First thing is to make sure management understands the magnitude of the problem. Once they are on board, you need to get a policy approved that has some teeth.
Something along the lines of NO COMPANY DATA STORED ON LOCAL HARDDRIVES. Period. Put sanctions in there. Firing on second offence, depending on the magnitude of the loss, is what we had at my last job.
Make them use mapped network drives for everything. We used H: for home directories and P: for the user-public directories. Get a time-table for moving over existing data.
Get a good backup drive, like a DLT or DAT autochanger from HP. Grok backup systems. GFS (Grandfather, Father, Son) is a real good one.
Once you choose a backup system, make sure you get the annual budget approved for media. You DO NOT reuse tapes. bad habit. Also, price an off-site storage facility that does weekly pickup.
Educate the end users. Send out a company-wide e-mail with the new policy and make sure it is brought up at the next "everyone" meeting. Make sure it is part of the new-hire orientation.
Then, you will need to create a chart of end-users that you can check off as you manually go around making sure drives are clean.
You will need to periodically check to make sure there are no repeat violators. This is what the chart is for.
I hope you have balls, because some of the worst offenders will be management. You'll need to give them a deadline -- after which point you'll wipe or reimage their drive. Follow thru.
While you're at it, look into Norton Ghost so you can create images of each system that can be pushed out from the server if necessary. Also make sure there is anti-virus software on each machine and you have a central definitions distribution point (like an NT server). Update definitions no later than once a week.
Don't even THINK about trying to backup each individual harddrive. It'll never work.
This is top-notch software. Well, PRO/Engineer is, anyway. The last shop I worked at used Cadence but we had a lot of PRO/E CAD people who had come from Lockheed-Martin.
OTOH, this is not cheap software. Usually several thousand $$ a seat.
The one on 436/17-92 in front of the Jai Alai was one of them. One in Sanford on 17-92 was another. I stay away from I-Drive.:-)
If you want good selection and personal service, go to the place on 434 just west of 17-92. In a small strip, and it is bigger than it looks. A hell of a lot better organized than Sci-Fi City, but Sci-Fi also has a lot of RPG stuff.
Enterprise 1701 moved to E. Colonial and changed their name to Sci-Fi City. Still a pretty good place. They have a gaming area, lots of anime, miniatures, comics, role-playing, etc.
They're just east of 436, on the south side of the road. In the same plaza as the big Home Depot and Petsmart.
Three or four comic book stores in the Orlando area went out of business quickly about 2 years ago. While scrounging for stuff, I got to speak with the owners and they ALL had the same story.
They made more money in one month selling their inventory on Ebay than they did in a year selling inthe "real" world. They pointed out all the really good stuff was gone, and I wasn't going to find what I was looking for.
All were quite happy with the situation and planned to continue selling at online auctions.
A side note is that in the last week I've sold 5 books on Amazon that I no longer wanted. I got decent money, too, not like the $1 or so at a garage sale. I *HATE* throwing books out -- they need to go to a good home.
Most people are idiots when it comes to technology, that isn't a surprise. Look back when cable modems first started to take off and you'll see lots of stories of people running PC Anywhere without a password, or using Windows File Sharing and sharing their entire drive.
Computers are complicated devices. Unless they are stripped down to do only one or two functions, like a play-only VCR, the majority of the public will not understand. Many of them don't WANT to understand -- they just want their e-mail, IM, MP3s and pr0n.
Case in point -- KaZaA. It is KNOWN spyware, and has an embedded secondary network (Britewave?) yet despite this being well publicized (CNN, FoxNews, regular geek news like Slashdot) it is wildly popular.
Why? It is *very* convenient, and people will put up with a ton of shit for convenience.
What would be a real interesting study, is get this one publicized as all get out then do it again in 1 year. I bet the stats would be about the same.
You can make your own UL-based distro, but you can NOT use the trademarked "Powered by United Linux" or probably anything similar.
"Compatible with United Linux-based distros" would probably be fine.
Think of UL as a formalized LSB implementation. The addons might be proprietary (like an MS Exchange client, Lotus Notes client) or it might be bundled with something like Oracle, SAP or UniCenter.
The config used was a Smoothwall Linux install with Apache on a non-standard (high) port. No mail (how does the server report problems), no FTP/SSH (how do you update files on the server), no nothing.
That isn't real world.
As far as the "honeypot" goes, that is utter bullshit.
