Has it occurred to you to ask the manufacturer (Alias|Wavefront) rather than asking a bunch of high school and college kids on Slashdot?
Has it occured to you that many of us aren't high school or college kids, and many of us have years of experience that might be helpful. Disregarding the fact that usenet is the correct place to ask questions like this because you'll get a more targeted audience, there's an excelent chance that an ask slashdot question will recieve a few truly insightful answers from people who do something like this for a living, and know what they're talking about.
If you don't like Ask Slashdot, there's a nice checkbox that'll remove it from your front page.
Some Fujitsu drives manufactured just over a year ago have faulty chips from Cirrus Logic on them that cause the controllers to fail. Check the article for details. I believe there's a class action lawsuit in progress that you can join.
Agreed, but if you're running an Oracle database on top of all this, Id venture to guess you had a fairly large budget.
You don't need the multiple thousand dollar switch to do Fibre Channel in a dual host configuration either. You can have a full working dual loop setup for under $2k+HBAs.
I am quite impressed with the total cost of this though. At $10/per Firewire card, you can have a setup to play with that is cheaper than just the cables in a shared SCSI cluster.
The Firewire cards needed to build a cluster can cost as little as 10% as much as the required FiberChannel hardware
Not to mention the FiberChannel switch. The Brocade [brocade.com] fiber switch we use to tie our three SGI Origins to our SAN's storage RAID was over CA$12K when we bought it.
Yeah, but you only get 20% of the speed. Fibre Channel is at 2048Mbps now, compared to the 400Mbps of Firewire.
How many people have used a computer but never, ever used Windows
That is not entirely the point. What matters is which system they used first. A significant percentage of computer users used a Mac before Windows.
And if it's a Windowsism, why do Mac apps have context menus at all?
Because alot of mac software is written by microsoft, and the context menu was brought to the mac with it. Pre OSX, context menus were an OS extension. They weren't part of the system. Also, apple is trying to get people to switch, so they want windows users to be comfortable on the new platform.
I don't know what your issue is. Most mac users use and enjoy the one button mouse. For everybody else, you can spend an additional $5 for another button.
EVERY Mac user I know does the same thing upon un-boxing their new Macs
All you've demonstarated is that you know a very small portion of the mac user community. Most mac users use the included mouse. It's no harder to hold down the option key then it is to right click, and there's no memorization invloved. Did you have to memorize the right mouse button?
OS X is even written with a two button scrollwheel mouse in mind!
Really? I don't see how you mean. I have a mac with a three button scrollwheel mouse (I *HATE* scrollwheels, though), and the only time I use a button other then the left one is in Microsoft software.
That's an overstatement. ATA/IDE/whatever storage is pretty cheap, but SCSI and Fibre Channel disks are still pricey. In order to protect a 1 TB filesystem with RAID 0+1, you'd have to have 2 TB worth of (let's say) Fibre Channel drives. That extra terabyte would cost you many thousands of dollars. But to protect the same filesystem with RAID 3 or RAID 5, you only have to have (at least) one spare drive. That's a lot cheaper than the 6 or 8 or 16 or whatever drives you'd have to buy to mirror the whole filesystem.
Basically, you've got to spend 33% more on your storage to use 0+1 instead of RAID 5. For smaller arrays, that's not a lot of actual dollars. The only real place I see for RAID 5 anymore is for low-end, medium-reliability configurations with more than 500Gb of storage. If you're a low end user with a 1TB array, then you'll go for the 33% savings. If you're not backing up your array, you'll probably not be spending the extra 33% (even though you probably should!). If you're backing up your terabyte array, and you are in an enterprise environment, the cost of your backup solution and your dedicated redundant caching controllers dwarfs the 33% increase in cost to go from raid 5 to raid 0+1. Also, the cost increase is typically justified by the increased performance of the 0+1 solution.
You're right, RIAD 5 isn't dead, but it's niche is shrinking.
It took me a minute to find the button that was going to give me a context menu. Sigh. I always thought that it was just an old joke/troll but seriously, why?
Because the right-click context menu is a windows-ism, and as such, people who have never used windows don't care. In fact, if you gave them another button they wouldn't use it, much like how windows users don't care they are missing the ever-so-useful middle button.
