We use RAID-3 exclusively, because our stuff requires deterministic read speeds. It's also a lot simpler to design software RAID-3 implementations because the parity generation and the rebuild algorithms are so much simpler.
We're going to start using RAID-5 for some of our new applications, though, because we just signed up to bundle HDS 9960 storage systems with our application. So that's going to be kinda exciting.
In my experience, obviously different from yours, RAID-3 and RAID-5 come up about 50/50. It just depends on what you do with it.
I'm well aware of the different pros and cons of RAID types.:) I do it for a living.
Are you sure about that?
FYI, there is no such thing as a "parity disk" when it comes to RAID.
In a RAID-3 implementation, parity data is generated for each stripe unit and stored on one disk of the array. In RAID-5, the parity data is stored across all disks of the array, a little bit in every stripe unit. (RAID-4 implements parity on the block level instead of the stripe level; it doesn't really have any advantages, so it's almost never used.)
"Quorum disks" are, as you said, something entirely else. They're related to a particular type of implementation of failover clustering, widely considered to be inferior to true highly available systems.
Perhaps you're confusing RAID with high availability. That would explain your response, I think.
In short, you're either wrong, or your post was so unclear that you might as well be wrong.
Having a "parity disk" would be contradictory to the whole point of RAID, as it represents a single point of failure for your storage. Not good.
False. Consider a three-way parity set: disks one and two contain data, and three contains parity. If you lose disk 1, you can reconstruct it from disk 2 XOR'd (or whatever; the method depends on the parity generation scheme and is irrelevant) with the parity disk, and vice-versa. And if you lose the parity disk, you reconstruct it from disks 1 and 2 XOR'd (or whatever) together. There is no single point of failure there.
In fact, set rebuilds are significantly simpler in a RAID-3 implementation than they are in RAID-5.
I ask again: are you absolutely sure that you do this for a living?
In order to have 1TB worth of DRAM on a scale next to 1TB of physical storage, you're going to have to amass like 2TB of storage on the plate in order to have just the 1TB worth of usable free space.
That isn't true at all. If you wanted to, you could mirror all of your data on two separate JBODs-- RAID level 1-- but that's not efficient. If you use RAID 3 or RAID 5, you'll never use more than 33% of your storage for parity data. As the size of your RAID set increases, the percent allocated for parity data goes down. In a 10-disk set, one disk is used for parity (in the case of RAID-3), which is only 10% of your total storage. (In the case of RAID-5, you'd still use only 10%, but you'd use 10% of each disk instead of one whole disk.)
The problem with your argument is that you assume that the only people who profit from MPEG-4 are the people who license the patent.
Not at all. I'm assuming that the only way the MPEG-4 Forum can recoup the expense of developing and publishing their spec is through licensing it, which I think is true. But if Bob Smith from Piedmont, Arizona, can write a codec or whatever that uses the MPEG-4 format without paying for it, and publish it for anybody to use, then the MPEG-4 Forum will most likely never recoup their costs. Therefore they won't be too eager to spend tons of money to develop MPEG-5.
Side note: this is similar to the Microsoft fallacy of "GPL software is bad for business".
I don't think so. I'm just providing an explanation of why patents are widely considered to be a good and important thing.
The GPL issue is something entirely else. I'll never use it myself, but that's a different conversation.
That's a pretty arrogant assumption, if you think about it. (Now don't get me wrong, I'm not accusing you of being arrogant!)
If you look at the balance sheet of the typical drug company, you will see that the LEGAL dept's budget is scary compared to R&D.
I don't believe that's the case. The drug companies in which I hold stock don't publish their specific budget figures in their annual reports, so I can find no example to prove or disprove your statement. Can you?
Except that IP laws are CLEARLY socialist in nature...
By definition, any policy that advances the good of an individual (and corporations are individuals, for legal purposes) over the "common good" (whatever that means) cannot be called socialist. Socialist policies attempt to advance the good of society (hence the name), not the individual. Communist policies advance the good of society at the expense of the individual.
What it does is STIFLE innovation (in that market) AFTER the patent is granted, UNTIL it expires.
Example?
You need to check your naivete at the door before trolling on a topic you obviously don't have much first hand experience in.
Here we are talking about the difference between real-world capitalist economics and pseudo-socialist regulated economics, and you talk about my naivete. Funny.
