Only one really springs to mind right now: Dell Precision 17 7000 Series.
For the same money as a tricked-out MacBook 'Pro' you get a Xeon CPU with 64GB of ECC RAM, and plenty of useful configuration and IO options.
Also: next-day on-site service for 3 years, extendable to 5 for a bit more money.
Having used a laptop for paid work in the field I can tell you for sure that not having undiagnosable OS crashes or silent data corruption is, y'know, professional.
Having the ability to call out an engineer with parts to fix the damn thing if it breaks while you're on-site at a client location and thus get working again quickly? That's also professional.
Having a sleek Apple MacBook. Sure, that conveys the impression of professionalism, but right now, that's all it is.. superficial. It's finely crafted consumer-grade appliance tech undeserving of the 'Pro' moniker right now.
Historically TRIM was a smart addition to block-storage commands in the enterprise SAN space, to better enable things like thin-provisioning, which at least at the time went with server virtualisation like eggs go with bacon.
It just-so-happened that TRIM also fulfilled a very similar, and similarly smart, use-case on flash-based SSDs of all levels soon after/at the same time.
Given the characteristics of the current flash devices in SSDs, it's a perfectly reasonable thing to add to filesystems, SANs and drives. It's vitally important that it be correctly implemented, but frankly, that's true for everything in IT.
Lots of 'not's there.. sometimes it's a good thing to leave those things open - to open up the board, and get an opportunity for yourself from your opponent's next play. I've seen boards shut down to a stalemate by adhering to those 'not' rules.
Way more than 3 - 'architecture' surely is a composite of many separable things including: supported instructions, no.&size of registers, no and types of functional units, width and speed, and latency of internal interconnects, pipelining and speculation capacity, HT/SMT functionality (or not), no of caches, cache styles, sizes, speeds, etc., no and sophistication of memory controllers, memory type, , the number, speed and widths of other off-chip interconnects (parallel multi-resident busses or serial links), etc., etc.
It's a bit like the toothpaste market.. throw enough brands and variants out there to ensure that you have something to appeal to everyone so that someone picks one of your products instead of the other guy's. There aren't really that many companies making toothpaste, but they all have several brands and a bunch of options.
Seriously, with the combination of North Sea oil and high fuel taxes, making the motorists stop and go at every set of lights by making sure they were deliberately out of sync seemed like an easy and inoffensive way to bring in tax revenue without hurting anyone.
Only recently have they permitted traffic regulators to synchronize the lights for the benefit of motorists, society, the environment, and utlimately the tax coffers too.
Systems thinking.. meh, they haven't even heard of it.
If we presume that this position were true, how would you explain the fact that commercial pharma companies have invented, and continue to invent new antibiotics, antifungals and particularly new vaccines which are intended to actively *prevent* illness and death, including some common causes of cancer like HPV?
Frankly there's a shitload of money in cures. And keeping patients alive by prevention, cure or treatment means you can sell them other stuff later - so they're all good.
But more than that - it's the right thing to do and it's often personal.. mostly the individuals doing the research really want to help people. You don't become a medical/pharma scientist for the flashy lifestyle (hah) or the opportunities to be evil. You may doubt it but the regulators (science, medical, fianncial, etc.) are actually all pretty much down on evil. And let's face it there are lots of opportunities to be evil without getting an extensive, expensive, education, and working in a field that's subject to public opporbrium at every turn even while you daily work for the public's health and welfare. Quite a lot, probably most of the researchers go into the field because it's a rare chance to make a positive difference to large chunks of humanity.
Remember the people who work in pharma are just that - people. They are also patients themselves or relatives/friends of patients, they're users of their own products, as well as being employees, shareholders, perhaps managers and members of broader society.
They also, somehow, have to find ways to do all that good while running a financial machine called a company - so they have to pick and choose where they put their effort to both do good and self-sustain that operation.
That's why government funding and foundations like the Bill and Melinda Gates foundation are so important - where some disease area couldn't become a sustainable financial prospect for a pharma company, those funders can help to get the science done by drawing on money generated elsewhere.
Nuhuh. This is because the BSD license is semantically freer than GPL in precisely this case:
Apple are free to release their putative fix to the community, or not - their free choice. That's one more freedom, relative to being obliged to release any changes they make which lead to a binary release outisde of Apple, which the GPL would oblige.
