An SLA really is nothing more than a contracted warrenty that says what is covered and what is not covered.
Um. No, it isn't. An SLA is a contract that commits the vendor to deliver certain things. It's not a warranty. A warranty says, "If this device fails for any of the following reasons, we (the vendor) will accept complete or partial responsibility for fixing it." An SLA says, "This device will not fail, within these given parameters." SLAs include such things as vendor penalty clauses in the event that the level of service promised isn't delivered.
An SLA and a warranty are very different things. I think it's important that we clarify this if we're to discuss this issue at all.
Now, as I understand it, Microsoft works with certain partners to offer complete managed solutions for their Datacenter product line. I'd imagine that a managed solution-- in which the hardware, software, service, and maintenance are all provided by a single vendor-- would come with an SLA. But anything less than that, no. As far as I know, there's no way Microsoft would ever offer an SLA on their software by itself.
Now, if I'm right about all those things, then I'd go on to imagine that there are-- or, at least, could be-- companies out there who will offer you a similar fully managed solution based on Red Hat Linux, complete with SLA.
I'm no Microsoft apologist, and neither am I a particular fan of Linux. I'm just trying to make sure we're comparing apples to apples here.
Ok. The word 'cluster' is also singular. Big deal.
The word "cluster" is collective. Like "dozen." It refers to a group of things as a unit. It's not possible to have a cluster of one.
Clusters of computers are not supercomputers. They're different. You can try all you like to say that the definition of "supercomputer" is arbitrary; I suppose it is, in the sense that all definitions are fundamentally arbitrary. A supercomputer is a single computer. A cluster of computers is not a single computer. Ergo, a cluster is not the same thing as a supercomputer.
If I could convince all the world's population to do maths with an abacus... that would be a supercomputer too.
No, it wouldn't. A computer-- and, by extension, a supercomputer-- is a mechanical device. A group of people could perform the same job as a supercomputer, but the group wouldn't be a supercomputer.
I think you're possibly confused by metaphors. When you say, "A group of people is a supercomputer," you're employing a metaphor to say, "A group of people shares many important characteristics with a supercomputer." Don't be fooled by this. It's not literally true, and shouldn't be accepted as such.
All things considered, pure probability would tell us that it's incredibly unlikely that you, r0t, are actually capable of comprehending, much less doing anything with, the source code to an operating system. Whether you personally have those skills or not isn't my point; I'm saying that the argument that Linux is superior to Windows because it comes with source code is specious at best. To the vast majority of users and potential users, the source code to the operating system would do nothing more than occupy CDROM-- and, if installed, hard drive-- space.
First, you're wrong. The sampling method, by which one collects a number of data points and uses them to extrapolate trends, is perfectly valid. In fact, it's really the only way to do predictive analysis.
Second, there's a better example of flawed logic; it's called a misaligned syllogism. "All men are mortal. Socrates is mortal. Therefore all men are Socrates."
Massive license fees? How do you define "massive?" A Windows server license will cost you a few thousand bucks, depending on configuration. That's a one-time charge. The guy who maintains it for you will cost you tens of thousands of dollars per year. The cost of the software license is very small in proportion to the recurring costs of owning the system. This is why it's not at all obvious that Linux-- which costs nothing to license-- should be cheaper to own that Windows.
I guess some people think you're just being an asshole; the moderations seem to imply this, anyway. But I'm curious. What kind of SLA is Microsoft willing to offer you? I've never investigated getting one, myself, and frankly the thought has never even occurred to me.
The important thing to notice about the word "supercomputer" is that it's singular. A supercomputer is a single system image; this is implicit in the definition. This is not to say that supercomputing clusters aren't worthy; it's just that they're different in important ways from single-system-image supercomputers.
Some classes of problems aren't suited for cluster computation. I won't pretend to be educated enough to tell you exactly which problems can and can't be adapted for cluster computation, but consider the nature of clusters to see my point. Clusters are highly scalable, but the inter-node latency is huge. An interconnect like Myrinet can get your remote messaging latencies down to the microsecond range, but the far more common MPI/PVM-over-Ethernet solution is a thousand times slower than that. This makes it somewhat inefficient for node N to try to access a bank of memory on node M. In order for a cluster to be efficient, each node should have sufficient physical memory to hold it's entire data set, and each node should be able to operate more-or-less autonomously, without having to contact other nodes.
Supercomputers are fundamentally different from clusters. In some cases, you can do the same job with either a supercomputer or a cluster. Some jobs are better suited to clusters, while some are better suited to supercomputers. Some jobs, as I mentioned above, are better suited to arrays than to either clusters or supercomputers. It just depends on the job.
Wrong. Render farms are neither clusters nor supercomputers. At best, a render farm might be considered an array.
