OTOH, you can buy a nice *off-the-shelf* opteron based workstation (e.g. supermicro motherboard) with four 12 core CPUs and 128GB RAM in a 4RU case with 6 or more hot swap bays for about $5-6K. Not as many cores and less RAM but:
- it's all in one machine - so better for big memory hungry jobs
- there's a lot of stuff that works better with a single multicore SMP machine than with MPI
- it's a lot less stuffing around than building a DIY cluster
- it will have a real warranty
- takes only 4 or 5RU.
- there's room for some GPUs in that case later too.
or use a 16GB or 32GB USB flash (or better yet, a small SSD - swapping to USB flash would suck) as the boot drive on most machines and have one machine (the head node) with hard disks as a file server - NFS will do for small to medium size clusters (anywhere from a handful of nodes to a few hundred nodes). The OP is going to need a head node anyway to run Slurm or Torque as the scheduler/resource-manager (yes, i have built clusters before).
put a 2nd NIC in the head node, so the compute nodes can run on a private 192.168 network (you'll need a 24 or 48 port switch as well), and also install DHCP, tftp, and apache. Set up the last three to allow the compute nodes to netboot clonezillla....install everything you'll need on one compute node (openmpi, libatlas, octave, R, open source and proprietary scientific software as needed, etc) and use clonezilla to mass produce the rest (also allows you to quickly and easily add new nodes or replace failed nodes). LDAP or NIS will be needed for sharing account/auth details between machines.
i built something quite similar to this last year (but using some sunfire 1RU opteron rackmount servers as the compute nodes)
I'd go for an x4 CPU, they're not that much more than an x3 and the extra core is useful. 8GB RAM too, 2x4GB only costs about $40). given the budget, it's probably not worth getting a custom power supply for the tray-mounted motherboards, so each will need its own dedicated PSU
each node is going to cost somewhere around $250 (very rough estimates: $50 for the m/b, $40 for 8GB RAM, $50 CPU, $50 PSU, $60 for 32GB SSD - but possibly a fair bit cheaper as a bulk purchase), and the head node will cost roughly triple that (you'll need a case w/ hot-swap bays for the drives - a Norco 4224 is probably overkill but at well under $400 for 4RU with 24 SAS/SATA hot-swap bays, it would be hard to find a significantly cheaper case even with less drive bays) so for $6K you can build a cluster with 20 x 4 core compute nodes plus a good head node for the scheduler & file server). 80 compute cores for $6K. that's good, even considering that with cheap crap motherboards you'll have a noticable failure rate. the cluster i built last year with name brand hardware cost closer to $50K. I could build a better system today (far less nodes with a lot more cores and RAM each), also with name brand hardware, for about $20K - $30K
trays for the motherboards, the rack(s), and cooling will cost extra. as will licenses for any proprietary software they might need to run (could easily cost as much - or more! - as the hardware). if the OP's friend is at a university, she can probably scavenge an old rack or two from another dept, but even if she has to buy one new she could easily build 15+ compute nodes entirely within the $6K budget
it is not categorically irrational to believe in God
actually, it is. Believing in something for which there is absolutely no evidence, and in fact is contra-indicated by the evidence, is pretty much the canonical example of irrational
believing in "God" is *precisely* as irrational as believing in Invisible Pink Unicorns - no more, no less.
Don't worry, you won't miss out on the fun. Apple have their own plans for turning OS X into a desktop OS suitable for cretins who only ever use mobile phones too.
thanks for the tip about clementine - i'll check it out.
i switched from amarok to exaile when amarok 2 fucked up what had been a great music player / organiser. exaile's OK, but not as good as what amarok used to be. hopefully clementine will be close.
Americans don't actually *hate* taxes. It's just that they've been programmed with a pavlovian response of stubborn anger and hostility when they hear the term. It's dog-whistle politics at its worst, and a guaranteed reliable way of getting Americans to act (and vote) against their own best interests.
worst of all, american political missionaries have done an unpleasantly effective job of exporting their cowardly fear of the word "tax" to other countries
these infrared sensors and controllers will soon be installed in police cars and possibly in police uniforms if they're small enough. and anywhere else that governments, as well as the rich and powerful, want to make it harder for ordinary citizens to gather evidence of their abuses of power.
there's this term called 'fair use' that you need to learn about. it specifically allows that sort of copying for academic purposes.
it's part of what we, the people, retained when we authorised the creation of socialist monopoly copyright laws to restrict our natural right to read and copy whatever information want to.
the lawsuit isn't a complaint that fair-use rights are being exceeded. it is an attempt to undermine - or completely destroy - existing fair-use rights.
