days/week. One of the reasons I've held off on a hybrid is the battery issue with my driving. So yes, there are those of us to whom that issue matters
Actually, at the moment, only jailbreakers can be *safe* from this vulnerability. Google "PDF Loading Warner". Ironic, isn't it?
While I usually buy books I want to read, sometimes it's a book I don't want to own, and the library doesn't have, so I head to B&N and read it there... BUT, I (and lots of other people) usually sit in the in-store starbucks, drink coffee... have a snack... they most likely make more money on my food and drink (the coffee at least has much higher margins) than they would have if I had bought the book and had coffee at home :-p
Heh, trust a /. thread to put me on the side of defending MS ::rolls eyes::
I think the point all the way up above that Double Drop was trying to make, the point that I was trying to make as well later on, is that having everything easily centrally configurable makes things *easier*. Why else do you think PITA software like Centrify is so popular (which, btw, uses AD as it's backend, and exports to everything else from that AD database)?
Yes, you can deal with web traffic at the proxy/firewall/gateway, and you should since in a large environment your network *is* going to be to some extent heterogeneous, but configuring it in AD is another tool in the arsenal, and being able to do *scriptable policies about specific users* without playing annoying games with a proxy box is useful too, as is customizing look and feel, enforcing search engine pref, etc. All from one centralized location, conveniently the same one that most of the rest of the network is probably configured from.
I'm kinda annoyed I ended up in the position of MS shill here, most of my work is on large linux clusters, my workstations are either osx or linux and my laptop is a macbook pro, I avoid using windows as much as possible BUT, they have their strengths, particularly in corporate deployment. And firefox, without AD integration, can't compete there.
You do realize you can configure more than just web traffic with AD, right?
Have you ever actually run an AD server?
I have, albiet a long time ago. The level of fine control you have over the clients, and IE, is fantastic, and far exceeds just "who can view what on the web with ease". Now, I was running the AD server for only ~90 users using 10 workstations for a small student group at college. I am by no means a MS guy normally so I imagine that convenience goes up exponentially in large corporate settings, and with admins who are more well versed in the minutiae of windows server than I was.
Funny you should post that, in NYC it would actually be police work, using the NYPD's Hercules Teams
Or different profit centers, HP's a much bigger server supplier than dell, not mention networking gear.
That said, you're probably right
My startac used to get *better* reception if I held the antenna!
For a QOS standpoint that makes sense, but the attenuation would be apparent to the proper testing done by the engineers in the lab *before* it was field tested by employees making calls, and indeed today engadget has a story that the engineers knew this could be a bad problem
Not all adults have credit cards (and honestly, a lot of adults that have credit cards shouldnt :-p)