In other words, it's amoral. Which some people actually consider "evil".
Just because some people apply that equation doesn't make it correct. Good is good, evil is evil. These are moral judgments. Business isn't a moral activity (usually; you cite good examples of situations that are both business and moral issues, but those are the exceptions that prove the rule).
That's what so frustrating to me. When people try to apply moral rules to situations that don't call for them, silliness ensues.
Never confuse the ownership of physical property with "intellectual" property. When people are allowed to own the very ideas you think, they might as well own YOU.
But, to address your point, consider a world in which there was no protection of intellectual property. Let's say it cost $1 million to develop MPEG-4. (I have no idea what it cost. That's just an example.) The people who developed have no protection; if they release their specification, anybody who wants to could develop a product based on it. If they don't release their spec, anybody who wants to could reverse-engineer it. The MPEG-4 Forums loses $1 million.
In that case, who will develop MPEG-5? Hobbyists who do it in their spare time, as opposed to for profit? Maybe. Scientists? Again, maybe. But there have been a lot of innovations in this world, and only a few have come from those kinds of sources.
This shouldn't be a new argument to you; the same one applies to drug companies. Patents on drugs are a good thing because it encourages companies to develop new drugs that improve the quality of life for all.
Their ability to set licensing terms can be, and should be, regulated by government so that said licensing (if there's any at all) is strictly so as to achieve the maximum public benefit.
WHAT?? That's awfully Marx-esque, don't you think?
This is a capitalist system, and as such the only responsibilities that a business has is to maximize value for its shareholders and to comply with the laws of the jurisdiction(s) in which it operates.
This system is not perfect, but any system based on the premise that you advocate has been demonstrated to be ultimately a failure.
I like the old saying: capitalism is the second-worst economic system yet devised. The worst economic system is... everything else.
After all, this is the PURPOSE of IP, is it not? To prevent competition so you can make money?
Actually, the purpose of intellectual property rules and laws is to encourage innovation by making it profitable to come up with expensive, exotic new ideas. Who would spend a billion dollars on something new if somebody else could reproduce it for 1% of that and put the innovator out of business?
A business and some point in it's life crosses a line where the focus of the company is no longer on the products it produces, but rather the aquistion of revenue.
Yes, it does. That point is called "incorporation."
The job of a business is to generate revenue. It exists for no other purpose. The only question is how it goes about doing that. Sometimes moral questions arise: is it morally right for drug companies to profit while people who can't afford their products suffer and die? That's a hard question.
The issue of licensing for the MPEG-4 standard seems kind of trivial by comparison, doesn't it?
The "subtle distinction" you refer to isn't lost on me, so much as it is irrelevant to the discussion at hand.
Imagine that a law is passed to the effect that people are not allowed to own personal or real property, they must lease it from certain governmentally-approved businesses.
Okay, I'm imagining it. It has nothing to do with what we're talking about, which is the MPEG-4 Forum's choice of licensing terms. What's your point?
Everything that happens in this world that is malevolent in nature....
Are you kidding me? Malevolent? Are you sitting there at your computer with a serious face telling me that the MPEG-4 Forum's licensing scheme is evil?
Of all thing, I would have thought that the events of the past four months would have given us all a sense of perspective. Blowing up buildings is malevolent. Killing people is malevolent. Charging a per-use-hour fee to your customers is business.
(Score: 5? Are you kidding me? In the absence of any mod points this week, I reply.)
I'm wondering in the next 10 years how many things we'll no longer own, but be charged as we use them. I think there was a word for this kind of situation... serfdom.
Here I illustrate the use of the "Mad-Libs" method of argument analysis: replace the significant noun and verb in the sentence with other similar words. If the argument sounds silly, draw your own conclusion.
"I'm wondering in the next 10 years how many roads we'll no longer own, but be charged as we drive on them. I think there was a word for this kind of situation... serfdom."
If you don't want to pay the toll, don't drive on the MPEG-4 parkway. It's that simple. There's no need for this sort of hyperbole. If we're talking about wrongful imprisonment or illegal taxation or something, get as morally indignant as you want. But don't dilute that kind of rhetoric by using it on mundane issues like this one.
I see a lot of comments about X terminals and one about a dual-head graphics card.
That's not what the question-asker-person was talking about.