Anyone who thinks that governments don't have legitimate reasons to limit who knows what information is out of touch with reality
Certainly, governements have legitimate reasons to want to keep things secret. I don't think anyone really disputes that.
What some people seem to forget is that a government's desire to keep secrets doesn't make it a crime for citizens - or even third parties - to want to find out or even try to find out....and as long as they commit no actual crimes like trespass or espionage while trying to find out, doing so is perfectly legal, legitimate, and moral
FWIW, I think citizens have an absolute, inalienable right to find out through any and all legal means what their government is up to if they want (and i don't think it's particularly wrong if they're willing to commit and bear the consequences of committing *minor* non-violent crimes in the process). remember, a democratic government - or even one that pretends to democratic processes - is the servant of the people, not the master.
any laws that attempt to criminalise the exercise of that right are bad laws, deserving contempt and civil disobedience
or they don't trust the new and enthusiastic "volunteers" with secret agendas due to the inevitable infiltration attempts from journalists, spook agencies, agent provocateurs, reactionaries with some kind of point to prove, and the occasional random lunatic.
a $20M penalty clause probably doesn't mean much to a real spook or even to a journo with a legal team, but it probably discourages the independent infiltrators
diabetes research
on
Is Sugar Toxic?
·
· Score: 5, Interesting
the toxicity of sugar (sucrose, glucose, fructose, etc) is one of things that almost no researcher in the know dares to mention publicly because it would be career (and funding) suicide. the processed food industry is far too powerful a lobby group.
but the researchers know. check medline. almost every research article on diabetes begins with words to the effect of "fed the rats sugar until they developed diabetes". feeding rats sugar is THE consistently reproducible method of inducing diabetes.
and this is what the processed food lobby is doing to consumers every day with sugar in absofuckinglutely everything. even things you think wouldn't have sugar because they're supposed to be salty or sour or savoury or anything-else-but-sweet have sugar in them. because it's cheap, it's addictive (esp. to children and adults with poor impulse control - i.e. most of the population), and it's a preservative.
sugar in our diet isn't bad when it's rare and unrefined (as it is in fruits and vegetables etc. and in our natural pre-agriculture diet it WAS rare, but it was a huge amount of easily absorbed energy which is why we evolved the ability to taste sweetness...and why we also evolved to *like* it). even when humans first discovered processed sugar from sugar cane a few hundred years ago it wasn't a huge problem because it was very expensive (like all spices were) - only the rich could afford it.
even the improvement of refinery processes that made sugar became extremely refined and extremely cheap wasn't that bad....it was only when "food" factories started putting it in *everything* so that it became almost impossible to avoid eating far too much of the stuff that it became a problem.
and this, btw, is also why the poor (and the time-poor) suffer from diabetes more than the rich do - the rich can afford to eat well. the poor can't (money-wise AND time-wise).
i'm a gnome user (in that i use the gnome panel[1] and most of the GUI apps i use are gnome) but i switched to the openbox window manager several years ago because the "standard" gnome window managers metacity and mutter absolutely suck. even worse than the software is the attitude ("design philosophy" if you want to be fancy) behind them - that taking away features is a Good Thing because users are too stupid to understand them and easily confused by choices.
this latest idiocy is just an extension of the initial practical reason i started using openbox - middle-click and right-click on the maximise button for vertical-only and horizontal-only maximise of a window. IIRC, after some argument a few years ago, the metacity devs agreed to add (or keep, i can't remember) the vert-max and horiz-max features, but refused to enable them via middle- or right- click on the max button....the ONLY way to access them is to manually configure the keyboard bindings to assign a key to them.
the gnome terminal, actually vte, also has an annoying broken-by-design bug of sending eight up/down arrow keys to the application running in the terminal when the scroll wheel is moved. the devs flat out refuse to acknowledge that this causes problems for programs like mutt and vi, and refuse to fix, and messes up middle-button pasting (because the scroll wheel is usually also the middle button, and it's almost impossible to click it without scrolling it a little at first) - even though several patches have been submitted over the years that the bug has been in the gnome bugzilla. because of this, i use mrxvt rather than gnome terminal - which, of course, has its own bugs but at least the dev doesn't suffer from the Gnome Developer Attitude Problem.
gnome has a lot of good software, but it also has a lot of rage-inducing idiocy like the above.
