You can blame the CERN Director-General for that one. Just about everyone else thought that the press release/conference nonsense was pretty poorly handled. But when the Director-General says jump, everyone else says "how high?".
I hope it doesn't make your math teacher pull her hair out. If so, she isn't very good at algebra (or recognizing where simple algebraic manipulation is applicable).
I actually got that one wrong, because I misread the division sign as an addition sign. (47*75) + 25 is a very different number... Guess I need to get my eyes checked.
The real breakdown there is whether or not you are paying for datacenter space too. If you have to pay for power, HVAC, etc out of your budget, then yes, rent from the cloud. If you're like most small research groups, you're part of a larger university or industry department, and you're already paying building overhead costs, which include some sort of server room space. For that sort of budget, I'd be astonished if you needed more than about 10 tiles of server room space, and probably a lot less than that. This is extremely workable in the sort of environment I described.
About a year ago, my research group (I'm a grad student in high-energy physics at a major American university) received 30k$ in supplemental funding via the American Recovery and Reinvestment Act. I used the money to build our group a 39-node (156 core * hyperthreading) compute cluster. For that price, we could have rented 94 days of 156 small EC2 instances. If you try to factor in maintenance and development time, you can get a little bit longer of course, but I certainly spent less than a year's worth of work on it, so you'd less than double the EC2 rental length at grad student pay rates...
More technical details: HEP is embarrassingly parallel tasks with IO needs ranging from very heavy to very light, and with fairly large data storage needs. I bought commodity desktop hardware (core i7 930, fairly cheap X58 chipset mobos, low-end name-brand memory, mid-range PSUs, cheap as near free chassis, 1.5TB WD HDD per node), and a 48-port Cisco gigabit switch (with our IO throughput needs for some jobs, and the OS install scheme, didn't want to skimp at all on the networking). We already had a general purpose server machine (file, NIS, electronic logbook, etc), which became the headnode.
Each node PXE boot + NFSROOT from the headnode, which reduces the maintenance dramatically. The disks in each node contain swap and data storage, and are tied together with Hadoop's HDFS or gfarm (admittedly, this part is still in flux, because extremely high inter-node network traffic causes nodes to hang spontaneously.... Still working on that aspect) which will, when/if it works, give us much greater data throughput by eliminating the bottleneck to the headnode for data.
If you'd like to chat more about building a small compute cluster on a shoestring budget according to this sort of model, please feel free to email me at jay ess double-u at eff enn ay ell dot gov.
Well shoot. I guess I'll just go hide under the bed now.
How many times have we invented away the looming Malthusian catastrophe, mostly just to satisfy our own greed? I think we will again, and I'd rather take that grand chance than cower in fear that we might offend Gaia.
Somebody on here once said, and I agree with them, "I like living in a country where the poor people are fat."
Because DEX was my dump stat, my parents had a standing agreement with the school(s, elementary, junior high, high) that I could type any assignment rather than handwriting it. Even as a physics grad student, I still typed my problem sets (in LaTeX, of course).
On the other hand, I have been hit (as a pedestrian, and only a glancing blow, but it still hurt) by a cyclist who was running a red light, riding on the wrong side of the road, and who failed to stop and render aid. When I lived closer to the local university in the undergraduate ghetto, I regularly had to stop my car while driving quite normally and legally down a narrow one-way street so as to avoid running over the cyclists coming the wrong way down the middle of the street. In the same neighbourhood, my wife pulled up to a stop sign and stopped (again on a one-way street). A cyclist coming down the cross street decided that he need to turn the wrong way down that street, didn't look, nearly plastered himself on the hood of her stationary car, and then stopped, gave her the finger, and stood in the road cussing her out. There're assholes everywhere, pedestrians and cyclists included. Perhaps Tucson possesses a uniquely law-abiding cyclist culture.
Nobody who is a scientist (except possibly for inflammatory gossip addicts like Dorigo) is claiming anything remotely resembling a discovery. Nature is, in my opinion, highly irresponsible for posting things like this, precisely because it leads to reactions like yours. It isn't screwy instrumentation or glitches, it isn't a discovery, it isn't an exclusion, it isn't a bird or a plane or superman, it's just a result that is not yet conclusive.
Scientific Linux won't, I expect. In my experience, SL has an extreme phobia of updating anything, ever, unless it becomes absolutely necessary. K-splice would fly in the face of this philosophy.
Considering your post as a short essay, I have one general comment: Keep it together. You started out quite coherent and interesting, but as the post continued, you got more grandiose, less coherent, less cohesive, and less comprehensible.
This is why I don't want socialized medicine in the US. Because then one can make the (admittedly valid in that context) legal argument that if I do anything that even might hurt myself, I am creating a cost to society and should be prevented. Then laws spring up that try to nerf the world and stop anyone from doing anything remotely dangerous.
