"First, even though Sun's E15k has 106 processors, only 72 of them are even directly connected to the memory fabric. The others are just PCI cards, with staggering latency, consequently they won't help transaction processing performance."
Uh, no. The extra CPUs connect to the centerplane. They are not on PCI cards. You are smoking crack. There is a whitepaper on Sun.com about this. Try reading.
"Third, although the p690 has only 32 processors, it has 64 cores. Each "processor" has two 4-way CPUs with a very low latency interconnect. The IBM product would be more accurately characterized as a 64-processor machine."
Wrong again. The p690 has FOUR MCMs. Each MCM has FOUR chips. Each chip has TWO CPU cores. 4x4x2=32. p690 has 32 CPUs. Not 64. The HPC model of the p690 turns off one of the cores on each chip. The HPC model of the p690 has 16 CPUs. Not 32.
"HP's new 16 processor box gets almost the same tpc rating as Sun's 64 processor E10k. IBM's old p680 with 24 processors almost doubled the performance of the E10k w/ 64 processors."
If HP and IBM are so much faster than Sun on TPC-C, why are they not equally faster on every other benchmark, like TPC-H, Oracle Apps, PeopleSoft? HP has yet to run a Oracle Applications, or SAP or other benchmark on their new server. Just try to get a PeopeSoft benchmark on an HP box. They haven't published one in almost two years! What is HP afraid of?
The truth is, if the TPC had the guts to replace their lame TPC-C benchmark with a 10,000 table database with 100% non-local lookups, it might be worth something. But right now it is useless.
"IBM's AIX partitioning code (originally introduced in the SP frames) predates Sun's earliest attempts by at least two years."
Wrong. The IBM SP2 and Cray CS6400 came out in about the same timeframe. Besides, I would not call the nodes of a distributed memory MPP system partitions, since individual nodes are not sharing memory. Now something like the SGI Origin, where nodes can be shared memory or distributed might count.
"First of all, let me mention that the RS/6k S80 (two releases ago - prior to the p680) outdid the 10k, at reduced cost, with the previous generation of procs."
This is bullcrap. IBM was able to get a high TPC-C score. TPC-C is a really lame, 6-year database benchmark designed when the typical RISC CPU ran at 80 MHz, had 256KB cache, 4GB RAM max, less than a 1 GB/sec backplane, and storage was done on 5,400 RPM SCSI-2 drives.
The S80 got a high score because it's CPU had a very short pipeline (5-stage) which a database benchmark with a lot of locality benefits from tremedously. Another benefit was the S80's CPUs had twice the cache as the rest of the industry. Throw in some really good database benchmark engineers and you can really tune for TPC-C's locality. Also, the S80 could hold as memory as the E10K, so a big Oracle SGA means fewer disk accesses.
Then with the p680, IBM boosted performance by turning on an experimental CPU feature called Hardware Multi Threading. Real cool, but no datacenter manager in his right mind would run HMT on a production machine.
What all this proved was nothing. When you looked at real-world bechmarks such as Oracle Applications, PeopleSoft, SAP, etc., Sun servers with 400 MHz UltraSPARC II CPUs performed as well as, and sometimes better than IBM boxes with 450 MHz CPUs.
The whole pitch IBM was making that the S80's 450 MHz CPUs were equal to three 400 MHz E10K CPUs were absolute fraudulent. Fortunately, in my job, I got the chance to blow IBM's story out of the water. There was no more fun to see an IBM sales rep at a total loss for words in front of our CIO when I called him on the 3 IBM CPUs = 1 Sun CPU. Of course, when the CIO said: "You know, I really don't mind marketing, but I can't stand lying." The IBM guy started sweating.
I guess the reality was we didn't really like IBM anyway, were mostly a Sun and HP shop. But the way IBM muscled their way into the opportunity really pissed us all off, but we decided to give them a shot. But they made EMC sales reps look like kindergarten teachers.
"16 processors outdid 64 in many, many tests (including ones with real-world data movement)."
Total BS. IBM did no S80/p680 benchmarks with "real-world data movement" because the S80 has a big bottleneck called the I/O hub, which is by design limited to 1 GB/sec input max. So IBM did no TPC-H or similar data warehousing benchmark. In fact, the last one IBM did was an old SP2 result.