[1] the gnome panel annoys me too - buggy bloated crap. i've looked around for something to replace it with but haven't found one yet.
the trouble is that it *doesn't* fix the problem, it just hides it.
rebooting - or even re-imaging - can be fine as a quick-and-dirty fix to get critical infrastructure back up and running as quickly as possible, but is IS NOT and CAN NOT be a substitute for actually figuring out what caused the problem and fixing it so that it doesn't happen again
That applies to a much lesser extent (or not at all) outside the US.
what makes you think I'm in the US? I'm not. I live in Australia, and (as in most of the rest of the world) almost all mobile phones here are sold indirectly through bundling deals with telcos.
I haven't read or even heard of the book before, but just reading the wikipedia article about the author confirms my initial thought that a *very* short summary of her thesis might be that the actual crime of child abuse may be compounded many times over by the constant barrage of people telling the victim that they ought to be severely traumatised actually makes them far more traumatised than they would otherwise be, in a manner reminiscent of the way that deluded, obsessive, or even outright malicious "therapists" *create* False Memory Syndrome in susceptible people.
i.e. that the crime of abuse may have happened, but that the trauma may be partly or even wholly created after the fact.
it's not even a particularly surprising or unusual thesis...the Satanic Ritual Abuse moral panic of the 1980s clearly shows that it's possible to create such trauma even in children who *haven't* been sexually abused.
damn. clicked 'submit' rather than 'continue editing'.
i meant to add that it's in the telcos' interests for mobile phone prices to stay high. It makes it much easier for them to trap people into long-term contracts. if "current" phone prices were only $100 or $200 rather than $600-$900, there's be a lot more people who just bought their phones outright and shopped around for the best voice&data deal.
in short: phone prices are artificially high just to make the bundled phone+overpriced plans seem a much better deal than buying the phone outright.
The trend for electronics to decrease in price does not seem to have started yet for mobile
That's because the end-user isn't really the manufacturer's customer. The telcos are. Almost all mobile phones are sold by Telcos (or their agents) in a bundle with a voice and/or data plan.
This has an enormous anti-competitive effect on the market. More than enough to counteract the pricing trends in the direct-to-consumer electronics markets
It's also why many features that end-users would like (e.g. an answering machine app, an an alternative to the telco-provided voicemail) are missing. There's no great incentive to create them...and, for some features, a huge disincentive if they're the kind of feature that might piss off a telco
Slashdot Mirror
OTOH, you can buy a nice *off-the-shelf* opteron based workstation (e.g. supermicro motherboard) with four 12 core CPUs and 128GB RAM in a 4RU case with 6 or more hot swap bays for about $5-6K. Not as many cores and less RAM but:
or use a 16GB or 32GB USB flash (or better yet, a small SSD - swapping to USB flash would suck) as the boot drive on most machines and have one machine (the head node) with hard disks as a file server - NFS will do for small to medium size clusters (anywhere from a handful of nodes to a few hundred nodes). The OP is going to need a head node anyway to run Slurm or Torque as the scheduler/resource-manager (yes, i have built clusters before).
put a 2nd NIC in the head node, so the compute nodes can run on a private 192.168 network (you'll need a 24 or 48 port switch as well), and also install DHCP, tftp, and apache. Set up the last three to allow the compute nodes to netboot clonezillla....install everything you'll need on one compute node (openmpi, libatlas, octave, R, open source and proprietary scientific software as needed, etc) and use clonezilla to mass produce the rest (also allows you to quickly and easily add new nodes or replace failed nodes). LDAP or NIS will be needed for sharing account/auth details between machines.
i built something quite similar to this last year (but using some sunfire 1RU opteron rackmount servers as the compute nodes)
I'd go for an x4 CPU, they're not that much more than an x3 and the extra core is useful. 8GB RAM too, 2x4GB only costs about $40). given the budget, it's probably not worth getting a custom power supply for the tray-mounted motherboards, so each will need its own dedicated PSU
each node is going to cost somewhere around $250 (very rough estimates: $50 for the m/b, $40 for 8GB RAM, $50 CPU, $50 PSU, $60 for 32GB SSD - but possibly a fair bit cheaper as a bulk purchase), and the head node will cost roughly triple that (you'll need a case w/ hot-swap bays for the drives - a Norco 4224 is probably overkill but at well under $400 for 4RU with 24 SAS/SATA hot-swap bays, it would be hard to find a significantly cheaper case even with less drive bays) so for $6K you can build a cluster with 20 x 4 core compute nodes plus a good head node for the scheduler & file server). 80 compute cores for $6K. that's good, even considering that with cheap crap motherboards you'll have a noticable failure rate. the cluster i built last year with name brand hardware cost closer to $50K. I could build a better system today (far less nodes with a lot more cores and RAM each), also with name brand hardware, for about $20K - $30K
trays for the motherboards, the rack(s), and cooling will cost extra. as will licenses for any proprietary software they might need to run (could easily cost as much - or more! - as the hardware). if the OP's friend is at a university, she can probably scavenge an old rack or two from another dept, but even if she has to buy one new she could easily build 15+ compute nodes entirely within the $6K budget
actually, it is. Believing in something for which there is absolutely no evidence, and in fact is contra-indicated by the evidence, is pretty much the canonical example of irrational
believing in "God" is *precisely* as irrational as believing in Invisible Pink Unicorns - no more, no less.