Exactly what DDR bottleneck are you talking about? Specifically, where Windows has a workaround and Linux doesn't.
Memory bandwidth has always been a bottleneck on systems, and it probably always will. However, this is a hardware issue and not an OS one, as far as I understand it.
Finally, an admin "in the know" that chose a Mac server was probably on drugs. PPC hardware is very nice, but the extra cost -- not to mention the skills needed to deal with the lack of server software... a questionable choice for any serious environment.
This should make companies like IBM that officially support 3 or 4 different Linux versions happy. This should consolidate things and make life a little easier.
OTOH, is this going to be like the OPEC of Linux? They "standardize" on one distribution in public, claiming to fight the common enemy but in private they still stab each other in the back and snipe at each other?
In defense of databases, they are probably the single most scalable, performance-tuned app in existance. LOTS of people put LOTS of money and LOTS of effort into addressing database performance.
Yes, mega-databases require high priests to manage properly but nothing beats Oracle, DB2, Sybase and the like for massive data storage, retrieval and searching.
Add in proper asynch I/O, raw partition access, transaction support, dedicated monitoring and backup engines and you have a system that is damned hard to beat for large mail storage.
From the brief bios, and Sequent pedigree, it looks like there is a lot of focus on high-end features like NUMA, async I/O and the like.
Other commercial organizations, notably SGI, are also putting forth effort in those areas. There is actually quite a bit of overlap.
Since these are "open source" projects, do you collaborate with your traditional "enemies" such as SGI and Sun on Linux? What is your management's attitude towards that type of collaboration? If not, do you "look" at the work the others are doing in comparison to what you are doing?
One argument for the GPL and against "look alikes"
on
Debian And WineX
·
· Score: 5, Insightful
If you want it to be truely free, use the GPL license. If you don't care, then use the BSD, Artistic, X11 or what-have-you. This is a good example of what can happen.
This is why I bought a PlayStation 2 -- NOTHING is free, and I don't expect any of it to be, so I'm not disappointed. I can just sit down and PLAY GAMES without making moral decisions.
Word is Spiderman cost 30% - 40% more than Star Wars to make. It also opened on 1,000 more screens than Star Wars (meaning another 1,000 or so duplicates to make and distribute).
Star Wars opened mid-week (Thursday), instead of on a Friday.
Star Wars cost about 1/2 of what Spiderman did to advertise.
Star Wars was/is pulling in about DOUBLE what Spiderman was/is on a per-theater basis.
Natalie Portman is hot, but Kirsten Dunst's tits are perfection embodied.
Many exploitable holes such as these can be attributed in part to the management mentality that one or two over-worked, under trained "computer people" can handle professional system/network administration.
Frequently SysAdmins started their jobs in another field, like Engineering, and were sort of migrated over. Little formal training was given, let alone budget for. Most smaller (sub-Fortune 500) operations were more of a congealed mass than a designed network.
Then, when the LAN wasn't hooked to the Internet, and some poor schmuck install MS BackOffice and wanted to instal SMS Server, it told him he had to install SQL Server. A couple of quick clicks and you're done. Odds are, he clicked thru the admin password not thinking he'd EVER touch MS SQL other than as a backend for SMS.
Pity the new admin who inherits such a setup. You think a new admin is given time to actually check a network configuration out, much less do a proper security, performance, license audit? Nope. Get in and tell me why Outlook is saying my deleted folder is empty. I haven't emptied it since 1998 and everything was always there before when I needed it!
Stupid worms/viruses/exploits will prevail until the MENTALITY of management changes. Computers run most modern business, they are not an afterthought. The people that take care of them should be properly trained, with proper budgets. Periodic PM (preventative maintenance) needs to be allowed, scheduled and performed.
I feel pity for the admins who have to deal with these worms. I feel nothing but contempt for the management process that let them get in this position.
Well, for one example, the new RedHat 7.3.iso files have MD5 sums embedded in them. From the boot prompt, type "linux mediacheck" and it will prompt for a disk to be validated.
A feature to take detatched/attached MD5 sums, GPG signatures or the like could be pretty easily added in.
Movies or television shows? Totally different ballgame. Production costs, quality, effort involved, etc. are miniscule compared to a feature-length movie.
If you want to talk TV shows...
The current model is based off of a certain number of episodes per year, shown a week apart. Half the year gets reruns so people can catch up.
There is the problem -- people no longer need half-a-year to catch up. They can get the episodes they missed by downloading them. If the industry wants to compete, then do it with convenience.