"As for partitioning... hmmm... let's think. IBM has been doing logical partitioning in AS/400 for a while, and on the S/390 (now the z-series) for quite some time... a few decades now. A lot of that experience went into this."
Decades, huh? The S/390 got LPARS in 1987. That is 14 years ago. That is 1.4 decades, not a "few decades". Maybe it is the IBM math that is the problem. Sort of like how a 32-CPU p690 becomes a 16 CPU p690 when running HPC code.
Most of the customers at this level are more interested in hypereliability than hyperperformance. And most are not going to trust version 1.0 of AIX LPARs (actually AIX VM Guest OS would be a more accurate mainframe feature comparison).
That said I don't doubt you got a really powerful CPU is a tight litte package, but don't overhype it.
You know, I really don't mind marketing, but I can't stand lying.
"First of all, yes it has 32 CPUs... but each CPU has two cores on it, i.e... it's more comparable to a 64 CPU box."
Wrong, it is 4 MCMs each consisting of 4 dual core chips, for a total fo 4x4x2=32 CPUs.
Yes the POWER4 CPUs have high benchmarks, but for HPC work you have to turn half of them off, turning each dual-core into a single core.
No, the IBM box does not run "Domains". It runs software based partitions. Think multiple VMWare sessions, not super secure, fault-isolated mainframe partitions, and not the total electrical isolation Sun's domains provide.
"Also, as IBM invented partition (xx/360-390 and AS/400)"
Uh, no. Amdahl invented mainframe partitioning in 1985 with the Multi-Domain Feature (MDF). It took IBM two years to clone this feature.
And don't think Sun engineers were not involved in the CS6400 project. The CS6400 was not a Cray product, it came from Floating Point Systems (FPS), who was working with Sun to build a large SMP SPARC system. After all, the code name for the CS6400 was "SuperDragon". The code name for the Sun SPARCCenter 2000 was "Dragon". The CS6400 used four of SC2000 busses to create its interconnect.
Her name is Charlotte Blackwood. I knew here back when she worked at the Navy Fighter Weapons School when it was at Mirimar. When Top Gun moved to that $hithole NAS Fallon, Charlie took the east coast gig with asymmetric warfare.
Did you notice the point-to-point switched architecture that sits on top of the PCI-X bridges?
Wanna be that ends up being Infiniband? It will. You will be able to cable PCI-X expansion racks through IB switiches into the server.
And don't think IB will replace Fibre-Channel. FC will still be the route from controller to disk, but expect a switch layer that translates from IB to FC. No more HBAs.
Also, once 10Gbit Ethernet comes along, do you really think you are going to have servers with a a dedicated four-processor set of 2.5 GHz CPUs to handle the packets? No, you will have a 10Gb network processor with a 10000BaseFX port on one side and an IB port on the other.
The person that wants to add more monitors can plug additional graphics processors into IB ports.
Anyone familiar with SGI's XIO architecture can see what IB has to offer. IB IS XIO, but instead of card slots and pins, it uses optical cables and ports.
All of this means it will be much easier to deal with different speeds and protocols of networks, be they Ethernet, Fibre-Channel, or Infiniband.
Infiniband is similar to Sun's UPA and SGI's XIO. Direct, switched connection to memory. Makes tremendous sense for a graphics connection, unless it has a weakness I do not know about.
You have obviously been reading IBM Marketing LIEterature. IBM may have sold more S80s than E10Ks, but Sun sold far more E10K's than IBM sold S390s (a more valid comparison). And Sun sold far more E6500's than IBM sold S80s (a more valid comparison). I mean come on, an RS6000 can be anything from a uniprocessor workstation up to a 24-way S80. It's like saying AMD sold more Athlon-based systems than Compaq sold Tandems. What is the point?
"a.root, anyone?"
Uh, yeah, more IBM LIES. The S80 replaced a Sun E450, not an E10K.
"they made their machines better"
To make the S80, IBM crammed twice as many processors into an S70 and did not increase the system bandwidth to compensate. You call that better?