There are plenty of folk who see that as a feature not a flaw.
Bad if he hadn't found anything, yes, but if he'd found 50,000 ways that didn't work, that would've been fine. Seriously. Even better if he'd published a report of some kind about all the things that hadn't worked and why.
As for why he gets the patent, well, inventors are named on patents, that's pretty much how it works, but the rights to exploit the patent are probably more complex, and possibly equitable, depending on all the factors cited.
Sure, under a relatively benign government, with relatively benign police this surveillance, and the laws that mandate it and hold no enforcable checks against abuses aren't hurting us much.
Yet.
It only takes a little more zeal or stupidity in these places to lead to genuine abuses. And then the lack of checks will become nasty.
And if we get a government which really distrusts its people, and has a strong leader, well, it's happened too many times before in too many places in too many ways to claim that we'll be immune.
We may have invented modern democracy, justice and liberty, but that won't stop us throwing it all away piece by tiny piece if we're not careful. And right now we're not being careful.
Patents, great, I'd love a good lawyer to shake one of them down:
Lawyer: 'So, when and how did you invent this gene?' Holder: 'Uh, well, uh, we more sorta found it, y'know, in us.' Lawyer: 'Oh, and how did you find it?' Holder: 'Uh, with some techniques invented by someone else.' Lawyer: 'Oh right, and these techniques are only available to you, and not used by other experts in the field?' Holder: 'Uh, not really, almost everyone's doing it now.' Lawyer: 'Right..'
Implants constantly monitoring blood chemical levels and reporting to your PDA would be kinda nice.
A tattoo on your arm that responds chromatically to insulin, sugar, etc. etc. levels in the blood and that could be read by something like a barcode reader would be good. Doesn't even have to be visible to humans, just the machines.
The whole puncturing yourself to get at real actual blood thing is Not Good(tm).
I personally use green on black, with a red cursor.
There is method in this:
1) it uses a single gun on CRTs which means there are fewer alignment problems even on shonky monitors,
2) it's typically the brightest phosphor, on CRTs, and I think the brightest filter on LCDs too, to my eyes,
3) the human eye picks up green very well, (might explain (2) to some extent),
4) picking a single colour means spectacle lens-wearers don't get chromatic aberrations which arise when looking at an angle through the lens.
The other thing I'd say is, pay extra if necessary for spectacle lenses with the anti-reflective coating. Ambient lighting glare on normal specs is a nightmare especially if you're using a screen all day and the coating does work.
Ironically Opteron would be a less good fit than, say, Xeon, for the engineering that SGI have done for MIPS and Itanium, exactly because it has native support for ccNUMA and has integral memory controllers.
Someone else pointed out the scaling numbers. Opteron scales better than 8 CPUs. 8CPUs is what you can do without glue chipsets, which is pretty darned great.
Newisys have a chipset that extends the CC and addressing of Opterons so that you can put upto 32 in a system.
When dual opterons are available, that'll be 64 cores in a single system, which is where Altix was about a year ago.
This is all reimplementation of stuff that SGI are already doing with CPUs that do all their memory access over the same buses as they do everything else.
So we're very unlikely to get SGI goodness and Opteron goodness in the same box any time soon. Which is a little sad, but no biggy really.
Xeons kick butt too - the top Xeon, Opteron and Itanium performance numbers and prices (for server use remember) are actually surprisingly close given they're all clean-sheet approaches wrt each other.
You can rest assured that if Itanic is killed off the SGI line would jump straight over to Xeon.
The next generation of Xeon and Itanic will have compatible buses according to the current roadmaps, and SGI would haev much less reengineering to do than if they wanted to use Opteron.
SGI's current plans are far more interesting then just coping with CPU supply changes though - they want to add vector processors and FPGAs into the mix.
Slashdot Mirror
Only one really springs to mind right now: Dell Precision 17 7000 Series.
For the same money as a tricked-out MacBook 'Pro' you get a Xeon CPU with 64GB of ECC RAM, and plenty of useful configuration and IO options.
Also: next-day on-site service for 3 years, extendable to 5 for a bit more money.
Having used a laptop for paid work in the field I can tell you for sure that not having undiagnosable OS crashes or silent data corruption is, y'know, professional.
Having the ability to call out an engineer with parts to fix the damn thing if it breaks while you're on-site at a client location and thus get working again quickly? That's also professional.