Make a central location (i.e.- getSouthPark.com) that people can go to and d/l an episode for $5 or so. High-speed servers that make your P2P look like shit. Don't have to hunt, don't have to worry you're getting inferior VHS to DivX after-the-dog-chewed-the-tape copies, don't have to wait. Hell, $5 for ones with the commercials or $7.50 sans ads.
Lots of people would jump on that. Add a subscription service for a show -- get all the episodes sent directly to your TiVo for $50 a season. Sort of like "League Pass" with the sports.
The problem is the model is changing and the industry execs don't want to change with it. They are comfortable.
Just to point out, as the article details, a DIGITAL bootleg of Spiderman was out on the Net the day before it hit the theater. The result? The theatrical release STILL was the largest grossing opening day (and weekend) ever. Its second weekend was the largest second weekend for a movie ever. Its third weekend (this weekend) is sitting at $46 million which is, surprise, the largest third weekend ever.
Star Wars II: Attack of the Clones.
Ditto. Movie out in digital piracy a week before opening, and it still makes obscene amounts of money ($86 million this weekend and $110+ million so far).
Wanna check on the sales of Star Wars I: Phantom Menace when released on VHS/DVD? One of the best sellers; ditto for The Matrix -- both of which were floating the web in DivX format before they hit the theaters, much less DVD/VHS.
The last 4 years (1998-2001) are the best on record for revenue generated and attendence at theaters. DVD/VHS sales are thru the roof.
In the "perfect" world, where movies are uncopiable and you have to see it at the theater and/or purchase a legitimate copy, the industry would see only a paltry rise in revenue compared to today -- not the $3 Billion touted by Mr. Valenti.
Most people who get rips would either do without altogether, or wait until the DVD/VHS that THEY WERE GOING TO PURCHASE ANYWAY became available.
Hmmm. It was my understanding that the 0.5W figure for the ULV P3M was in "Deep Sleep" mode. I was also assuming that when running a task, the CPU would be a full-tilt for any of the types of applications a "supercomputer" would be needed for. I see where Intel is reporting the AVERAGE power of the unit running TYPICAL OFFICE APPLICATIONS. The problem with these measurements is the CPU is 99% idle when people are typing in word -- it doesn't matter if the CPU is running at 700 MHz or 7 MHz, you aren't going to out-type it.
The ULV P3M runs a 100 MHz bus, like the 633 Crusoe but the 677 Crusoe runs a 133 MHz bus like some of the LV P3Ms.
The final problem with the P3M is the thermal diode. To control heat, once the core CPU temp reaches a certain number (100 deg F, I think -- the "maximum junction temperature"), it clocks down to reduce heat. Again, that's fine for someone typing in Word or Excel. It can clock up for the 3 seconds needed to run that macro, but for sustained high-performance computing, it will be a problem.
I'll agree that Intel is very competitive in the laptop CPU market and their LV and ULV, SpeedStep enabled chips are great in that market -- hell, I'm typing this on an IBM laptop with a SpeedStep enabled 1 GHz P3M, and it blows the doors off the Dell P3-450 I just got rid of.
However, for sustained computing where you aren't relying on user input to clock-down between, I think the fewer transistors on the Crusoe generate a hell of a lot less heat and use lots less electricity. Transmeta has some nice thermal photos on their website, but I believe they are comparing with the "old", non-SpeedStep P3M and not any of the LV/ULV stuff.
Using this site as an example to estimate power usage, we get: 240 computer blades in Green Destiny x 6,480 hours uptime (9 months) = 1,555,200 computer hours of uptime
Assuming the only thing changed on the blade is the CPU -- and North Bridge chipset, since the Crusoe includes a North Bridge on die and the P-III does not -- at full blast the Crusoe consumes about 1.75W of power and the P-III + NB consumes between 4.5 - 8 W, depending on chip model. However, the 4.5W number is an approximation from the 0.13 micron ULV P-IIIM chip running in "Battery Saving" mode, or SpeedStepped down to 300 MHz. Running at full 700 MHz tilt, with NB, we are still talking 5.75W of power consumed.
1,555,200 * 0.0175Kw * 0.10 (dollar per KwH power cost) = $2,721.60 electricity cost/year (Crusoe) 1,555,200 * 0.0575Kw * 0.10 (dollar per KwH power cost) = $8,942.40 electricity cost/year (Intel)
A saving of approx. $6,200/year in direct electric costs.