"Now Sun will work harder on their lineup to try and outsell IBM"
When it comes to S390s, RS6000s, and AS400s, Sun outsells IBM every day.
SLC-6 was originally built for the MOL program. I thought it had been used for Titan III launches, but I could be wrong. I know the last plan was to modify it to accept the Titan IV launchers.
S-80's will include hot-swappable memory and CPU's...so that won't be an issue for much longer.
Sun has been swapping processor boards on running systems for quite some time now. Designing a computer that can do this is _not_ an easy thing to do. The E10K is a second generation machine (the Cray CS6400 was the first generation). Dynamic Reconfiguration ("DR") requires all device drivers to be tested and stable during this operation. Memory has to be "drained" from the banks on the board being removed and processes have to be migrated off of the processors on the board being removed. The hardware on the system board has to support DR, the backplane has to support DR, the control board has to support DR, the operating system has to support DR, and the Service Processor has to control DR.
Again, _not_ easy. I do expect that IBM, given its extensive experience making mainframes, could definately provide this capability to a UNIX system if they put their minds to it. Heck, they put LPARS on the AS400.
BTW, the E10K has three times the system memory bandwidth of the S80. That is why IBM will never publish a Stream benchmark for the S80.
This all great that they are making a better big engines for trucks, but can somebody make a more reliable 2-cycle engine for chainsaws, weedeaters, and the like?
The end of the engineering shortage was caused by a couple of things. The problems in the Boston area were caused in part by companies moving to the sunbelt. There was certainly more of a shortage in Texas and Florida than Massachusetts and New Hampshire.
The other factor was the reduction in defense spending. Defense spending drove much of the technology research in the Boston area. After Vietnam, and continuing through the Carter administration, spending declined. The Reagan "buildup" was short lived, for the most part purchased weapons that had already been designed (i.e., B-1 bomber), and the research spending was more focused on SDI. The fall of the Soviet Union ended most of the remaining research.
As an Aerospace Engineering undergraduate student in the 1980's, I saw much of this first hand. Many students entered the Aerospace curriculum in '86/'87, and then the bottom fell out, leaving many who graduated from '88-'90 unable to find jobs, or losing the jobs they had.
Fortunately, commercial business in technology research started taking off in the early '80s, and has continued. We are in period of fundamental change, from the industrial age to the information age, and we are probably only about 25% of the way through the transformation. The tech worker shortage will not end any time soon. There is simply too much to be done.
Finally, while it is easy to be critical of immigration as a solution to the tech shortage, one only has to look at where these immigrants are coming from: Countries with high education standards, and a local economy that cannot provided the necessary jobs. This means India and the Orient. These are not sweat-shop workers, in many cases these people have advanced degrees, and are very good at what they do. They also have much better work ethics and stronger loyalty than the typical American worker. The real issue in the future will be whether they immigrate to America to pursue employment, or if major companies outsource entire development departments to front companies in India and Taiwan--which will happen if restrictions against immigration are put in place.
The best thing a young person can do is get a college degree, and earn a reputation as a serious person, one that your references will attest to. And realize these two fundamental facts of life:
No one owes you a job. You have to earn it.
You have to pay your dues in this world. Your boss' feel that they paid their dues, and expect the same of you. This may mean working as a contract worker for a year or two to gain a level of experience to apply for a higher position.
My company has over 1900 open positions. My boss spent six months trying to fill a position recently, inteviewing about 60 different people. But the only thing a boss likes less than filling a position is firing somebody. That means the boss has to have the utmost confidence that a person is the right person for the job.
My joystick disappeared when I added SP5. I reinstall it but it disappears after reboot. Plus, the Flight Sim 98 install did a rip and replace on my Sounblaster Drivers and removed the volume control from the tray. I now have to fly in windows 95.
Sun is rumored to have purchased StarDivision? That would require an SEC filing. Publically traded companies cannot buy other companies in secret. Go back to your room, turn your a.m. radio to Art Bell, and masturbate to your alien abduction/sexual probing fantasies.