I'd far rather allow people to take risks in the full knowledge that they are responsible for their own insurance (or lack thereof).
Steve Jobs, Steve Wozniak, Bill Gates, Paul Allen, and Mark Zuckerberg all dropped out of college, and they all went on to have fairly decent careers.
Yep. And you know their names. Compare the number of people with successful careers (in fields traditionally pursued by college graduates) after dropping out of college to the number who dropped out of college. Now compare the number of people with successful careers (in fields traditionally pursued by college graduates) after finishing college to the number who finished college. Which probability would you like to have of achieving a successful career?
TFS certainly sounds like Apple desires to pursue that sort of anticompetitive situation, however.
Which makes a lot of sense, considering that Apple's razors-and-blades, vertical-marketplace model for iTunes (and the various iDevices) doesn't make as much sense with the world of TV, where your Sony, Samsung, or (egads!) Westinghouse set is just as happy with a Google TV box, or a Roku, or one of many other media devices, as it is with an Apple TV attached.
By implication, the writer of TFS believes that Apple would be happier if a significant portion of the TV market was made up of devices that did not work or did not work as well with their competitors boxes.
That was me. In the analysis released on Friday, D0 does not perform the delta R reweighting (this was a specific criticism that they sought to address). In spite of no delta R reweighting, they still do not see the bump. There are some systematic errors that they handle differently from CDF which are quite likely to explain the result. Some of my colleagues at CDF are investigating (and were investigating before this D0 release, because of a suggestion by a D0 physicist at the release of the original bump paper) these systematics and their effect on our ability to model the data well. I can't really comment further until results are released, however.
It is probably more important to note that D0 handles their systematics a little bit differently. This is natural, because their detector is different, so their systematic errors are different. But there appears to be one particular systematic that they handled better than CDF, and CDF's less-good handling produced a bump-shaped mismodeling of the data. I don't want to say too much, but hint: quarks and gluons don't necessarily quite act alike in some important particulars.
Slashdot Mirror
You can blame the CERN Director-General for that one. Just about everyone else thought that the press release/conference nonsense was pretty poorly handled. But when the Director-General says jump, everyone else says "how high?".
I hope it doesn't make your math teacher pull her hair out. If so, she isn't very good at algebra (or recognizing where simple algebraic manipulation is applicable).
I actually got that one wrong, because I misread the division sign as an addition sign. (47*75) + 25 is a very different number... Guess I need to get my eyes checked.
Unfortunately, Commerce Clause.
Banking on the assumption that laws not currently enforced will never be enforced doesn't sound to me like especially sound business practice...
It's a lot cheaper on the small scale to build towers and put them on a meatrack than to get rackmount chassis et al. A *lot* cheaper.
The real breakdown there is whether or not you are paying for datacenter space too. If you have to pay for power, HVAC, etc out of your budget, then yes, rent from the cloud. If you're like most small research groups, you're part of a larger university or industry department, and you're already paying building overhead costs, which include some sort of server room space. For that sort of budget, I'd be astonished if you needed more than about 10 tiles of server room space, and probably a lot less than that. This is extremely workable in the sort of environment I described.
About a year ago, my research group (I'm a grad student in high-energy physics at a major American university) received 30k$ in supplemental funding via the American Recovery and Reinvestment Act. I used the money to build our group a 39-node (156 core * hyperthreading) compute cluster. For that price, we could have rented 94 days of 156 small EC2 instances. If you try to factor in maintenance and development time, you can get a little bit longer of course, but I certainly spent less than a year's worth of work on it, so you'd less than double the EC2 rental length at grad student pay rates...
More technical details: HEP is embarrassingly parallel tasks with IO needs ranging from very heavy to very light, and with fairly large data storage needs. I bought commodity desktop hardware (core i7 930, fairly cheap X58 chipset mobos, low-end name-brand memory, mid-range PSUs, cheap as near free chassis, 1.5TB WD HDD per node), and a 48-port Cisco gigabit switch (with our IO throughput needs for some jobs, and the OS install scheme, didn't want to skimp at all on the networking). We already had a general purpose server machine (file, NIS, electronic logbook, etc), which became the headnode.
Each node PXE boot + NFSROOT from the headnode, which reduces the maintenance dramatically. The disks in each node contain swap and data storage, and are tied together with Hadoop's HDFS or gfarm (admittedly, this part is still in flux, because extremely high inter-node network traffic causes nodes to hang spontaneously.... Still working on that aspect) which will, when/if it works, give us much greater data throughput by eliminating the bottleneck to the headnode for data.