However, the big savings comes from the heat dissipation of the units. While the newer LV/ULV P-IIIs do not require active cooling, they still run quite a bit warmer than the Crusoe units. As a result, you don't stick a rack full of them in a room that isn't temperature controlled. The difference in the air conditioning bill can easily reach tens of thousands of dollars.
In business, there are two types of money/budgets. One-time grants and acquisition budgets are large chunks of cash. Recurring expense and operations budgets are smaller. Being able to get a large chunk of cash to BUY a cluster/supercomputer is one thing. Being able to go back year-after-year and get the funds to keep it running is another project altogether. $15,000 - $20,000/year for electricity used in running/cooling computers is a LOT of money to some people. This doesn't include construction or maintenance costs on a custom facility/room.
As far as reduced administration costs go, many conventional supercomputers required chilled water and other special considerations for operation. People with experience managing things like Sun E15000s and Cray T3Es are few and far between. They are the last of the "high priesthood" of computer administrators and cost a LOT of money to employ.
A blade server, on the other hand, is a bunch of x86 computers running Linux -- nothing a couple of grad students can't learn the ins-and-outs of over a term. Maintenance contracts, spare parts, etc. are also TONS cheaper for the blade/cluster solution as opposed to high-end SGIs, Suns, Fujitsu and Cray super-computers.
Another site with a bit of good supporting information is PC Stats.
Re:I don't understand the difference, either.
on
StarOffice 6.0
·
· Score: 2, Informative
The difference?
A database program, for one. More licensed clip art and fonts, for another.
The entire article can be summed up as "MPLS is coming. Soon everything will be IP-based."
Cisco is very big in Layer-3 switching and MPLS, but their Layer-2 switches (ATM and such) are trash compared to Nortel and Lucent. They are pushing big to move large telcos to MPLS and replace their ATM and Frame Relay networks.
Slashdot Mirror
First thing is to make sure management understands the magnitude of the problem. Once they are on board, you need to get a policy approved that has some teeth.
Something along the lines of NO COMPANY DATA STORED ON LOCAL HARDDRIVES. Period. Put sanctions in there. Firing on second offence, depending on the magnitude of the loss, is what we had at my last job.
Make them use mapped network drives for everything. We used H: for home directories and P: for the user-public directories. Get a time-table for moving over existing data.
Get a good backup drive, like a DLT or DAT autochanger from HP. Grok backup systems. GFS (Grandfather, Father, Son) is a real good one.
Once you choose a backup system, make sure you get the annual budget approved for media. You DO NOT reuse tapes. bad habit. Also, price an off-site storage facility that does weekly pickup.
Educate the end users. Send out a company-wide e-mail with the new policy and make sure it is brought up at the next "everyone" meeting. Make sure it is part of the new-hire orientation.
Then, you will need to create a chart of end-users that you can check off as you manually go around making sure drives are clean.
You will need to periodically check to make sure there are no repeat violators. This is what the chart is for.
I hope you have balls, because some of the worst offenders will be management. You'll need to give them a deadline -- after which point you'll wipe or reimage their drive. Follow thru.
While you're at it, look into Norton Ghost so you can create images of each system that can be pushed out from the server if necessary. Also make sure there is anti-virus software on each machine and you have a central definitions distribution point (like an NT server). Update definitions no later than once a week.
Don't even THINK about trying to backup each individual harddrive. It'll never work.
Welcome to Hell.
This is top-notch software. Well, PRO/Engineer is, anyway. The last shop I worked at used Cadence but we had a lot of PRO/E CAD people who had come from Lockheed-Martin.
OTOH, this is not cheap software. Usually several thousand $$ a seat.
I do this with books I can't sell. Magazines in Doctor's offices, too. They need something other than "People". It's a good idea.
You have to leave a note with the book, though, or people frequently won't touch it thinking it belongs to someone.
The one on 436/17-92 in front of the Jai Alai was one of them. One in Sanford on 17-92 was another. I stay away from I-Drive. :-)
If you want good selection and personal service, go to the place on 434 just west of 17-92. In a small strip, and it is bigger than it looks. A hell of a lot better organized than Sci-Fi City, but Sci-Fi also has a lot of RPG stuff.
Enterprise 1701 moved to E. Colonial and changed their name to Sci-Fi City. Still a pretty good place. They have a gaming area, lots of anime, miniatures, comics, role-playing, etc.