... however, your best comparison, i.e., the telcos, is different. Long-distance telephone service is an open-standards commodity, as is local service provider dial tone. What I mean by this is any company can make a phone (application) that will work with any local service provider and make long distance phone calls regardless of long distance provider. I still have a phone I bought back in 1987 -- it's been through seven locations, two countries, and four states. It worked in all of them -- Talk about application portability.
The fact that it is (damn near) impossible for ACME software to produce a 100% binary compatible Windows clone is why competion is so difficult on the x86 PC platform.
That is why I think that MS-Windows 98 and NT Workstation licensing fees should be set by a PSC. There is no other provider of 100% Win32 compatible operating systems.
Nor could there be. Unless MS GPLs their code. Now there's a solution.
Just like municipalities have Public Service Commisions that regulate the rate the utility monopolies can charge, there should be a commision that oversees the "Microsoft Tax" to ensure OEMs and customers are not screwed.
Breaking up Microsoft will just create several monopoly companies.
Anyone who is a CS-type that thinks FORTRAN is dead just needs to talk to a mechanical or aerospace engineer. FORTRAN is still the primary language of engineering. Anything that is real-number math intensive is done in FORTRAN: CFD, simulations, etc. Granted, the FORTRAN code is often linked to other written in C or C++. The one person who said he was told in an interview "nobody progams in FORTAN anymore" was obviously not interviewing for an engineering position at Boeing, Ford, or Sverdrup.
Isn't Anakin/Vader about 6'5"? Isn't Leo about 5'6"? How does one explain the difference? I guess in Episode 3 Anakin/Vader will be strapped to a medieval rack and stretched.
I have both Netscape 4.51 and IE 5.0 on my machine, an AMD K6-2 266 running NT 4.0 SP3. I find IE to be slower at accessing sites and rendering HTML. Maybe I spend most of my time on sites that are optimtzed for Netscape, I do not know. I agree about the stability, sometimes, after a running for a while, Netscape decides it wants 98% of the CPU. I still can't figure that one out.
That would be the "Magic Bus" ...
on
SGI Name Change
·
· Score: 1
... an 18-wheeler chock full of SGI, ehm, sgi, technology. A trade show on wheels, it has (at least when I worked it) a small O2K, deskside Onyx2, about a half a dozen workstations (Octanes and O2s) and in the very back, a kick-ass three-pipe Onyx2 Infinite Reality dual-rack.
Algore is a dork, not a nerd. The most technical savvy politition of this decade was Newt Gingrich. I remember watching his televised college class back in the late '80s, and thinking "This guys a politician?" Unfortunately, Newt has now been reduced to making $50,000/night on the chicken and mashed potatos circuit, and as such, he will only tell those groups what they want to hear, not what he thinks is interesting or important.
> Ronald Reagan and his extremely big > government policies ("Star Wars" > defense system, drug war, Iran/Contra, > etc.).
Yes the defense buildup had much to do with the deficits. However, Reagan originally intended to eliminate or reduce many social programs and corportate tax breaks (i.e., Windfall Profits on oil). These reductions were impossible to get through a congress that was more concerned with bring home the pork. The benefit of the military buildup was that it directly led to the econmic collapse of the Soviet Union, which also caused the politcal collapse of totalitarian Soviet Communism. Hence, we have been able to reduce our defense spending and infrastructure by over 40% since its high of 1985-86.
> has actually reduced the size of > government from its high under > Ronald Reagan.
To use a Clintonian term: It depends on what "reduced" means.
Non-defense spending, and non-defense government employment has risen under Clinton. All of the total net reductions have come from the Department of Defense, to include both the uniformed military, and the civil servants. Since the reduction of the DOD was result of policies of previous administrations, it is incorrect to give Clinton credit for this.
Slashdot Mirror
"First, even though Sun's E15k has 106 processors, only 72 of them are even directly connected to the memory fabric. The others are just PCI cards, with staggering latency, consequently they won't help transaction processing performance."
Uh, no. The extra CPUs connect to the centerplane. They are not on PCI cards. You are smoking crack. There is a whitepaper on Sun.com about this. Try reading.