A supercomputer is a single system image. Some people call large clusters "supercomputers," but technically they're wrong.
A cluster is an interconnected group of computers that can communicate with each other. Usually a cluster depends on some kind of software layer to allow programs to run across multiple systems, something like MPI. Clusters are tightly interconnected many-to-many systems.
An array has a single job control system and a number of job execution systems. Batch jobs are submitted by users to the job control system, which doles them out to the various execution systems and then collects the results. The execution nodes don't talk to each other, and one job runs on one execution node at a time. Render farms are basically arrays; each execution node works on rendering a single frame of a multiframe animation. Because each frame can be rendered independently, without any dependencies on the previous and subsequent frames, rendering is particularly well suited to array computing.
With 2048 processors and an assumed 1GB/node that's 1TB of low latency, super high-jiz RAM.
How do you define "low latency?" From a first glance at the evidence, it appears that this cluster just uses plain old TCP/IP over Ethernet as its node interconnect. That's not exactly low latency access to remote memory, you know.
I had a boss one time who wanted all computers in the office named after biblical figures. The big servers were Matthew, Mark, and Luke, but the names got pretty obscure after that. One girl in the art department-- tattoos and piercings, never wore underwear, religious affiliations left as an exercise for the reader-- suggested we put an end to it by naming the boss's laptop Lucifer.
Not wanting to get both fired and excommunicated, I instead talked him out of it by appealing to simple reason. Nobody can spell Zephaniah, I told him.
Roman gods? No. Quaoar was the name of the one and only god in the mythology of one particular native american tribe, the name of which escapes me. For some reason, I remember this fact from freshman year anthropology.
Besides, I hereby announce that I'm taking bets. The official name of this body-- if it ever gets one-- will be Persephone.
Okay, I guess you're right. The normal state is to have two of each of your chromosomes, while haploid cells only have one of each. So while a sperm only has half of your genetic material, it's got a full set.
And I did pay attention in biology; it's just that it was so long ago.;-)
Gives new meaning to that old story about virii(or is it viruses?) being spread when people shake hands...
It's "viruses." And, in case you've been living your new life in the off-world colonies or something ("A chance to begin again!"), viruses are spread by shaking hands!
"The Gernsback Continuum" is his best short story; that is, it's my favorite of his short stories. It's also included in Burning Chrome, I think. Worth a read.
Don't forget that your sperm only hold half of your genes. You're overestimating the number of bytes transmitted a bit there.
It's kind of like the old saying that I'm too lazy to look up right now. "Never underestimate the bandwidth of a station wagon full of tapes tooling down the expressway."
Not to be pedantic, but Aqua is a user interface appearance. What you're talking about is Quartz/AppKit, which is the user interface code that implements the Aqua appearance.
And from a read of the article, it sounds like there are no plans to create a Quartz/AppKit version of OpenOffice. They're talking about removing dependencies on X11 so it'll run on OS X without an X server, but they're not talking about replacing their homegrown and decidedly un-Mac-like UI code with true OS X UI code.
It's a shame. If that's the path they choose, the best they'll be able to do is a poor imitation of a true OS X application.
although for the size of our projects C++ is the obvious lnaguage of choice
Um... no. The language that your people are comfortable using is the language of choice. No other meaningful criteria exists.
Let's say there were some magic language, Foo, that was absolutely perfect for computational chemistry. It has standard library calls like solveThisHardProblemQuickly() and such. Just perfect.
Like any other programming language, Foo has a strict syntax. It's easy to write a buggy program in Foo, but tricky to write a perfect one. All languages are like this.
Your Fortran programmers decide they're going to abandon their preference and write in Foo instead. How many months do you think it would take for them to become conversant in Foo? How many trivial bugs will slip through in that time because your programmers weren't experienced enough with Foo to catch them? How far behind will you be at the end of the whole process?
And that's assuming that Foo is perfect. If Foo has its own idiosyncrasies and quirks, you can comfortable double those estimates.
There is only one inviolable law of computer programming: do what works.
I wrote (well, wrote most of) the engine module for an F-16 flight simulator. All in Fortran 77. Did a bunch of work on the simexec, too, which was in Ada 83.
I'm conflicted. I'm not sure if I'm bragging, or confessing.
Slashdot Mirror
An SLA really is nothing more than a contracted warrenty that says what is covered and what is not covered.
Um. No, it isn't. An SLA is a contract that commits the vendor to deliver certain things. It's not a warranty. A warranty says, "If this device fails for any of the following reasons, we (the vendor) will accept complete or partial responsibility for fixing it." An SLA says, "This device will not fail, within these given parameters." SLAs include such things as vendor penalty clauses in the event that the level of service promised isn't delivered.