The Octane system that was referred to in the question is designed to support two users logged into the machine at the same time, each using a fully functional hardware graphics pipe. From the user's point of view, it's exactly like having your own Octane, complete with V12 ("really fast") graphics.
You can't do that with VNC or an X terminal, because those both involve running your application on the server and displaying its output over the LAN. That's not as fast as working on the local machine, and it doesn't provide you with hardware-accelerated graphics.
A dual-head output from your graphics card also wouldn't be enough, because to implement a system like this you'd have to have two graphics cards, one for each user.
SGI Onyx systems have supported this kind of thing for years; at one point it was marketed as a product called "GroupStation." The way they set it up wasn't particularly exotic, although it may depend on some features unique to the Xsgi X server. For instance, Xsgi has a "-pipe" option that tells it to manage the given graphics pipe; graphics pipes are numbered by the kernel at boot time, so on any multipipe machine you can refer to a specific pipe as 0, 1, 2, or whatever. I don't know if that option is unique to Xsgi or what.
Each user gets a mouse, keyboard, and monitor. The X server on the Onyx would be configured to run one instance of X on each graphics pipe, instead of one instance of X with a screen for each pipe. (In other words,:0.0 and:1.0 instead of:0.0 and:0.1.) Each X display gets an instance of xdm, so when you sit down you log in and go like you're the only user on the machine. You set up the bindings between mouse/keyboard devices and pipes in the Xservers config file, I think.
It should be possible, at least in theory, to do this with XFree86. I know multiple X sessions on one Linux machine are supported; I've done it myself with multiple XDMCP clients running at the same time. If XFree86 supports binding mouse and keyboard devices to X displays, then you should be ready to roll.
The only bad news is that you'd be limited to PCI graphics cards for your 2nd through nth heads, unless you have a motherboard with more than one AGP slot. None of the PCs at my office have more than one, so I'm not even sure whether such a beast exists.
So to sum up, I think this depends more on XFree86 than it does on Linux. If you want to try it, I'd start with the XFree86 documentation to find out if the features you'd require are supported.
This new VHS has the potential to hold up to 28Mbps (Megabits per second) of High Definition signal.
Whaaa? SMPTE has long since specified that over-the-air HDTV will be broadcast on a 19 Mbit channel. That's maximum; broadcasters can mux more than one lower-bandwidth channel into that piece of broadcast spectrum if they choose to.
So it a major feature of this new format is home recording of HDTV, those extra 9 megabits are completely wasted. There's no information there. That's nine million zeros per second.;-)
Not to nitpick, but the parity computations only impact writes, not reads, on reads RAID-5 is essentially a RAID-0.
Depends on the RAID controller. Some-- although damn it because I can't remember their names-- do parity on the read and on the write for an extra level of error checking. The idea is to catch bad data before it gets to the OS.
The human body (used here only because it's the most familiar to the average person) works. It has some problems, but the design is solid.
I think the human body can only be said to work in the statistical sense. Pick any given cell, and you'll find that the body (any complex organism, really) is a pretty dangerous place. The body works as designed because the component parts are unbelievably vast in numbers and practically (in fact, literally) disposable.
Stephenson dealt with an idea like this in The Diamond Age, his book about nanotech. The idea is that, because of an absurd but logical application of economies of scale, it's about as expensive to produce one nanotechnological computer as it is to produce one trillion of them.
If we lived in a world like that, where fairly autonomous disposable computers could be practically manufactured and used, the "computer network as biological system" idea might make some sense.
Remember that life as we have observed it is basically tuned to the idea that the problem is hard, but the raw materials are cheap and time is no object. The only thing that situation has in common with our world is that the problems are hard; in our case, the materials are really expensive (in dollars, but also in labor and opportunity cost) and time is of the essence.
That's not an area in which biology does very well.
In short, this is an R14k single proc Octane, with no XIO, not too sure about the backplane, as there do not seem to be any fuel related docs up on techpubs yet.
I have a PowerPoint that I don't think I can redistribute, but the info in it isn't secret any more.
Think Origin 300 for Fuel's internals. The CPU is connected to the Bedrock ASIC (memory controller and system crossbow) via a 200 MHz 64-bit FSB (1.6 GB per second). The Bedrock interfaces to the RAM over a 200 MHz DDR bus (3.2 GB per second) and to the XBridge, which is another crossbow and protocol translator.