"Third, although the p690 has only 32 processors, it has 64 cores. Each "processor" has two 4-way CPUs with a very low latency interconnect. The IBM product would be more accurately characterized as a 64-processor machine."
Wrong again. The p690 has FOUR MCMs. Each MCM has FOUR chips. Each chip has TWO CPU cores. 4x4x2=32. p690 has 32 CPUs. Not 64. The HPC model of the p690 turns off one of the cores on each chip. The HPC model of the p690 has 16 CPUs. Not 32.
"HP's new 16 processor box gets almost the same tpc rating as Sun's 64 processor E10k. IBM's old p680 with 24 processors almost doubled the performance of the E10k w/ 64 processors."
If HP and IBM are so much faster than Sun on TPC-C, why are they not equally faster on every other benchmark, like TPC-H, Oracle Apps, PeopleSoft? HP has yet to run a Oracle Applications, or SAP or other benchmark on their new server. Just try to get a PeopeSoft benchmark on an HP box. They haven't published one in almost two years! What is HP afraid of?
The truth is, if the TPC had the guts to replace their lame TPC-C benchmark with a 10,000 table database with 100% non-local lookups, it might be worth something. But right now it is useless.
"IBM's AIX partitioning code (originally introduced in the SP frames) predates Sun's earliest attempts by at least two years."
Wrong. The IBM SP2 and Cray CS6400 came out in about the same timeframe. Besides, I would not call the nodes of a distributed memory MPP system partitions, since individual nodes are not sharing memory. Now something like the SGI Origin, where nodes can be shared memory or distributed might count.
"First of all, let me mention that the RS/6k S80 (two releases ago - prior to the p680) outdid the 10k, at reduced cost, with the previous generation of procs."
This is bullcrap. IBM was able to get a high TPC-C score. TPC-C is a really lame, 6-year database benchmark designed when the typical RISC CPU ran at 80 MHz, had 256KB cache, 4GB RAM max, less than a 1 GB/sec backplane, and storage was done on 5,400 RPM SCSI-2 drives.
The S80 got a high score because it's CPU had a very short pipeline (5-stage) which a database benchmark with a lot of locality benefits from tremedously. Another benefit was the S80's CPUs had twice the cache as the rest of the industry. Throw in some really good database benchmark engineers and you can really tune for TPC-C's locality. Also, the S80 could hold as memory as the E10K, so a big Oracle SGA means fewer disk accesses.
Then with the p680, IBM boosted performance by turning on an experimental CPU feature called Hardware Multi Threading. Real cool, but no datacenter manager in his right mind would run HMT on a production machine.
What all this proved was nothing. When you looked at real-world bechmarks such as Oracle Applications, PeopleSoft, SAP, etc., Sun servers with 400 MHz UltraSPARC II CPUs performed as well as, and sometimes better than IBM boxes with 450 MHz CPUs.
The whole pitch IBM was making that the S80's 450 MHz CPUs were equal to three 400 MHz E10K CPUs were absolute fraudulent. Fortunately, in my job, I got the chance to blow IBM's story out of the water. There was no more fun to see an IBM sales rep at a total loss for words in front of our CIO when I called him on the 3 IBM CPUs = 1 Sun CPU. Of course, when the CIO said: "You know, I really don't mind marketing, but I can't stand lying." The IBM guy started sweating.
I guess the reality was we didn't really like IBM anyway, were mostly a Sun and HP shop. But the way IBM muscled their way into the opportunity really pissed us all off, but we decided to give them a shot. But they made EMC sales reps look like kindergarten teachers.
"16 processors outdid 64 in many, many tests (including ones with real-world data movement)."
Total BS. IBM did no S80/p680 benchmarks with "real-world data movement" because the S80 has a big bottleneck called the I/O hub, which is by design limited to 1 GB/sec input max. So IBM did no TPC-H or similar data warehousing benchmark. In fact, the last one IBM did was an old SP2 result.
"As for partitioning... hmmm... let's think. IBM has been doing logical partitioning in AS/400 for a while, and on the S/390 (now the z-series) for quite some time... a few decades now. A lot of that experience went into this."