An SLA and a warranty are very different things. I think it's important that we clarify this if we're to discuss this issue at all.
Now, as I understand it, Microsoft works with certain partners to offer complete managed solutions for their Datacenter product line. I'd imagine that a managed solution-- in which the hardware, software, service, and maintenance are all provided by a single vendor-- would come with an SLA. But anything less than that, no. As far as I know, there's no way Microsoft would ever offer an SLA on their software by itself.
Now, if I'm right about all those things, then I'd go on to imagine that there are-- or, at least, could be-- companies out there who will offer you a similar fully managed solution based on Red Hat Linux, complete with SLA.
I'm no Microsoft apologist, and neither am I a particular fan of Linux. I'm just trying to make sure we're comparing apples to apples here.
Ok. The word 'cluster' is also singular. Big deal.
The word "cluster" is collective. Like "dozen." It refers to a group of things as a unit. It's not possible to have a cluster of one.
Clusters of computers are not supercomputers. They're different. You can try all you like to say that the definition of "supercomputer" is arbitrary; I suppose it is, in the sense that all definitions are fundamentally arbitrary. A supercomputer is a single computer. A cluster of computers is not a single computer. Ergo, a cluster is not the same thing as a supercomputer.
If I could convince all the world's population to do maths with an abacus... that would be a supercomputer too.
No, it wouldn't. A computer-- and, by extension, a supercomputer-- is a mechanical device. A group of people could perform the same job as a supercomputer, but the group wouldn't be a supercomputer.
I think you're possibly confused by metaphors. When you say, "A group of people is a supercomputer," you're employing a metaphor to say, "A group of people shares many important characteristics with a supercomputer." Don't be fooled by this. It's not literally true, and shouldn't be accepted as such.
All things considered, pure probability would tell us that it's incredibly unlikely that you, r0t, are actually capable of comprehending, much less doing anything with, the source code to an operating system. Whether you personally have those skills or not isn't my point; I'm saying that the argument that Linux is superior to Windows because it comes with source code is specious at best. To the vast majority of users and potential users, the source code to the operating system would do nothing more than occupy CDROM-- and, if installed, hard drive-- space.
First, you're wrong. The sampling method, by which one collects a number of data points and uses them to extrapolate trends, is perfectly valid. In fact, it's really the only way to do predictive analysis.
Second, there's a better example of flawed logic; it's called a misaligned syllogism. "All men are mortal. Socrates is mortal. Therefore all men are Socrates."
It's "total cost of ownership," not "total cost overall."
Massive license fees? How do you define "massive?" A Windows server license will cost you a few thousand bucks, depending on configuration. That's a one-time charge. The guy who maintains it for you will cost you tens of thousands of dollars per year. The cost of the software license is very small in proportion to the recurring costs of owning the system. This is why it's not at all obvious that Linux-- which costs nothing to license-- should be cheaper to own that Windows.
I guess some people think you're just being an asshole; the moderations seem to imply this, anyway. But I'm curious. What kind of SLA is Microsoft willing to offer you? I've never investigated getting one, myself, and frankly the thought has never even occurred to me.
Can you provide details?
Without it, nothing works, the sky falls and chaos reigns - basically the worst parts of Ghostbusters.
Annie Potts?
The important thing to notice about the word "supercomputer" is that it's singular. A supercomputer is a single system image; this is implicit in the definition. This is not to say that supercomputing clusters aren't worthy; it's just that they're different in important ways from single-system-image supercomputers.
Some classes of problems aren't suited for cluster computation. I won't pretend to be educated enough to tell you exactly which problems can and can't be adapted for cluster computation, but consider the nature of clusters to see my point. Clusters are highly scalable, but the inter-node latency is huge. An interconnect like Myrinet can get your remote messaging latencies down to the microsecond range, but the far more common MPI/PVM-over-Ethernet solution is a thousand times slower than that. This makes it somewhat inefficient for node N to try to access a bank of memory on node M. In order for a cluster to be efficient, each node should have sufficient physical memory to hold it's entire data set, and each node should be able to operate more-or-less autonomously, without having to contact other nodes.
Supercomputers are fundamentally different from clusters. In some cases, you can do the same job with either a supercomputer or a cluster. Some jobs are better suited to clusters, while some are better suited to supercomputers. Some jobs, as I mentioned above, are better suited to arrays than to either clusters or supercomputers. It just depends on the job.
Wrong. Render farms are neither clusters nor supercomputers. At best, a render farm might be considered an array.
A supercomputer is a single system image. Some people call large clusters "supercomputers," but technically they're wrong.
A cluster is an interconnected group of computers that can communicate with each other. Usually a cluster depends on some kind of software layer to allow programs to run across multiple systems, something like MPI. Clusters are tightly interconnected many-to-many systems.