They're just east of 436, on the south side of the road. In the same plaza as the big Home Depot and Petsmart.
Three or four comic book stores in the Orlando area went out of business quickly about 2 years ago. While scrounging for stuff, I got to speak with the owners and they ALL had the same story.
They made more money in one month selling their inventory on Ebay than they did in a year selling inthe "real" world. They pointed out all the really good stuff was gone, and I wasn't going to find what I was looking for.
All were quite happy with the situation and planned to continue selling at online auctions.
A side note is that in the last week I've sold 5 books on Amazon that I no longer wanted. I got decent money, too, not like the $1 or so at a garage sale. I *HATE* throwing books out -- they need to go to a good home.
Most people are idiots when it comes to technology, that isn't a surprise. Look back when cable modems first started to take off and you'll see lots of stories of people running PC Anywhere without a password, or using Windows File Sharing and sharing their entire drive.
Computers are complicated devices. Unless they are stripped down to do only one or two functions, like a play-only VCR, the majority of the public will not understand. Many of them don't WANT to understand -- they just want their e-mail, IM, MP3s and pr0n.
Case in point -- KaZaA. It is KNOWN spyware, and has an embedded secondary network (Britewave?) yet despite this being well publicized (CNN, FoxNews, regular geek news like Slashdot) it is wildly popular.
Why? It is *very* convenient, and people will put up with a ton of shit for convenience.
What would be a real interesting study, is get this one publicized as all get out then do it again in 1 year. I bet the stats would be about the same.
You can make your own UL-based distro, but you can NOT use the trademarked "Powered by United Linux" or probably anything similar.
"Compatible with United Linux-based distros" would probably be fine.
Think of UL as a formalized LSB implementation. The addons might be proprietary (like an MS Exchange client, Lotus Notes client) or it might be bundled with something like Oracle, SAP or UniCenter.
The config used was a Smoothwall Linux install with Apache on a non-standard (high) port. No mail (how does the server report problems), no FTP/SSH (how do you update files on the server), no nothing.
That isn't real world.
As far as the "honeypot" goes, that is utter bullshit.
That sounded like a Troll.
Exactly what DDR bottleneck are you talking about? Specifically, where Windows has a workaround and Linux doesn't.
Memory bandwidth has always been a bottleneck on systems, and it probably always will. However, this is a hardware issue and not an OS one, as far as I understand it.
Finally, an admin "in the know" that chose a Mac server was probably on drugs. PPC hardware is very nice, but the extra cost -- not to mention the skills needed to deal with the lack of server software... a questionable choice for any serious environment.
This should make companies like IBM that officially support 3 or 4 different Linux versions happy. This should consolidate things and make life a little easier.
OTOH, is this going to be like the OPEC of Linux? They "standardize" on one distribution in public, claiming to fight the common enemy but in private they still stab each other in the back and snipe at each other?
Good points.
In defense of databases, they are probably the single most scalable, performance-tuned app in existance. LOTS of people put LOTS of money and LOTS of effort into addressing database performance.
Yes, mega-databases require high priests to manage properly but nothing beats Oracle, DB2, Sybase and the like for massive data storage, retrieval and searching.
Add in proper asynch I/O, raw partition access, transaction support, dedicated monitoring and backup engines and you have a system that is damned hard to beat for large mail storage.
From the brief bios, and Sequent pedigree, it looks like there is a lot of focus on high-end features like NUMA, async I/O and the like.
Other commercial organizations, notably SGI, are also putting forth effort in those areas. There is actually quite a bit of overlap.
Since these are "open source" projects, do you collaborate with your traditional "enemies" such as SGI and Sun on Linux? What is your management's attitude towards that type of collaboration? If not, do you "look" at the work the others are doing in comparison to what you are doing?
If you want it to be truely free, use the GPL license. If you don't care, then use the BSD, Artistic, X11 or what-have-you. This is a good example of what can happen.
This is why I bought a PlayStation 2 -- NOTHING is free, and I don't expect any of it to be, so I'm not disappointed. I can just sit down and PLAY GAMES without making moral decisions.
Word is Spiderman cost 30% - 40% more than Star Wars to make. It also opened on 1,000 more screens than Star Wars (meaning another 1,000 or so duplicates to make and distribute).
Star Wars opened mid-week (Thursday), instead of on a Friday.
Star Wars cost about 1/2 of what Spiderman did to advertise.
Star Wars was/is pulling in about DOUBLE what Spiderman was/is on a per-theater basis.
