You know what I miss? I miss being able to type text input as fast as my hands could go, without an hourglass or hanging pause. And it's not like I type any faster now than I did on a C64.
Admittedly, my experience is a few years out of date, but it used to be that the immediate answer to this question was Scyld, the direct descendant of the original "Beowulf" cluster created at Goddard Space Flight Center by Donald Becker. We used it for 3d rendering and video processing and it was really slick, and being based on RHEL it was easy to get people who knew how to work on it/software updates/support in forums, etc.
I've only seen one comment in support of Scyld here, has it fallen out of favor for some reason?
I'm really thinking that this article is leaving some very important details out... It's really strange that a money-making data center would have physical space as it's primary limiting factor. Things like power, cooling, network, etc are usually far more important than square feet of tile, especially when anyone with an experience in data centers isn't going to put it in a high-value real estate market, it's going to be out in some industrial/commercial zone in the burbs where land/power/water are cheaper. It's not like the developers sending programs/renders to the cluster need to be anywhere near it physically.
I'm guessing these folks are addressing some sort of unique problem that they have to solve this way, and they don't bother to explain that to us in the article.
What are you talking about? Many corporations now use internal webmail systems, and web front ends to things like Outlook, ticketing systems, inventory, etc, plus web-based document storage.
Hell, I bet your bank runs a web front end to the tellers and managers.
Just because something is web based doesn't mean it's sitting outside of your control.
This needs to get modded up. If you don't have a degree, you can't get past the filter system that HR puts in front of managers at most corporations. HR doesn't care/know if you can do the job, they just have a list of checkboxes that need to be filled before they pass the resume to anyone hiring, and a degree is almost always on that list of checkboxes.
But if you KNOW the managers, or someone who works with them, you can get your resume past the HR filter. Also, if someone the manager trusts 'vouches' for you, it gives them some comfort that they don't get from someone coming in from outside. In this case, a degree isn't important if you can do the job.
A great way to get known is to work in phone support in smaller companies that do support in-house (a call center won't do), where you'll meet the ops/dev staff in the halls, and usually don't need a degree. If you work hard at improving the processes in place, and do it well, it will get noticed. The NOC is also a good place, since they're often desperate for staff that will work midnight on Saturday, and you'll have access to learn a lot about operations functions that outside hires with degrees can't match.
The bottom line is that a degree is a checkbox on a form to an HR manager, while an operations manager wants someone who can do the work. If you can get past the first, you can get a really well paying job with the second if they know you can do what they need.
Unless you happen to (really, not just-in-case) need images from crash-dumps or if you have some sort of suspend-to-disk system, you should NEVER use swap.
Ever sit and listen to Windows randomly chug the disk for a while? Often, that's it doing stupid crap with swap when you want it to be working on something else. Swap is SOOO slow that even scenarios where the OS tries to shove stuff it thinks you'll never use again, which should be a good idea, results in you sitting and waiting while that happens at really irritating times.
On servers, it's much worse. A host constantly hitting the disk for swap will have much higher power and cooling costs, and the life of the drives (although still very long) will be reduced. Using anything mechanical you don't have to is bad.
Just buy more RAM. Anything arguments against are edge cases that you should engineer yourself out of anyway.
The article says you would be creating your API, not restricting you to the APIs of Direct X or OpenGL. Software houses would probably build/trade/sell CUDA implementations much like they do engines these days.
Try Crossfit. It's an open-source style method of fitness where the workout of the day is posted every day, using (mostly) simple equipment, is easily scaled to your needs, and is meant to work your whole body instead of focusing on just cardio or just weights.
My experience (which admittedly may vary) is that you can do these exercises at home, or go to a gym/dojo that offers Crossfit and get a very affordable trainer in a very easygoing atmosphere. Usually there are lots of other beginners so there's very little macho BS and everyone is helpful.
Yes. That's it, that's exactly what he was asking.
OH, if you're looking for an ANSWER, that would be that an SSD takes on average 50% of the power, but perhaps 1/3 to 1/5 of the capacity of a similar form-factor hard drive. Meaning that per-drive they use less and per gig they use more.