XBridge interfaces to the graphics board, which has a new type of interface that's sort of like a cross between XIO and PCI. It's proprietary, and it's 1.6 GB per sec.
XBridge also interfaces to the two external PCI busses-- one 64 bit/33 MHz with two slots, and one 64 bit/66 MHz with two slots. Also, XBridge connects the internal stuff like the serial ports, Ethernet, internal SCSI bus, and so on.
This is exactly the same internal architecture used in the Origin/Onyx 300 and Origin/Onyx 3000.
Also, it says it has space for up to three drives, but doesn't mention any sort of hardware-based RAID feature.
Because one of those three drives is the system drive. You can only do striping or mirroring with two disks, and the OS supports both of those in software through XVM or XLV. There's basically no reason to have internal hardware RAID.
This coupled with the fact that no SGI system has ever had internal hardware RAID. It seems that the customers aren't exactly clamoring for it.
It's probably worth mentioning that Fuel maxes out at 4 GB of RAM while Octane2 can have 8 GB.
Another key differentiator is that Fuel is exclusively a single-user workstation. Octane2 has a dual-head option (two V12 graphics pipes, each with an option dual-channel display adapter) and can take two mouse/keyboard combos. Two users can use an Octane2 at the same time with full graphics functionality. Can't do that with Fuel.
Sheeeeeit... when I take a ~7 hour flight from DC to London or Paris, I sleep for at least 2-3 hours of it!
Depends on how you fly. The best way I've found to go LA-Sydney is to leave at about 11:30 at night and get in at 8:00 in the morning Sydney time. The flight is so long that by the time you get there you've had a few hours of sleep and your body clock is totally reset.
But you can't always do it that way. I flew from Sydney to LA from noon to noon and didn't get a wink of sleep the whole way. My body just couldn't deal. So 14 hours of batter would have been a good thing.
Slashdot Mirror
Okay, I knew there had to be some explanation. ;-)
We use RAID-3 exclusively, because our stuff requires deterministic read speeds. It's also a lot simpler to design software RAID-3 implementations because the parity generation and the rebuild algorithms are so much simpler.
We're going to start using RAID-5 for some of our new applications, though, because we just signed up to bundle HDS 9960 storage systems with our application. So that's going to be kinda exciting.
In my experience, obviously different from yours, RAID-3 and RAID-5 come up about 50/50. It just depends on what you do with it.
I'm well aware of the different pros and cons of RAID types. :) I do it for a living.
Are you sure about that?
FYI, there is no such thing as a "parity disk" when it comes to RAID.
In a RAID-3 implementation, parity data is generated for each stripe unit and stored on one disk of the array. In RAID-5, the parity data is stored across all disks of the array, a little bit in every stripe unit. (RAID-4 implements parity on the block level instead of the stripe level; it doesn't really have any advantages, so it's almost never used.)
"Quorum disks" are, as you said, something entirely else. They're related to a particular type of implementation of failover clustering, widely considered to be inferior to true highly available systems.
Perhaps you're confusing RAID with high availability. That would explain your response, I think.
In short, you're either wrong, or your post was so unclear that you might as well be wrong.
Having a "parity disk" would be contradictory to the whole point of RAID, as it represents a single point of failure for your storage. Not good.
False. Consider a three-way parity set: disks one and two contain data, and three contains parity. If you lose disk 1, you can reconstruct it from disk 2 XOR'd (or whatever; the method depends on the parity generation scheme and is irrelevant) with the parity disk, and vice-versa. And if you lose the parity disk, you reconstruct it from disks 1 and 2 XOR'd (or whatever) together. There is no single point of failure there.
In fact, set rebuilds are significantly simpler in a RAID-3 implementation than they are in RAID-5.
I ask again: are you absolutely sure that you do this for a living?
In order to have 1TB worth of DRAM on a scale next to 1TB of physical storage, you're going to have to amass like 2TB of storage on the plate in order to have just the 1TB worth of usable free space.
That isn't true at all. If you wanted to, you could mirror all of your data on two separate JBODs-- RAID level 1-- but that's not efficient. If you use RAID 3 or RAID 5, you'll never use more than 33% of your storage for parity data. As the size of your RAID set increases, the percent allocated for parity data goes down. In a 10-disk set, one disk is used for parity (in the case of RAID-3), which is only 10% of your total storage. (In the case of RAID-5, you'd still use only 10%, but you'd use 10% of each disk instead of one whole disk.)