Natalie Portman is hot, but Kirsten Dunst's tits are perfection embodied.
Teach them to do a half-and-half. Write down half, just don't stick it on the monitor. In a wallet or locked desk drawer is good.
Memorize the other half -- and make it an acronym, like ROFL! or an intentional mizspelling.
Many exploitable holes such as these can be attributed in part to the management mentality that one or two over-worked, under trained "computer people" can handle professional system/network administration.
Frequently SysAdmins started their jobs in another field, like Engineering, and were sort of migrated over. Little formal training was given, let alone budget for. Most smaller (sub-Fortune 500) operations were more of a congealed mass than a designed network.
Then, when the LAN wasn't hooked to the Internet, and some poor schmuck install MS BackOffice and wanted to instal SMS Server, it told him he had to install SQL Server. A couple of quick clicks and you're done. Odds are, he clicked thru the admin password not thinking he'd EVER touch MS SQL other than as a backend for SMS.
Pity the new admin who inherits such a setup. You think a new admin is given time to actually check a network configuration out, much less do a proper security, performance, license audit? Nope. Get in and tell me why Outlook is saying my deleted folder is empty. I haven't emptied it since 1998 and everything was always there before when I needed it!
Stupid worms/viruses/exploits will prevail until the MENTALITY of management changes. Computers run most modern business, they are not an afterthought. The people that take care of them should be properly trained, with proper budgets. Periodic PM (preventative maintenance) needs to be allowed, scheduled and performed.
I feel pity for the admins who have to deal with these worms. I feel nothing but contempt for the management process that let them get in this position.
Okay, how about:
.aa-.aj.
split -b 65m filename.iso filename.iso.
breaking the 650+ Mb iso in about ten 65 Mb chunks with the suffixes
Share them on Gnutella, KaZaA and any other P2P services.
Once downloaded, cat all the files together into one and check the MD5 sum (also downloaded, or embedded like RedHat 7.3 does).
Well, for one example, the new RedHat 7.3 .iso files have MD5 sums embedded in them. From the boot prompt, type "linux mediacheck" and it will prompt for a disk to be validated.
A feature to take detatched/attached MD5 sums, GPG signatures or the like could be pretty easily added in.
You're right, it is needed.
Movies or television shows? Totally different ballgame. Production costs, quality, effort involved, etc. are miniscule compared to a feature-length movie.
If you want to talk TV shows...
The current model is based off of a certain number of episodes per year, shown a week apart. Half the year gets reruns so people can catch up.
There is the problem -- people no longer need half-a-year to catch up. They can get the episodes they missed by downloading them. If the industry wants to compete, then do it with convenience.
Make a central location (i.e.- getSouthPark.com) that people can go to and d/l an episode for $5 or so. High-speed servers that make your P2P look like shit. Don't have to hunt, don't have to worry you're getting inferior VHS to DivX after-the-dog-chewed-the-tape copies, don't have to wait. Hell, $5 for ones with the commercials or $7.50 sans ads.
Lots of people would jump on that. Add a subscription service for a show -- get all the episodes sent directly to your TiVo for $50 a season. Sort of like "League Pass" with the sports.
The problem is the model is changing and the industry execs don't want to change with it. They are comfortable.
Just to point out, as the article details, a DIGITAL bootleg of Spiderman was out on the Net the day before it hit the theater. The result? The theatrical release STILL was the largest grossing opening day (and weekend) ever. Its second weekend was the largest second weekend for a movie ever. Its third weekend (this weekend) is sitting at $46 million which is, surprise, the largest third weekend ever.
Star Wars II: Attack of the Clones.
Ditto. Movie out in digital piracy a week before opening, and it still makes obscene amounts of money ($86 million this weekend and $110+ million so far).
Wanna check on the sales of Star Wars I: Phantom Menace when released on VHS/DVD? One of the best sellers; ditto for The Matrix -- both of which were floating the web in DivX format before they hit the theaters, much less DVD/VHS.
The last 4 years (1998-2001) are the best on record for revenue generated and attendence at theaters. DVD/VHS sales are thru the roof.
In the "perfect" world, where movies are uncopiable and you have to see it at the theater and/or purchase a legitimate copy, the industry would see only a paltry rise in revenue compared to today -- not the $3 Billion touted by Mr. Valenti.
Most people who get rips would either do without altogether, or wait until the DVD/VHS that THEY WERE GOING TO PURCHASE ANYWAY became available.