Decades, huh? The S/390 got LPARS in 1987. That is 14 years ago. That is 1.4 decades, not a "few decades". Maybe it is the IBM math that is the problem. Sort of like how a 32-CPU p690 becomes a 16 CPU p690 when running HPC code.
Most of the customers at this level are more interested in hypereliability than hyperperformance. And most are not going to trust version 1.0 of AIX LPARs (actually AIX VM Guest OS would be a more accurate mainframe feature comparison).
That said I don't doubt you got a really powerful CPU is a tight litte package, but don't overhype it.
You know, I really don't mind marketing, but I can't stand lying.
"First of all, yes it has 32 CPUs... but each CPU has two cores on it, i.e... it's more comparable to a 64 CPU box."
Wrong, it is 4 MCMs each consisting of 4 dual core chips, for a total fo 4x4x2=32 CPUs.
Yes the POWER4 CPUs have high benchmarks, but for HPC work you have to turn half of them off, turning each dual-core into a single core.
No, the IBM box does not run "Domains". It runs software based partitions. Think multiple VMWare sessions, not super secure, fault-isolated mainframe partitions, and not the total electrical isolation Sun's domains provide.
"Also, as IBM invented partition (xx/360-390 and AS/400)"
Uh, no. Amdahl invented mainframe partitioning in 1985 with the Multi-Domain Feature (MDF). It took IBM two years to clone this feature.
And don't think Sun engineers were not involved in the CS6400 project. The CS6400 was not a Cray product, it came from Floating Point Systems (FPS), who was working with Sun to build a large SMP SPARC system. After all, the code name for the CS6400 was "SuperDragon". The code name for the Sun SPARCCenter 2000 was "Dragon". The CS6400 used four of SC2000 busses to create its interconnect.
Her name is Charlotte Blackwood. I knew here back when she worked at the Navy Fighter Weapons School when it was at Mirimar. When Top Gun moved to that $hithole NAS Fallon, Charlie took the east coast gig with asymmetric warfare.
Let's see:
...
...
...
...
...
...
...
Scored higher on his SAT than former Senator Bill Bradley, and by some reports Al Gore as well
Graduated Yale with a Bachlor's degree and a higher GPA than Al Gore's Harvard undergraduate GPA
Graduated from the U.S. Air Force's Undergraduate Pilot Training (UPT) and flew supersonic jet fighters
Graduated Harvard Business with an MBA (that's a Masters of Business Adminstration for you folks in Palm Beach County Florida)
Yeah, a real idiot. Let's compare to Al Gore:
Graduated from Harvard with a lower GPA than Bush's Yale GPA
Flunked out of Divinity School
Flunked out of Law School
And Al Gore is considered a Genious and Bush the idiot? Whatever!
Interesting article, but ...
Did you notice the point-to-point switched architecture that sits on top of the PCI-X bridges?
Wanna be that ends up being Infiniband? It will. You will be able to cable PCI-X expansion racks through IB switiches into the server.
And don't think IB will replace Fibre-Channel. FC will still be the route from controller to disk, but expect a switch layer that translates from IB to FC. No more HBAs.
Also, once 10Gbit Ethernet comes along, do you really think you are going to have servers with a a dedicated four-processor set of 2.5 GHz CPUs to handle the packets? No, you will have a 10Gb network processor with a 10000BaseFX port on one side and an IB port on the other.
The person that wants to add more monitors can plug additional graphics processors into IB ports.
Anyone familiar with SGI's XIO architecture can see what IB has to offer. IB IS XIO, but instead of card slots and pins, it uses optical cables and ports.
All of this means it will be much easier to deal with different speeds and protocols of networks, be they Ethernet, Fibre-Channel, or Infiniband.
Infiniband is similar to Sun's UPA and SGI's XIO. Direct, switched connection to memory. Makes tremendous sense for a graphics connection, unless it has a weakness I do not know about.
"IBM outsold Sun's E10K with their RS/6000"
You have obviously been reading IBM Marketing LIEterature. IBM may have sold more S80s than E10Ks, but Sun sold far more E10K's than IBM sold S390s (a more valid comparison). And Sun sold far more E6500's than IBM sold S80s (a more valid comparison). I mean come on, an RS6000 can be anything from a uniprocessor workstation up to a 24-way S80. It's like saying AMD sold more Athlon-based systems than Compaq sold Tandems. What is the point?