The problem with your argument is that you assume that the only people who profit from MPEG-4 are the people who license the patent.
;-)
Not at all. I'm assuming that the only way the MPEG-4 Forum can recoup the expense of developing and publishing their spec is through licensing it, which I think is true. But if Bob Smith from Piedmont, Arizona, can write a codec or whatever that uses the MPEG-4 format without paying for it, and publish it for anybody to use, then the MPEG-4 Forum will most likely never recoup their costs. Therefore they won't be too eager to spend tons of money to develop MPEG-5.
Side note: this is similar to the Microsoft fallacy of "GPL software is bad for business".
I don't think so. I'm just providing an explanation of why patents are widely considered to be a good and important thing.
The GPL issue is something entirely else. I'll never use it myself, but that's a different conversation.
That's a pretty arrogant assumption, if you think about it. (Now don't get me wrong, I'm not accusing you of being arrogant!)
Oh, heck. Go right ahead. Everybody else does.
If you look at the balance sheet of the typical drug company, you will see that the LEGAL dept's budget is scary compared to R&D.
I don't believe that's the case. The drug companies in which I hold stock don't publish their specific budget figures in their annual reports, so I can find no example to prove or disprove your statement. Can you?
Except that IP laws are CLEARLY socialist in nature...
By definition, any policy that advances the good of an individual (and corporations are individuals, for legal purposes) over the "common good" (whatever that means) cannot be called socialist. Socialist policies attempt to advance the good of society (hence the name), not the individual. Communist policies advance the good of society at the expense of the individual.
What it does is STIFLE innovation (in that market) AFTER the patent is granted, UNTIL it expires.
Example?
You need to check your naivete at the door before trolling on a topic you obviously don't have much first hand experience in.
Here we are talking about the difference between real-world capitalist economics and pseudo-socialist regulated economics, and you talk about my naivete. Funny.
In other words, it's amoral. Which some people actually consider "evil".
Just because some people apply that equation doesn't make it correct. Good is good, evil is evil. These are moral judgments. Business isn't a moral activity (usually; you cite good examples of situations that are both business and moral issues, but those are the exceptions that prove the rule).
That's what so frustrating to me. When people try to apply moral rules to situations that don't call for them, silliness ensues.
Never confuse the ownership of physical property with "intellectual" property. When people are allowed to own the very ideas you think, they might as well own YOU.
This is, of course, a "straw man" fallacy.
But, to address your point, consider a world in which there was no protection of intellectual property. Let's say it cost $1 million to develop MPEG-4. (I have no idea what it cost. That's just an example.) The people who developed have no protection; if they release their specification, anybody who wants to could develop a product based on it. If they don't release their spec, anybody who wants to could reverse-engineer it. The MPEG-4 Forums loses $1 million.
In that case, who will develop MPEG-5? Hobbyists who do it in their spare time, as opposed to for profit? Maybe. Scientists? Again, maybe. But there have been a lot of innovations in this world, and only a few have come from those kinds of sources.
This shouldn't be a new argument to you; the same one applies to drug companies. Patents on drugs are a good thing because it encourages companies to develop new drugs that improve the quality of life for all.
Their ability to set licensing terms can be, and should be, regulated by government so that said licensing (if there's any at all) is strictly so as to achieve the maximum public benefit.
WHAT?? That's awfully Marx-esque, don't you think?
This is a capitalist system, and as such the only responsibilities that a business has is to maximize value for its shareholders and to comply with the laws of the jurisdiction(s) in which it operates.
This system is not perfect, but any system based on the premise that you advocate has been demonstrated to be ultimately a failure.
I like the old saying: capitalism is the second-worst economic system yet devised. The worst economic system is... everything else.
After all, this is the PURPOSE of IP, is it not? To prevent competition so you can make money?
Actually, the purpose of intellectual property rules and laws is to encourage innovation by making it profitable to come up with expensive, exotic new ideas. Who would spend a billion dollars on something new if somebody else could reproduce it for 1% of that and put the innovator out of business?
A business and some point in it's life crosses a line where the focus of the company is no longer on the products it produces, but rather the aquistion of revenue.