it's a bit dumbass to not realise that "little known" was an instance of this new and never-before-seen thing called sarcasm.
Don't worry, you won't miss out on the fun. Apple have their own plans for turning OS X into a desktop OS suitable for cretins who only ever use mobile phones too.
thanks for the tip about clementine - i'll check it out. i switched from amarok to exaile when amarok 2 fucked up what had been a great music player / organiser. exaile's OK, but not as good as what amarok used to be. hopefully clementine will be close.
Americans don't actually *hate* taxes. It's just that they've been programmed with a pavlovian response of stubborn anger and hostility when they hear the term. It's dog-whistle politics at its worst, and a guaranteed reliable way of getting Americans to act (and vote) against their own best interests.
worst of all, american political missionaries have done an unpleasantly effective job of exporting their cowardly fear of the word "tax" to other countries
I bet you're glad that you were saved from the government death lists :(
these infrared sensors and controllers will soon be installed in police cars and possibly in police uniforms if they're small enough. and anywhere else that governments, as well as the rich and powerful, want to make it harder for ordinary citizens to gather evidence of their abuses of power.
you must be an american. only yanks think that caveat emptor is an admirable business strategy rather than a warning against scumbags.
only drug traffickers use bitcoin.
only criminals care about privacy.
only paedophiles use encryption.
only terrorists use p2p.
cue the sound of a million angry bloggers inventing insults for the king of thailand.
there's this term called 'fair use' that you need to learn about. it specifically allows that sort of copying for academic purposes.
it's part of what we, the people, retained when we authorised the creation of socialist monopoly copyright laws to restrict our natural right to read and copy whatever information want to.
the lawsuit isn't a complaint that fair-use rights are being exceeded. it is an attempt to undermine - or completely destroy - existing fair-use rights.
Certainly, governements have legitimate reasons to want to keep things secret. I don't think anyone really disputes that.
What some people seem to forget is that a government's desire to keep secrets doesn't make it a crime for citizens - or even third parties - to want to find out or even try to find out....and as long as they commit no actual crimes like trespass or espionage while trying to find out, doing so is perfectly legal, legitimate, and moral
FWIW, I think citizens have an absolute, inalienable right to find out through any and all legal means what their government is up to if they want (and i don't think it's particularly wrong if they're willing to commit and bear the consequences of committing *minor* non-violent crimes in the process). remember, a democratic government - or even one that pretends to democratic processes - is the servant of the people, not the master.
any laws that attempt to criminalise the exercise of that right are bad laws, deserving contempt and civil disobedience
or they don't trust the new and enthusiastic "volunteers" with secret agendas due to the inevitable infiltration attempts from journalists, spook agencies, agent provocateurs, reactionaries with some kind of point to prove, and the occasional random lunatic.
a $20M penalty clause probably doesn't mean much to a real spook or even to a journo with a legal team, but it probably discourages the independent infiltrators
the toxicity of sugar (sucrose, glucose, fructose, etc) is one of things that almost no researcher in the know dares to mention publicly because it would be career (and funding) suicide. the processed food industry is far too powerful a lobby group.
but the researchers know. check medline. almost every research article on diabetes begins with words to the effect of "fed the rats sugar until they developed diabetes". feeding rats sugar is THE consistently reproducible method of inducing diabetes.
and this is what the processed food lobby is doing to consumers every day with sugar in absofuckinglutely everything. even things you think wouldn't have sugar because they're supposed to be salty or sour or savoury or anything-else-but-sweet have sugar in them. because it's cheap, it's addictive (esp. to children and adults with poor impulse control - i.e. most of the population), and it's a preservative.
sugar in our diet isn't bad when it's rare and unrefined (as it is in fruits and vegetables etc. and in our natural pre-agriculture diet it WAS rare, but it was a huge amount of easily absorbed energy which is why we evolved the ability to taste sweetness...and why we also evolved to *like* it). even when humans first discovered processed sugar from sugar cane a few hundred years ago it wasn't a huge problem because it was very expensive (like all spices were) - only the rich could afford it.