If you'd like to chat more about building a small compute cluster on a shoestring budget according to this sort of model, please feel free to email me at jay ess double-u at eff enn ay ell dot gov.
Well shoot. I guess I'll just go hide under the bed now.
How many times have we invented away the looming Malthusian catastrophe, mostly just to satisfy our own greed? I think we will again, and I'd rather take that grand chance than cower in fear that we might offend Gaia.
Somebody on here once said, and I agree with them, "I like living in a country where the poor people are fat."
Because DEX was my dump stat, my parents had a standing agreement with the school(s, elementary, junior high, high) that I could type any assignment rather than handwriting it. Even as a physics grad student, I still typed my problem sets (in LaTeX, of course).
I might do that, if it could emulate vim.
The Kindle DX has a screen size of 9.7".
On the other hand, I have been hit (as a pedestrian, and only a glancing blow, but it still hurt) by a cyclist who was running a red light, riding on the wrong side of the road, and who failed to stop and render aid. When I lived closer to the local university in the undergraduate ghetto, I regularly had to stop my car while driving quite normally and legally down a narrow one-way street so as to avoid running over the cyclists coming the wrong way down the middle of the street. In the same neighbourhood, my wife pulled up to a stop sign and stopped (again on a one-way street). A cyclist coming down the cross street decided that he need to turn the wrong way down that street, didn't look, nearly plastered himself on the hood of her stationary car, and then stopped, gave her the finger, and stood in the road cussing her out. There're assholes everywhere, pedestrians and cyclists included. Perhaps Tucson possesses a uniquely law-abiding cyclist culture.
Sure. That's a totally reasonable approach for them to take. It's much better than outlawing dangerous activities.
Nobody who is a scientist (except possibly for inflammatory gossip addicts like Dorigo) is claiming anything remotely resembling a discovery. Nature is, in my opinion, highly irresponsible for posting things like this, precisely because it leads to reactions like yours. It isn't screwy instrumentation or glitches, it isn't a discovery, it isn't an exclusion, it isn't a bird or a plane or superman, it's just a result that is not yet conclusive.
Scientific Linux won't, I expect. In my experience, SL has an extreme phobia of updating anything, ever, unless it becomes absolutely necessary. K-splice would fly in the face of this philosophy.
Considering your post as a short essay, I have one general comment: Keep it together. You started out quite coherent and interesting, but as the post continued, you got more grandiose, less coherent, less cohesive, and less comprehensible.
This.
This is why I don't want socialized medicine in the US. Because then one can make the (admittedly valid in that context) legal argument that if I do anything that even might hurt myself, I am creating a cost to society and should be prevented. Then laws spring up that try to nerf the world and stop anyone from doing anything remotely dangerous.
I'd far rather allow people to take risks in the full knowledge that they are responsible for their own insurance (or lack thereof).
You should check your local laws.
Nah, I'm not really that into Pokemon.
Steve Jobs, Steve Wozniak, Bill Gates, Paul Allen, and Mark Zuckerberg all dropped out of college, and they all went on to have fairly decent careers.
Yep. And you know their names. Compare the number of people with successful careers (in fields traditionally pursued by college graduates) after dropping out of college to the number who dropped out of college. Now compare the number of people with successful careers (in fields traditionally pursued by college graduates) after finishing college to the number who finished college. Which probability would you like to have of achieving a successful career?
Which makes a lot of sense, considering that Apple's razors-and-blades, vertical-marketplace model for iTunes (and the various iDevices) doesn't make as much sense with the world of TV, where your Sony, Samsung, or (egads!) Westinghouse set is just as happy with a Google TV box, or a Roku, or one of many other media devices, as it is with an Apple TV attached.
By implication, the writer of TFS believes that Apple would be happier if a significant portion of the TV market was made up of devices that did not work or did not work as well with their competitors boxes.
The technological singularity will occur in 2015, but we won't have human-level machine translation for another 14 years....
You don't need to set oracle to evil, it already has that value. You just need: assert oracle==evil
That was me. In the analysis released on Friday, D0 does not perform the delta R reweighting (this was a specific criticism that they sought to address). In spite of no delta R reweighting, they still do not see the bump. There are some systematic errors that they handle differently from CDF which are quite likely to explain the result. Some of my colleagues at CDF are investigating (and were investigating before this D0 release, because of a suggestion by a D0 physicist at the release of the original bump paper) these systematics and their effect on our ability to model the data well. I can't really comment further until results are released, however.
It is probably more important to note that D0 handles their systematics a little bit differently. This is natural, because their detector is different, so their systematic errors are different. But there appears to be one particular systematic that they handled better than CDF, and CDF's less-good handling produced a bump-shaped mismodeling of the data. I don't want to say too much, but hint: quarks and gluons don't necessarily quite act alike in some important particulars.