Hmmm. It was my understanding that the 0.5W figure for the ULV P3M was in "Deep Sleep" mode. I was also assuming that when running a task, the CPU would be a full-tilt for any of the types of applications a "supercomputer" would be needed for. I see where Intel is reporting the AVERAGE power of the unit running TYPICAL OFFICE APPLICATIONS. The problem with these measurements is the CPU is 99% idle when people are typing in word -- it doesn't matter if the CPU is running at 700 MHz or 7 MHz, you aren't going to out-type it.
The ULV P3M runs a 100 MHz bus, like the 633 Crusoe but the 677 Crusoe runs a 133 MHz bus like some of the LV P3Ms.
The final problem with the P3M is the thermal diode. To control heat, once the core CPU temp reaches a certain number (100 deg F, I think -- the "maximum junction temperature"), it clocks down to reduce heat. Again, that's fine for someone typing in Word or Excel. It can clock up for the 3 seconds needed to run that macro, but for sustained high-performance computing, it will be a problem.
I'll agree that Intel is very competitive in the laptop CPU market and their LV and ULV, SpeedStep enabled chips are great in that market -- hell, I'm typing this on an IBM laptop with a SpeedStep enabled 1 GHz P3M, and it blows the doors off the Dell P3-450 I just got rid of.
However, for sustained computing where you aren't relying on user input to clock-down between, I think the fewer transistors on the Crusoe generate a hell of a lot less heat and use lots less electricity. Transmeta has some nice thermal photos on their website, but I believe they are comparing with the "old", non-SpeedStep P3M and not any of the LV/ULV stuff.
Using this site as an example to estimate power usage, we get:
240 computer blades in Green Destiny x 6,480 hours uptime (9 months) = 1,555,200 computer hours of uptime
Assuming the only thing changed on the blade is the CPU -- and North Bridge chipset, since the Crusoe includes
a North Bridge on die and the P-III does not -- at full blast the Crusoe consumes about 1.75W of power and the
P-III + NB consumes between 4.5 - 8 W, depending on chip model. However, the 4.5W number is an approximation
from the 0.13 micron ULV P-IIIM chip running in "Battery Saving" mode, or SpeedStepped down to 300 MHz. Running
at full 700 MHz tilt, with NB, we are still talking 5.75W of power consumed.
1,555,200 * 0.0175Kw * 0.10 (dollar per KwH power cost) = $2,721.60 electricity cost/year (Crusoe)
1,555,200 * 0.0575Kw * 0.10 (dollar per KwH power cost) = $8,942.40 electricity cost/year (Intel)
A saving of approx. $6,200/year in direct electric costs.
However, the big savings comes from the heat dissipation of the units. While the newer LV/ULV P-IIIs do not require
active cooling, they still run quite a bit warmer than the Crusoe units. As a result, you don't stick a rack
full of them in a room that isn't temperature controlled. The difference in the air conditioning bill can
easily reach tens of thousands of dollars.
In business, there are two types of money/budgets. One-time grants and acquisition budgets are large chunks of
cash. Recurring expense and operations budgets are smaller. Being able to get a large chunk of cash to BUY a
cluster/supercomputer is one thing. Being able to go back year-after-year and get the funds to keep it running
is another project altogether. $15,000 - $20,000/year for electricity used in running/cooling computers is a
LOT of money to some people. This doesn't include construction or maintenance costs on a custom facility/room.
As far as reduced administration costs go, many conventional supercomputers required chilled water and other
special considerations for operation. People with experience managing things like Sun E15000s and Cray T3Es
are few and far between. They are the last of the "high priesthood" of computer administrators and cost a LOT
of money to employ.
A blade server, on the other hand, is a bunch of x86 computers running Linux -- nothing a couple of grad students
can't learn the ins-and-outs of over a term. Maintenance contracts, spare parts, etc. are also TONS cheaper for
the blade/cluster solution as opposed to high-end SGIs, Suns, Fujitsu and Cray super-computers.
Another site with a bit of good supporting information is
PC Stats.
The difference?
A database program, for one. More licensed clip art and fonts, for another.
The entire article can be summed up as "MPLS is coming. Soon everything will be IP-based."
Cisco is very big in Layer-3 switching and MPLS, but their Layer-2 switches (ATM and such) are trash compared to Nortel and Lucent. They are pushing big to move large telcos to MPLS and replace their ATM and Frame Relay networks.