Yes, it does. That point is called "incorporation."
The job of a business is to generate revenue. It exists for no other purpose. The only question is how it goes about doing that. Sometimes moral questions arise: is it morally right for drug companies to profit while people who can't afford their products suffer and die? That's a hard question.
The issue of licensing for the MPEG-4 standard seems kind of trivial by comparison, doesn't it?
The "subtle distinction" you refer to isn't lost on me, so much as it is irrelevant to the discussion at hand.
Imagine that a law is passed to the effect that people are not allowed to own personal or real property, they must lease it from certain governmentally-approved businesses.
Okay, I'm imagining it. It has nothing to do with what we're talking about, which is the MPEG-4 Forum's choice of licensing terms. What's your point?
Everything that happens in this world that is malevolent in nature....
Are you kidding me? Malevolent? Are you sitting there at your computer with a serious face telling me that the MPEG-4 Forum's licensing scheme is evil?
Of all thing, I would have thought that the events of the past four months would have given us all a sense of perspective. Blowing up buildings is malevolent. Killing people is malevolent. Charging a per-use-hour fee to your customers is business.
Get a grip.
(Score: 5? Are you kidding me? In the absence of any mod points this week, I reply.)
I'm wondering in the next 10 years how many things we'll no longer own, but be charged as we use them. I think there was a word for this kind of situation... serfdom.
Here I illustrate the use of the "Mad-Libs" method of argument analysis: replace the significant noun and verb in the sentence with other similar words. If the argument sounds silly, draw your own conclusion.
"I'm wondering in the next 10 years how many roads we'll no longer own, but be charged as we drive on them. I think there was a word for this kind of situation... serfdom."
If you don't want to pay the toll, don't drive on the MPEG-4 parkway. It's that simple. There's no need for this sort of hyperbole. If we're talking about wrongful imprisonment or illegal taxation or something, get as morally indignant as you want. But don't dilute that kind of rhetoric by using it on mundane issues like this one.
I see a lot of comments about X terminals and one about a dual-head graphics card.
:0.0 and :1.0 instead of :0.0 and :0.1.) Each X display gets an instance of xdm, so when you sit down you log in and go like you're the only user on the machine. You set up the bindings between mouse/keyboard devices and pipes in the Xservers config file, I think.
That's not what the question-asker-person was talking about.
The Octane system that was referred to in the question is designed to support two users logged into the machine at the same time, each using a fully functional hardware graphics pipe. From the user's point of view, it's exactly like having your own Octane, complete with V12 ("really fast") graphics.
You can't do that with VNC or an X terminal, because those both involve running your application on the server and displaying its output over the LAN. That's not as fast as working on the local machine, and it doesn't provide you with hardware-accelerated graphics.
A dual-head output from your graphics card also wouldn't be enough, because to implement a system like this you'd have to have two graphics cards, one for each user.
SGI Onyx systems have supported this kind of thing for years; at one point it was marketed as a product called "GroupStation." The way they set it up wasn't particularly exotic, although it may depend on some features unique to the Xsgi X server. For instance, Xsgi has a "-pipe" option that tells it to manage the given graphics pipe; graphics pipes are numbered by the kernel at boot time, so on any multipipe machine you can refer to a specific pipe as 0, 1, 2, or whatever. I don't know if that option is unique to Xsgi or what.
Each user gets a mouse, keyboard, and monitor. The X server on the Onyx would be configured to run one instance of X on each graphics pipe, instead of one instance of X with a screen for each pipe. (In other words,
It should be possible, at least in theory, to do this with XFree86. I know multiple X sessions on one Linux machine are supported; I've done it myself with multiple XDMCP clients running at the same time. If XFree86 supports binding mouse and keyboard devices to X displays, then you should be ready to roll.
The only bad news is that you'd be limited to PCI graphics cards for your 2nd through nth heads, unless you have a motherboard with more than one AGP slot. None of the PCs at my office have more than one, so I'm not even sure whether such a beast exists.
So to sum up, I think this depends more on XFree86 than it does on Linux. If you want to try it, I'd start with the XFree86 documentation to find out if the features you'd require are supported.
[A]ccording to the article they got a sixth-grader in Hawaii to set one up.
This is so simple, a sixth-grader could understand it!
Bring me a sixth grader!