"a.root, anyone?"
Uh, yeah, more IBM LIES. The S80 replaced a Sun E450, not an E10K.
"they made their machines better"
To make the S80, IBM crammed twice as many processors into an S70 and did not increase the system bandwidth to compensate. You call that better?
"Now Sun will work harder on their lineup to try and outsell IBM"
When it comes to S390s, RS6000s, and AS400s, Sun outsells IBM every day.
SLC-6 was originally built for the MOL program. I thought it had been used for Titan III launches, but I could be wrong. I know the last plan was to modify it to accept the Titan IV launchers.
Sun has been swapping processor boards on running systems for quite some time now. Designing a computer that can do this is _not_ an easy thing to do. The E10K is a second generation machine (the Cray CS6400 was the first generation). Dynamic Reconfiguration ("DR") requires all device drivers to be tested and stable during this operation. Memory has to be "drained" from the banks on the board being removed and processes have to be migrated off of the processors on the board being removed. The hardware on the system board has to support DR, the backplane has to support DR, the control board has to support DR, the operating system has to support DR, and the Service Processor has to control DR.
Again, _not_ easy. I do expect that IBM, given its extensive experience making mainframes, could definately provide this capability to a UNIX system if they put their minds to it. Heck, they put LPARS on the AS400.
BTW, the E10K has three times the system memory bandwidth of the S80. That is why IBM will never publish a Stream benchmark for the S80.
This all great that they are making a better big engines for trucks, but can somebody make a more reliable 2-cycle engine for chainsaws, weedeaters, and the like?
The end of the engineering shortage was caused by a couple of things. The problems in the Boston area were caused in part by companies moving to the sunbelt. There was certainly more of a shortage in Texas and Florida than Massachusetts and New Hampshire.
The other factor was the reduction in defense spending. Defense spending drove much of the technology research in the Boston area. After Vietnam, and continuing through the Carter administration, spending declined. The Reagan "buildup" was short lived, for the most part purchased weapons that had already been designed (i.e., B-1 bomber), and the research spending was more focused on SDI. The fall of the Soviet Union ended most of the remaining research.
As an Aerospace Engineering undergraduate student in the 1980's, I saw much of this first hand. Many students entered the Aerospace curriculum in '86/'87, and then the bottom fell out, leaving many who graduated from '88-'90 unable to find jobs, or losing the jobs they had.
Fortunately, commercial business in technology research started taking off in the early '80s, and has continued. We are in period of fundamental change, from the industrial age to the information age, and we are probably only about 25% of the way through the transformation. The tech worker shortage will not end any time soon. There is simply too much to be done.
Finally, while it is easy to be critical of immigration as a solution to the tech shortage, one only has to look at where these immigrants are coming from: Countries with high education standards, and a local economy that cannot provided the necessary jobs. This means India and the Orient. These are not sweat-shop workers, in many cases these people have advanced degrees, and are very good at what they do. They also have much better work ethics and stronger loyalty than the typical American worker. The real issue in the future will be whether they immigrate to America to pursue employment, or if major companies outsource entire development departments to front companies in India and Taiwan--which will happen if restrictions against immigration are put in place.
The best thing a young person can do is get a college degree, and earn a reputation as a serious person, one that your references will attest to. And realize these two fundamental facts of life:
No one owes you a job. You have to earn it.
You have to pay your dues in this world. Your boss' feel that they paid their dues, and expect the same of you. This may mean working as a contract worker for a year or two to gain a level of experience to apply for a higher position.
My company has over 1900 open positions. My boss spent six months trying to fill a position recently, inteviewing about 60 different people. But the only thing a boss likes less than filling a position is firing somebody. That means the boss has to have the utmost confidence that a person is the right person for the job.
My joystick disappeared when I added SP5. I reinstall it but it disappears after reboot. Plus, the Flight Sim 98 install did a rip and replace on my Sounblaster Drivers and removed the volume control from the tray. I now have to fly in windows 95.