All of these associations of Google with 1984 and the Panopticon leave out very important facets of the surveillance... Google records public spaces and ANYONE can look at Google and see what's up there, providing many useful benefits. The other two's monitoring target public AND private areas and are tools of an authority, and keep their results in the hands of that authority solely as a tool for control.
If Google Streets can be kept open and records data in the public sphere, the benefits for other uses will far outweigh it's value as a tool of oppression.
Next we'll be talking about harnessing unused RAM on all those workstations.
Unless for some reason there are extra disks on these hosts, it's not worth the effort of trying to access/lock/manage security/etc for storage that you'd have to access across a network, especially when users could change/reboot hosts and the increase in nodes overexposes you to failures.
Just buy disks of a size/speed/cost that's appropriate to give them room for anything that needs to be installed locally and set up a SAN or NAS for saving their files, where you can manage bulk storage most effectively. If you're feeling that you can invest a lot of time in the name of storage efficiency, give users individual iSCSI virtual drives on a NAS.
Except that IP addresses aren't domain names, they are owned by ARIN and allocated to companies, and ARIN is perfectly capable of taking them back if they're unused.
Missiles have to carry their propulsion units and fuel with them, adding to projectile weight, cost and complexity, and don't go anywhere near mach 8 (most aren't even supersonic). The value of the speed of railgun projeciles against high value land targets in advance of Marines is mentioned in the article, so they must be guided in some way.
Oh come on, by that argument I should be forced to load into RAM all the fonts for every written language and speech recognition programs for all of them.
If a 'feature' isn't used by the majority of users frequently, it shouldn't be using system resources. Just like you don't put data that doesn't get frequent hits in a cache.
>>If everyone did this then the big power companies would see a dent in their bottom line and so they start spreading lies like this.
They could really care less, except that this might take some of the pressure on them to risk capital in alternative power. It's not like we don't suck up everything the power companies can produce and demand more.
If Intel had a chipset that both AMD and Intel could use, and AMD used it, they would gain a great deal. Like selling more chipsets to motherboard manufacturers and getting a piece of even AMD systems, dictating the future of the bus, memory, and form factors. Plus, even if there wasn't an actual performance benefit (and there probably would be since Intel would have made the design, and have that intelligence in-house), they could easily give the impression that running Intel chips on Intel hardware was 'better'.
But will AMD bite? Is working with Intel chipsets cheap enough that it makes it worth it to lose maintaining it's own sockets and bus?
And, if it bites, will Intel turn around in six years and lock them out of the next bus, forcing them to recreate (an re-capitalize) the means to start over?
Seems like a good move for Intel even if it just gets them the Via chips. A bad move for AMD if they fall for it.
I owned just about every home computer of that era, and the 800 was definitely second best to the C64. It should have been, it was much older. A steel frame only counts for so much.
On the other hand, there really isn't anything in this article about what the new 'Commodore' gaming computers really are... and it sounds like just more leeching off of a dead name.
Whomever marked this as offtopic was a little quick on the gun. I believe the coward is referring to PAR files, a method of breaking up data and reassembling it commonly used on newsgroups.
Wouldn't it be easier to rig up an armature using old joystick potentiometers at each joint than use three cameras and try to extrapolate position data from that? You could easily have a single machine running a perl script collect the position data from all the sensors at one time into a file and then write it out as bvh/whatever or convert it later.
AOL is perfectly aware of how bad their rep is. That's why everything they do now is branded 'AIM'. Like OpenAim to allow third parties to make AIM clients, and AIMPages, their new blogging software. They also have AIM mail accounts on the AOL Webmail system, which uses the same spam filter and storage system (which may have been lame in 2002, but now is a leader in features and capability) as the in-client AOL mail service.
Of course, they tried to do this with Netscape (as their answer to NetZero) too, and failed to support it properly, and it never prospered. Hopefully, now that it's do or die, they won't fail to rebrand again.
Slashdot Mirror
You know what I miss? I miss being able to type text input as fast as my hands could go, without an hourglass or hanging pause. And it's not like I type any faster now than I did on a C64.