People who do care plug any old multi-button USB mouse into their mac and forget about it.
MacOS X version 2 MacOS X version 3 . . . MacOS X version 8
And so on. Wasn't so hard, now was it? It won't annoy me as much as advertising that uses three single word "sentences" that are also supposed to go together, either. Man, those piss me off.
I was pretty upset when I discovered that they only came with raid1 and raid0.
Standard practice nowadays is to use RAID 0 and RAID 1 together instead of using RAID 5. The data protection is better, and the performance is too. You should make mirror sets, and then stripe them rather than the other way around. That way your system can keep running at full speed if any single disk breaks. The other thing you may consider (I don't know if this is possible under OSX, but it should be) Is to RAID your partitions instead of partitioning your RAID. This should overcome your filesystem issue.
With the low cost of storage these days, RAID 5 is basically obsolete. Spend the extra few gigabytes, and use RAID 0+1
My TiVo generally has more accurate guide data than the DirecTV box. Don't ask me how they manage that, but it's true. Anyway, the TiVo is connected to the box with a serial cable. No IR necissary. You can also get the DirecTivo integrated with the sattelite box, and a similar model for AT&T digital cable. You just chose the wrong provider. Being TiVo friendly was the primary reason I chose DirecTV over dish network.
I don't even know where to start. You'd rather pay more per page to avoid the initail toner charge? My ink jet printer costs less than the cartridge. That's why it's in the box unused... Anyway, back on topic: It was alright that you would pay half the cost of the device just to use it as long as they folded that ammount into the overall price, but now that they've seperated it out, and the total price is still the same, it's unacceptable? No wonder so much stuff is $0.99, and we have a $0.009 at the end of the gas price. That stupid person marketing stuff really works.
Personally, I would love to see a USA federal tax on Gas. That would rise by 5 cents/year for the next 5 years and then rise by 10 / year.
What exactly would that acomplish? Without suitable public transportation, people will be forced to pay the higher prices just to get to work. Salaries will have to increase, and we'll have massive inflation. Oil consumption will not change.
Solve the problem (poor transportation), not the symptom (high oil consumption).
The user's private key is just for authentication. The data keys are seperate. To decrypt the outgoing data, the sysadmin would require the private key that was generated by the remote server. At best he could hope to pull the local key out of the clients system memory and decrypt the incoming stream, and if he can pull that off without the user noticing, he could probably get a job that pays much better than sysadmin.
Multiply that value by 512 bytes if you are using one address per hard drive sector.
Aside from the bad math, the issue here was that they memory mapped the disk. That means you have to byte address the disk instead of sector address the disk.
But how soon will IBM's pixie dust make possible disks with partitions larger than can be addressed in 64-bits?
Considering that you can byte address billions of gigabytes with a 64-bit address space, I'm going to go out on a limb and say that there will never be an IBM pixie dust hard drive that can't be mapped with 64 bits.
The processor with the highest number of valid virtual address bits currently available uses 54 bits of the available 64. Even this machine is not in danger of being unable to address all of a disk that is made with currently available technology.
Um, no. I hate to burst your "I bought an Xbox, so it damn well better succeed" bubble, but Microsoft is demonstrateably nowhere near having brought in $1 billion in revenue from the Xbox. Look at their last few quarterly reports. You'll see two things. First, they group Xbox revenue in with revenue from MSN subscriptions, PC games sales, and consumer software. Basically everything microsoft makes that's not Office, Windows, or a server product, and all the revenue (not profits, revenue) from MSN. Second you'll see that they bring in under $2 billion a year with all those things. There is NO WAY that Xbox accounts for half of Microsoft's non Windows/Office sales, especially since those numbers aren't significantly increased from the previous year when Xbox didn't exist. Not only that, but revenue figures don't take into account the expenditure for building each device.
Estimates I've seen tend to agree that Microsoft must sell between 10 and 20 games to break even on an Xbox sale. How many games do you have for your Xbox?