An array has a single job control system and a number of job execution systems. Batch jobs are submitted by users to the job control system, which doles them out to the various execution systems and then collects the results. The execution nodes don't talk to each other, and one job runs on one execution node at a time. Render farms are basically arrays; each execution node works on rendering a single frame of a multiframe animation. Because each frame can be rendered independently, without any dependencies on the previous and subsequent frames, rendering is particularly well suited to array computing.
With 2048 processors and an assumed 1GB/node that's 1TB of low latency, super high-jiz RAM.
How do you define "low latency?" From a first glance at the evidence, it appears that this cluster just uses plain old TCP/IP over Ethernet as its node interconnect. That's not exactly low latency access to remote memory, you know.
Just nitpickin'.
I had a boss one time who wanted all computers in the office named after biblical figures. The big servers were Matthew, Mark, and Luke, but the names got pretty obscure after that. One girl in the art department-- tattoos and piercings, never wore underwear, religious affiliations left as an exercise for the reader-- suggested we put an end to it by naming the boss's laptop Lucifer.
Not wanting to get both fired and excommunicated, I instead talked him out of it by appealing to simple reason. Nobody can spell Zephaniah, I told him.
I believe it was TIME magazine that reported this a number of months, if not a year or so ago, they claimed it to be "Planet X".
Don't forget, it's pronounced "Planet Ten," version 10.0.
Roman gods? No. Quaoar was the name of the one and only god in the mythology of one particular native american tribe, the name of which escapes me. For some reason, I remember this fact from freshman year anthropology.
Besides, I hereby announce that I'm taking bets. The official name of this body-- if it ever gets one-- will be Persephone.
Okay, I guess you're right. The normal state is to have two of each of your chromosomes, while haploid cells only have one of each. So while a sperm only has half of your genetic material, it's got a full set.
;-)
And I did pay attention in biology; it's just that it was so long ago.
As God is my witness, I have never seen a news day as slow as this one.
Gives new meaning to that old story about virii(or is it viruses?) being spread when people shake hands...
It's "viruses." And, in case you've been living your new life in the off-world colonies or something ("A chance to begin again!"), viruses are spread by shaking hands!
"The Gernsback Continuum" is his best short story; that is, it's my favorite of his short stories. It's also included in Burning Chrome, I think. Worth a read.
Don't forget that your sperm only hold half of your genes. You're overestimating the number of bytes transmitted a bit there.
It's kind of like the old saying that I'm too lazy to look up right now. "Never underestimate the bandwidth of a station wagon full of tapes tooling down the expressway."
...backbone providers have announced plans to pay people minimum wage to hold hands with each other as a backup backbone.
Sigh. The way the job markets looks right about now, I would take that job.
The guy in the old Micro Machines commercials
You mean the guy from the old Federal Express commercials? You youngsters, I swear....
2002-10-07 01:14:50 Download Porn Videos While You Kiss (articles,news) (rejected)
Perhaps now you're starting to understand the importance of a good title.
Not to be pedantic, but Aqua is a user interface appearance. What you're talking about is Quartz/AppKit, which is the user interface code that implements the Aqua appearance.
And from a read of the article, it sounds like there are no plans to create a Quartz/AppKit version of OpenOffice. They're talking about removing dependencies on X11 so it'll run on OS X without an X server, but they're not talking about replacing their homegrown and decidedly un-Mac-like UI code with true OS X UI code.
It's a shame. If that's the path they choose, the best they'll be able to do is a poor imitation of a true OS X application.
although for the size of our projects C++ is the obvious lnaguage of choice
Um... no. The language that your people are comfortable using is the language of choice. No other meaningful criteria exists.
Let's say there were some magic language, Foo, that was absolutely perfect for computational chemistry. It has standard library calls like solveThisHardProblemQuickly() and such. Just perfect.
Like any other programming language, Foo has a strict syntax. It's easy to write a buggy program in Foo, but tricky to write a perfect one. All languages are like this.
Your Fortran programmers decide they're going to abandon their preference and write in Foo instead. How many months do you think it would take for them to become conversant in Foo? How many trivial bugs will slip through in that time because your programmers weren't experienced enough with Foo to catch them? How far behind will you be at the end of the whole process?
And that's assuming that Foo is perfect. If Foo has its own idiosyncrasies and quirks, you can comfortable double those estimates.
There is only one inviolable law of computer programming: do what works.
I wrote (well, wrote most of) the engine module for an F-16 flight simulator. All in Fortran 77. Did a bunch of work on the simexec, too, which was in Ada 83.
I'm conflicted. I'm not sure if I'm bragging, or confessing.