Having a sleek Apple MacBook. Sure, that conveys the impression of professionalism, but right now, that's all it is.. superficial. It's finely crafted consumer-grade appliance tech undeserving of the 'Pro' moniker right now.
*So* not a kludge. o.O
Historically TRIM was a smart addition to block-storage commands in the enterprise SAN space, to better enable things like thin-provisioning, which at least at the time went with server virtualisation like eggs go with bacon.
It just-so-happened that TRIM also fulfilled a very similar, and similarly smart, use-case on flash-based SSDs of all levels soon after/at the same time.
Given the characteristics of the current flash devices in SSDs, it's a perfectly reasonable thing to add to filesystems, SANs and drives. It's vitally important that it be correctly implemented, but frankly, that's true for everything in IT.
Lots of 'not's there.. sometimes it's a good thing to leave those things open - to open up the board, and get an opportunity for yourself from your opponent's next play. I've seen boards shut down to a stalemate by adhering to those 'not' rules.
Way more than 3 - 'architecture' surely is a composite of many separable things including: supported instructions, no.&size of registers, no and types of functional units, width and speed, and latency of internal interconnects, pipelining and speculation capacity, HT/SMT functionality (or not), no of caches, cache styles, sizes, speeds, etc., no and sophistication of memory controllers, memory type, , the number, speed and widths of other off-chip interconnects (parallel multi-resident busses or serial links), etc., etc.
It's a bit like the toothpaste market.. throw enough brands and variants out there to ensure that you have something to appeal to everyone so that someone picks one of your products instead of the other guy's. There aren't really that many companies making toothpaste, but they all have several brands and a bunch of options.
Seriously, with the combination of North Sea oil and high fuel taxes, making the motorists stop and go at every set of lights by making sure they were deliberately out of sync seemed like an easy and inoffensive way to bring in tax revenue without hurting anyone.
Only recently have they permitted traffic regulators to synchronize the lights for the benefit of motorists, society, the environment, and utlimately the tax coffers too.
Systems thinking.. meh, they haven't even heard of it.
Considerably better than the US too if wikipedia's suicide rates for 2008 are to be credited.
It so won't be more powerful than Fermi. But it might be available in industrial quantities, instead of cottage-industry amounts.
If we presume that this position were true, how would you explain the fact that commercial pharma companies have invented, and continue to invent new antibiotics, antifungals and particularly new vaccines which are intended to actively *prevent* illness and death, including some common causes of cancer like HPV?
Frankly there's a shitload of money in cures. And keeping patients alive by prevention, cure or treatment means you can sell them other stuff later - so they're all good.
But more than that - it's the right thing to do and it's often personal.. mostly the individuals doing the research really want to help people. You don't become a medical/pharma scientist for the flashy lifestyle (hah) or the opportunities to be evil. You may doubt it but the regulators (science, medical, fianncial, etc.) are actually all pretty much down on evil. And let's face it there are lots of opportunities to be evil without getting an extensive, expensive, education, and working in a field that's subject to public opporbrium at every turn even while you daily work for the public's health and welfare. Quite a lot, probably most of the researchers go into the field because it's a rare chance to make a positive difference to large chunks of humanity.
Remember the people who work in pharma are just that - people. They are also patients themselves or relatives/friends of patients, they're users of their own products, as well as being employees, shareholders, perhaps managers and members of broader society.
They also, somehow, have to find ways to do all that good while running a financial machine called a company - so they have to pick and choose where they put their effort to both do good and self-sustain that operation.
That's why government funding and foundations like the Bill and Melinda Gates foundation are so important - where some disease area couldn't become a sustainable financial prospect for a pharma company, those funders can help to get the science done by drawing on money generated elsewhere.
In a word: 'Eve' (um, 'Online' :)
That's my reading, if these patents are granted.
It also pwns LDAP, and ORACLE, and well, pretty much everything that will run on multicore CPUs and GPGPUs.
Lovely, time to take up an honest trade.
http://www.usenix.org/events/lisa00/gilfix.html
Won best paper award that year istr.
Read the diffs to BIND and work it out. They're only hiding things from the bad guys to give you a few more hours window to implement the patches.. :)
One of its running mates is both the same!
(What's the difference between a duck.. ;)
And other IT professional associations and bodies around the world have them too.