Sun is rumored to have purchased StarDivision? That would require an SEC filing. Publically traded companies cannot buy other companies in secret. Go back to your room, turn your a.m. radio to Art Bell, and masturbate to your alien abduction/sexual probing fantasies.
... however, your best comparison, i.e., the telcos, is different. Long-distance telephone service is an open-standards commodity, as is local service provider dial tone. What I mean by this is any company can make a phone (application) that will work with any local service provider and make long distance phone calls regardless of long distance provider. I still have a phone I bought back in 1987 -- it's been through seven locations, two countries, and four states. It worked in all of them -- Talk about application portability.
The fact that it is (damn near) impossible for ACME software to produce a 100% binary compatible Windows clone is why competion is so difficult on the x86 PC platform.
That is why I think that MS-Windows 98 and NT Workstation licensing fees should be set by a PSC. There is no other provider of 100% Win32 compatible operating systems.
Nor could there be. Unless MS GPLs their code. Now there's a solution.
Just like municipalities have Public Service Commisions that regulate the rate the utility monopolies can charge, there should be a commision that oversees the "Microsoft Tax" to ensure OEMs and customers are not screwed.
Breaking up Microsoft will just create several monopoly companies.
Anyone who is a CS-type that thinks FORTRAN is dead just needs to talk to a mechanical or aerospace engineer. FORTRAN is still the primary language of engineering. Anything that is real-number math intensive is done in FORTRAN: CFD, simulations, etc. Granted, the FORTRAN code is often linked to other written in C or C++. The one person who said he was told in an interview "nobody progams in FORTAN anymore" was obviously not interviewing for an engineering position at Boeing, Ford, or Sverdrup.
Isn't Anakin/Vader about 6'5"? Isn't Leo about 5'6"? How does one explain the difference? I guess in Episode 3 Anakin/Vader will be strapped to a medieval rack and stretched.
Silicon Graphics, er, SGI was founded in 1982.
I have both Netscape 4.51 and IE 5.0 on my machine, an AMD K6-2 266 running NT 4.0 SP3. I find IE to be slower at accessing sites and rendering HTML. Maybe I spend most of my time on sites that are optimtzed for Netscape, I do not know. I agree about the stability, sometimes, after a running for a while, Netscape decides it wants 98% of the CPU. I still can't figure that one out.
... an 18-wheeler chock full of SGI, ehm, sgi, technology. A trade show on wheels, it has (at least when I worked it) a small O2K, deskside Onyx2, about a half a dozen workstations (Octanes and O2s) and in the very back, a kick-ass three-pipe Onyx2 Infinite Reality dual-rack.
Algore is a dork, not a nerd.
The most technical savvy politition of this decade was Newt Gingrich. I remember watching his televised college class back in the late '80s, and thinking "This guys a politician?" Unfortunately, Newt has now been reduced to making $50,000/night on the chicken and mashed potatos circuit, and as such, he will only tell those groups what they want to hear, not what he thinks is interesting or important.
> Ronald Reagan and his extremely big
> government policies ("Star Wars"
> defense system, drug war, Iran/Contra,
> etc.).
Yes the defense buildup had much to do with the deficits. However, Reagan originally intended to eliminate or reduce many social programs and corportate tax breaks (i.e., Windfall Profits on oil). These reductions were impossible to get through a congress that was more concerned with bring home the pork. The benefit of the military buildup was that it directly led to the econmic collapse of the Soviet Union, which also caused the politcal collapse of totalitarian Soviet Communism. Hence, we have been able to reduce our defense spending and infrastructure by over 40% since its high of 1985-86.
> has actually reduced the size of
> government from its high under
> Ronald Reagan.
To use a Clintonian term: It depends on what "reduced" means.
Non-defense spending, and non-defense government employment has risen under Clinton. All of the total net reductions have come from the Department of Defense, to include both the uniformed military, and the civil servants. Since the reduction of the DOD was result of policies of previous administrations, it is incorrect to give Clinton credit for this.
This is simply the start of a planned FUD campaign against Open Source by Microsoft. First, they obsfucate the issue and attempt to redefine it.
...
Stay tuned