even the improvement of refinery processes that made sugar became extremely refined and extremely cheap wasn't that bad....it was only when "food" factories started putting it in *everything* so that it became almost impossible to avoid eating far too much of the stuff that it became a problem.
and this, btw, is also why the poor (and the time-poor) suffer from diabetes more than the rich do - the rich can afford to eat well. the poor can't (money-wise AND time-wise).
ummm....who cares? why is their desire for incentives so important?
there are plenty of "content producers" willing to make stuff for free.
the most appropriate response would be to create an app to "Cure Heterosexuality" or "Cure Christianity".
i'm a gnome user (in that i use the gnome panel[1] and most of the GUI apps i use are gnome) but i switched to the openbox window manager several years ago because the "standard" gnome window managers metacity and mutter absolutely suck. even worse than the software is the attitude ("design philosophy" if you want to be fancy) behind them - that taking away features is a Good Thing because users are too stupid to understand them and easily confused by choices.
this latest idiocy is just an extension of the initial practical reason i started using openbox - middle-click and right-click on the maximise button for vertical-only and horizontal-only maximise of a window. IIRC, after some argument a few years ago, the metacity devs agreed to add (or keep, i can't remember) the vert-max and horiz-max features, but refused to enable them via middle- or right- click on the max button....the ONLY way to access them is to manually configure the keyboard bindings to assign a key to them.
the gnome terminal, actually vte, also has an annoying broken-by-design bug of sending eight up/down arrow keys to the application running in the terminal when the scroll wheel is moved. the devs flat out refuse to acknowledge that this causes problems for programs like mutt and vi, and refuse to fix, and messes up middle-button pasting (because the scroll wheel is usually also the middle button, and it's almost impossible to click it without scrolling it a little at first) - even though several patches have been submitted over the years that the bug has been in the gnome bugzilla. because of this, i use mrxvt rather than gnome terminal - which, of course, has its own bugs but at least the dev doesn't suffer from the Gnome Developer Attitude Problem.
gnome has a lot of good software, but it also has a lot of rage-inducing idiocy like the above.
[1] the gnome panel annoys me too - buggy bloated crap. i've looked around for something to replace it with but haven't found one yet.
the trouble is that it *doesn't* fix the problem, it just hides it.
rebooting - or even re-imaging - can be fine as a quick-and-dirty fix to get critical infrastructure back up and running as quickly as possible, but is IS NOT and CAN NOT be a substitute for actually figuring out what caused the problem and fixing it so that it doesn't happen again
what makes you think I'm in the US? I'm not. I live in Australia, and (as in most of the rest of the world) almost all mobile phones here are sold indirectly through bundling deals with telcos.
I haven't read or even heard of the book before, but just reading the wikipedia article about the author confirms my initial thought that a *very* short summary of her thesis might be that the actual crime of child abuse may be compounded many times over by the constant barrage of people telling the victim that they ought to be severely traumatised actually makes them far more traumatised than they would otherwise be, in a manner reminiscent of the way that deluded, obsessive, or even outright malicious "therapists" *create* False Memory Syndrome in susceptible people.
i.e. that the crime of abuse may have happened, but that the trauma may be partly or even wholly created after the fact.
it's not even a particularly surprising or unusual thesis...the Satanic Ritual Abuse moral panic of the 1980s clearly shows that it's possible to create such trauma even in children who *haven't* been sexually abused.
damn. clicked 'submit' rather than 'continue editing'.
i meant to add that it's in the telcos' interests for mobile phone prices to stay high. It makes it much easier for them to trap people into long-term contracts. if "current" phone prices were only $100 or $200 rather than $600-$900, there's be a lot more people who just bought their phones outright and shopped around for the best voice&data deal.
in short: phone prices are artificially high just to make the bundled phone+overpriced plans seem a much better deal than buying the phone outright.
That's because the end-user isn't really the manufacturer's customer. The telcos are. Almost all mobile phones are sold by Telcos (or their agents) in a bundle with a voice and/or data plan.
This has an enormous anti-competitive effect on the market. More than enough to counteract the pricing trends in the direct-to-consumer electronics markets
It's also why many features that end-users would like (e.g. an answering machine app, an an alternative to the telco-provided voicemail) are missing. There's no great incentive to create them...and, for some features, a huge disincentive if they're the kind of feature that might piss off a telco
where's the harm in that? it's got H2 and O in it!