Nuhuh. This is because the BSD license is semantically freer than GPL in precisely this case:
Apple are free to release their putative fix to the community, or not - their free choice. That's one more freedom, relative to being obliged to release any changes they make which lead to a binary release outisde of Apple, which the GPL would oblige.
There are plenty of folk who see that as a feature not a flaw.
Bad if he hadn't found anything, yes, but if he'd found 50,000 ways that didn't work, that would've been fine. Seriously. Even better if he'd published a report of some kind about all the things that hadn't worked and why.
As for why he gets the patent, well, inventors are named on patents, that's pretty much how it works, but the rights to exploit the patent are probably more complex, and possibly equitable, depending on all the factors cited.
Sure, under a relatively benign government, with relatively benign police this surveillance, and the laws that mandate it and hold no enforcable checks against abuses aren't hurting us much.
Yet.
It only takes a little more zeal or stupidity in these places to lead to genuine abuses. And then the lack of checks will become nasty.
And if we get a government which really distrusts its people, and has a strong leader, well, it's happened too many times before in too many places in too many ways to claim that we'll be immune.
We may have invented modern democracy, justice and liberty, but that won't stop us throwing it all away piece by tiny piece if we're not careful. And right now we're not being careful.
I don't know about other folk, but I subscribe to these:
http://www.acm.org/constitution/code.html
http://www.sage.org/ethics.mm
Ask your IT colleagues if they've heard of them.
Patents, great, I'd love a good lawyer to shake one of them down:
Lawyer: 'So, when and how did you invent this gene?'
Holder: 'Uh, well, uh, we more sorta found it, y'know, in us.'
Lawyer: 'Oh, and how did you find it?'
Holder: 'Uh, with some techniques invented by someone else.'
Lawyer: 'Oh right, and these techniques are only available to you, and not used by other experts in the field?'
Holder: 'Uh, not really, almost everyone's doing it now.'
Lawyer: 'Right..'
What's a patent again?
Buy a Sun, or an SGI.
Implants constantly monitoring blood chemical levels and reporting to your PDA would be kinda nice.
A tattoo on your arm that responds chromatically to insulin, sugar, etc. etc. levels in the blood and that could be read by something like a barcode reader would be good. Doesn't even have to be visible to humans, just the machines.
The whole puncturing yourself to get at real actual blood thing is Not Good(tm).
I personally use green on black, with a red cursor.
There is method in this:
1) it uses a single gun on CRTs which means there are fewer alignment problems even on shonky monitors,
2) it's typically the brightest phosphor, on CRTs, and I think the brightest filter on LCDs too, to my eyes,
3) the human eye picks up green very well, (might explain (2) to some extent),
4) picking a single colour means spectacle lens-wearers don't get chromatic aberrations which arise when looking at an angle through the lens.
The other thing I'd say is, pay extra if necessary for spectacle lenses with the anti-reflective coating. Ambient lighting glare on normal specs is a nightmare especially if you're using a screen all day and the coating does work.
Ironically Opteron would be a less good fit than, say, Xeon, for the engineering that SGI have done for MIPS and Itanium, exactly because it has native support for ccNUMA and has integral memory controllers.
Someone else pointed out the scaling numbers. Opteron scales better than 8 CPUs. 8CPUs is what you can do without glue chipsets, which is pretty darned great.
Newisys have a chipset that extends the CC and addressing of Opterons so that you can put upto 32 in a system.
When dual opterons are available, that'll be 64 cores in a single system, which is where Altix was about a year ago.
This is all reimplementation of stuff that SGI are already doing with CPUs that do all their memory access over the same buses as they do everything else.
So we're very unlikely to get SGI goodness and Opteron goodness in the same box any time soon. Which is a little sad, but no biggy really.
Xeons kick butt too - the top Xeon, Opteron and Itanium performance numbers and prices (for server use remember) are actually surprisingly close given they're all clean-sheet approaches wrt each other.
You can rest assured that if Itanic is killed off the SGI line would jump straight over to Xeon.
The next generation of Xeon and Itanic will have compatible buses according to the current roadmaps, and SGI would haev much less reengineering to do than if they wanted to use Opteron.
SGI's current plans are far more interesting then just coping with CPU supply changes though - they want to add vector processors and FPGAs into the mix.
In the meantime Altix kicks butt.
HP are dropping them from their high-end workstations.
Not their high-end servers.