Admittedly, my experience is a few years out of date, but it used to be that the immediate answer to this question was Scyld, the direct descendant of the original "Beowulf" cluster created at Goddard Space Flight Center by Donald Becker. We used it for 3d rendering and video processing and it was really slick, and being based on RHEL it was easy to get people who knew how to work on it/software updates/support in forums, etc.
I've only seen one comment in support of Scyld here, has it fallen out of favor for some reason?
And, because it's in Java in a browser, it's running choppy on my 4 core i7 with 12gb of ram. Kudos, you've made a pdp game bloated and unplayable.
I'm really thinking that this article is leaving some very important details out... It's really strange that a money-making data center would have physical space as it's primary limiting factor. Things like power, cooling, network, etc are usually far more important than square feet of tile, especially when anyone with an experience in data centers isn't going to put it in a high-value real estate market, it's going to be out in some industrial/commercial zone in the burbs where land/power/water are cheaper. It's not like the developers sending programs/renders to the cluster need to be anywhere near it physically.
I'm guessing these folks are addressing some sort of unique problem that they have to solve this way, and they don't bother to explain that to us in the article.
What are you talking about? Many corporations now use internal webmail systems, and web front ends to things like Outlook, ticketing systems, inventory, etc, plus web-based document storage.
Hell, I bet your bank runs a web front end to the tellers and managers.
Just because something is web based doesn't mean it's sitting outside of your control.
Sure you can, so long as they don't conflict with the previous licenses.
Dude. I tried to mod you funny and I fat-fingered you to overrated. So I'm replying to hopefully un-mod you. My apologies.
This needs to get modded up. If you don't have a degree, you can't get past the filter system that HR puts in front of managers at most corporations. HR doesn't care/know if you can do the job, they just have a list of checkboxes that need to be filled before they pass the resume to anyone hiring, and a degree is almost always on that list of checkboxes.
But if you KNOW the managers, or someone who works with them, you can get your resume past the HR filter. Also, if someone the manager trusts 'vouches' for you, it gives them some comfort that they don't get from someone coming in from outside. In this case, a degree isn't important if you can do the job.
A great way to get known is to work in phone support in smaller companies that do support in-house (a call center won't do), where you'll meet the ops/dev staff in the halls, and usually don't need a degree. If you work hard at improving the processes in place, and do it well, it will get noticed. The NOC is also a good place, since they're often desperate for staff that will work midnight on Saturday, and you'll have access to learn a lot about operations functions that outside hires with degrees can't match.
The bottom line is that a degree is a checkbox on a form to an HR manager, while an operations manager wants someone who can do the work. If you can get past the first, you can get a really well paying job with the second if they know you can do what they need.
Unless you happen to (really, not just-in-case) need images from crash-dumps or if you have some sort of suspend-to-disk system, you should NEVER use swap.
Ever sit and listen to Windows randomly chug the disk for a while? Often, that's it doing stupid crap with swap when you want it to be working on something else. Swap is SOOO slow that even scenarios where the OS tries to shove stuff it thinks you'll never use again, which should be a good idea, results in you sitting and waiting while that happens at really irritating times.
On servers, it's much worse. A host constantly hitting the disk for swap will have much higher power and cooling costs, and the life of the drives (although still very long) will be reduced. Using anything mechanical you don't have to is bad.
Just buy more RAM. Anything arguments against are edge cases that you should engineer yourself out of anyway.
The article says you would be creating your API, not restricting you to the APIs of Direct X or OpenGL. Software houses would probably build/trade/sell CUDA implementations much like they do engines these days.
Compared to even today's bargain hardware, stuff 5-6 years old doesn't even have the processor power to justify the electricity/waste heat/noise.
Try Crossfit. It's an open-source style method of fitness where the workout of the day is posted every day, using (mostly) simple equipment, is easily scaled to your needs, and is meant to work your whole body instead of focusing on just cardio or just weights.
My experience (which admittedly may vary) is that you can do these exercises at home, or go to a gym/dojo that offers Crossfit and get a very affordable trainer in a very easygoing atmosphere. Usually there are lots of other beginners so there's very little macho BS and everyone is helpful.
Yes. That's it, that's exactly what he was asking.
OH, if you're looking for an ANSWER, that would be that an SSD takes on average 50% of the power, but perhaps 1/3 to 1/5 of the capacity of a similar form-factor hard drive. Meaning that per-drive they use less and per gig they use more.