This new VHS has the potential to hold up to 28Mbps (Megabits per second) of High Definition signal.
;-)
Whaaa? SMPTE has long since specified that over-the-air HDTV will be broadcast on a 19 Mbit channel. That's maximum; broadcasters can mux more than one lower-bandwidth channel into that piece of broadcast spectrum if they choose to.
So it a major feature of this new format is home recording of HDTV, those extra 9 megabits are completely wasted. There's no information there. That's nine million zeros per second.
I work for a company developing a filesystem which is truly symmetric, and hence does not suffer from these issues.
How is your product different from CXFS?
Not to nitpick, but the parity computations only impact writes, not reads, on reads RAID-5 is essentially a RAID-0.
Depends on the RAID controller. Some-- although damn it because I can't remember their names-- do parity on the read and on the write for an extra level of error checking. The idea is to catch bad data before it gets to the OS.
The human body (used here only because it's the most familiar to the average person) works. It has some problems, but the design is solid.
I think the human body can only be said to work in the statistical sense. Pick any given cell, and you'll find that the body (any complex organism, really) is a pretty dangerous place. The body works as designed because the component parts are unbelievably vast in numbers and practically (in fact, literally) disposable.
Stephenson dealt with an idea like this in The Diamond Age, his book about nanotech. The idea is that, because of an absurd but logical application of economies of scale, it's about as expensive to produce one nanotechnological computer as it is to produce one trillion of them.
If we lived in a world like that, where fairly autonomous disposable computers could be practically manufactured and used, the "computer network as biological system" idea might make some sense.
Remember that life as we have observed it is basically tuned to the idea that the problem is hard, but the raw materials are cheap and time is no object. The only thing that situation has in common with our world is that the problems are hard; in our case, the materials are really expensive (in dollars, but also in labor and opportunity cost) and time is of the essence.
That's not an area in which biology does very well.
In short, this is an R14k single proc Octane, with no XIO, not too sure about the backplane, as there do not seem to be any fuel related docs up on techpubs yet.
I have a PowerPoint that I don't think I can redistribute, but the info in it isn't secret any more.
Think Origin 300 for Fuel's internals. The CPU is connected to the Bedrock ASIC (memory controller and system crossbow) via a 200 MHz 64-bit FSB (1.6 GB per second). The Bedrock interfaces to the RAM over a 200 MHz DDR bus (3.2 GB per second) and to the XBridge, which is another crossbow and protocol translator.
XBridge interfaces to the graphics board, which has a new type of interface that's sort of like a cross between XIO and PCI. It's proprietary, and it's 1.6 GB per sec.
XBridge also interfaces to the two external PCI busses-- one 64 bit/33 MHz with two slots, and one 64 bit/66 MHz with two slots. Also, XBridge connects the internal stuff like the serial ports, Ethernet, internal SCSI bus, and so on.
This is exactly the same internal architecture used in the Origin/Onyx 300 and Origin/Onyx 3000.
Also, it says it has space for up to three drives, but doesn't mention any sort of hardware-based RAID feature.
Because one of those three drives is the system drive. You can only do striping or mirroring with two disks, and the OS supports both of those in software through XVM or XLV. There's basically no reason to have internal hardware RAID.
This coupled with the fact that no SGI system has ever had internal hardware RAID. It seems that the customers aren't exactly clamoring for it.
Fuel has 3.2x RAM...
It's probably worth mentioning that Fuel maxes out at 4 GB of RAM while Octane2 can have 8 GB.
Another key differentiator is that Fuel is exclusively a single-user workstation. Octane2 has a dual-head option (two V12 graphics pipes, each with an option dual-channel display adapter) and can take two mouse/keyboard combos. Two users can use an Octane2 at the same time with full graphics functionality. Can't do that with Fuel.
Sheeeeeit... when I take a ~7 hour flight from DC to London or Paris, I sleep for at least 2-3 hours of it!
Depends on how you fly. The best way I've found to go LA-Sydney is to leave at about 11:30 at night and get in at 8:00 in the morning Sydney time. The flight is so long that by the time you get there you've had a few hours of sleep and your body clock is totally reset.
But you can't always do it that way. I flew from Sydney to LA from noon to noon and didn't get a wink of sleep the whole way. My body just couldn't deal. So 14 hours of batter would have been a good thing.