All of these associations of Google with 1984 and the Panopticon leave out very important facets of the surveillance... Google records public spaces and ANYONE can look at Google and see what's up there, providing many useful benefits. The other two's monitoring target public AND private areas and are tools of an authority, and keep their results in the hands of that authority solely as a tool for control.
If Google Streets can be kept open and records data in the public sphere, the benefits for other uses will far outweigh it's value as a tool of oppression.
Next we'll be talking about harnessing unused RAM on all those workstations.
Unless for some reason there are extra disks on these hosts, it's not worth the effort of trying to access/lock/manage security/etc for storage that you'd have to access across a network, especially when users could change/reboot hosts and the increase in nodes overexposes you to failures.
Just buy disks of a size/speed/cost that's appropriate to give them room for anything that needs to be installed locally and set up a SAN or NAS for saving their files, where you can manage bulk storage most effectively. If you're feeling that you can invest a lot of time in the name of storage efficiency, give users individual iSCSI virtual drives on a NAS.
Except that IP addresses aren't domain names, they are owned by ARIN and allocated to companies, and ARIN is perfectly capable of taking them back if they're unused.
Missiles have to carry their propulsion units and fuel with them, adding to projectile weight, cost and complexity, and don't go anywhere near mach 8 (most aren't even supersonic). The value of the speed of railgun projeciles against high value land targets in advance of Marines is mentioned in the article, so they must be guided in some way.
Let me amend that, it shouldn't be using resources until requested to do so. Of course making the feature unavailable isn't reasonable.
Oh come on, by that argument I should be forced to load into RAM all the fonts for every written language and speech recognition programs for all of them.
If a 'feature' isn't used by the majority of users frequently, it shouldn't be using system resources. Just like you don't put data that doesn't get frequent hits in a cache.
>>If everyone did this then the big power companies would see a dent in their bottom line and so they start spreading lies like this.
They could really care less, except that this might take some of the pressure on them to risk capital in alternative power. It's not like we don't suck up everything the power companies can produce and demand more.
If Intel had a chipset that both AMD and Intel could use, and AMD used it, they would gain a great deal. Like selling more chipsets to motherboard manufacturers and getting a piece of even AMD systems, dictating the future of the bus, memory, and form factors. Plus, even if there wasn't an actual performance benefit (and there probably would be since Intel would have made the design, and have that intelligence in-house), they could easily give the impression that running Intel chips on Intel hardware was 'better'.
But will AMD bite? Is working with Intel chipsets cheap enough that it makes it worth it to lose maintaining it's own sockets and bus?
And, if it bites, will Intel turn around in six years and lock them out of the next bus, forcing them to recreate (an re-capitalize) the means to start over?
Seems like a good move for Intel even if it just gets them the Via chips. A bad move for AMD if they fall for it.
I owned just about every home computer of that era, and the 800 was definitely second best to the C64. It should have been, it was much older. A steel frame only counts for so much.
On the other hand, there really isn't anything in this article about what the new 'Commodore' gaming computers really are... and it sounds like just more leeching off of a dead name.
Whomever marked this as offtopic was a little quick on the gun. I believe the coward is referring to PAR files, a method of breaking up data and reassembling it commonly used on newsgroups.
Wouldn't it be easier to rig up an armature using old joystick potentiometers at each joint than use three cameras and try to extrapolate position data from that? You could easily have a single machine running a perl script collect the position data from all the sensors at one time into a file and then write it out as bvh/whatever or convert it later.
Something like Gypsy, only lots cheaper.
AOL is perfectly aware of how bad their rep is. That's why everything they do now is branded 'AIM'. Like OpenAim to allow third parties to make AIM clients, and AIMPages, their new blogging software. They also have AIM mail accounts on the AOL Webmail system, which uses the same spam filter and storage system (which may have been lame in 2002, but now is a leader in features and capability) as the in-client AOL mail service.
Of course, they tried to do this with Netscape (as their answer to NetZero) too, and failed to support it properly, and it never prospered. Hopefully, now that it's do or die, they won't fail to rebrand again.