The xbox is selling very well
Again, bullshit. There's loads of market research that shows Xbox in an uphill battle for second place. http://www.instat.com/press.asp?ID=390&sku=IN02004 1ME
You need to shell out $3k for the numbers, but it's not to hard to figure it out from the abstract. If sony sold >20 million consoles in the last 12 months, 31 million consoles were sold overall, and Nintendo and Microsoft have sold roughly equal numbers of consoles, you can see that the Xbox is not selling very well compared to PS2. That means there are 40 million PS2 in people's houses, and ~5 million Xboxes out there. Also, if you head over to NPDFunworld, you'll see that for the last 6 months Xbox has had on average 1 game in the monthly top ten based on sales. That's not anything to write home about, especially when there are typically 6 PS2 games and 3 GBA games on the list.
Maybe someday there will be enough good Xbox exclusive titles out there to get more people to shell out the $200 for an Xbox, but with microsoft already planning on releasing Xbox 2 in just over 2 years, and 90% of the good xbox games being available on other platforms, Xbox sales may not be picking up anytime soon.
I'm assuming you have a fast net connection. If you don't, please disregard...
Download the debian boot-floppies. You can do a full net instalation, and you'll only install what you need instead of downloading the entire contents of some distro's ISOs.
If you're not afraid of trying a development kernel with a beta filesystem patch, then the debian installation process should be simple.
Seriously, though. It looks like this new filesystem patch isn't quite ready. They're still finding leaked blocks and other corruption. You should run a "stable" kernel on your primary box, or you should do frequent backups! My primary workstation is running 2.4.17 (Hasn't had any downtime since 2.4.17 was released, so I haven't upgraded.), and I reserve the 2.5 series kernels for the rack of test machines behind me. If they break it's OK, but if I loose my emacs session I get seriously pissed off.
-- No need for flameproof armor. My home box runs windows most of the time. I hack Linux all day at work, so I want to use my home box for gaming. No better OS for that right now than windows. --
I honestly want to know if it is a huge increase, or a small increase.
As in all things benchmarking related, the answer is: It depends. This will be significant for certain uses of your system, but unimportant for others. If you've got a very busy file server, a news server, or a build machine where you do alot of compilation, this will be very significant. For other tasks you might not even notice.
This is a kernel change, so there won't be any ISOs. Why not just try it now?
Slashdot Mirror
Probably way better than it did on the Honeywell 6180 it first ran on.
:)
Kids today... Thinking a 50mhz 486 is slow! 8MB of RAM? That's gigantic!
Has it occurred to you to ask the manufacturer (Alias|Wavefront) rather than asking a bunch of high school and college kids on Slashdot?
Has it occured to you that many of us aren't high school or college kids, and many of us have years of experience that might be helpful. Disregarding the fact that usenet is the correct place to ask questions like this because you'll get a more targeted audience, there's an excelent chance that an ask slashdot question will recieve a few truly insightful answers from people who do something like this for a living, and know what they're talking about.
If you don't like Ask Slashdot, there's a nice checkbox that'll remove it from your front page.
Some Fujitsu drives manufactured just over a year ago have faulty chips from Cirrus Logic on them that cause the controllers to fail. Check the article for details. I believe there's a class action lawsuit in progress that you can join.
Agreed, but if you're running an Oracle database on top of all this, Id venture to guess you had a fairly large budget.
You don't need the multiple thousand dollar switch to do Fibre Channel in a dual host configuration either. You can have a full working dual loop setup for under $2k+HBAs.
I am quite impressed with the total cost of this though. At $10/per Firewire card, you can have a setup to play with that is cheaper than just the cables in a shared SCSI cluster.
The Firewire cards needed to build a cluster can cost as little as 10% as much as the required FiberChannel hardware
Not to mention the FiberChannel switch. The Brocade [brocade.com] fiber switch we use to tie our three SGI Origins to our SAN's storage RAID was over CA$12K when we bought it.
Yeah, but you only get 20% of the speed. Fibre Channel is at 2048Mbps now, compared to the 400Mbps of Firewire.
Just so you don't confuse the newbies, I'm sure what you meant to say was
/etc/apt/sources.list"
echo "deb http://mplayer.nmeos.net/ unstable main" >>
apt-get update
apt-get install mplayer-686
apt-get install mencoder-686
At least that's what you say if you wanted it to work...
How many people have used a computer but never, ever used Windows
That is not entirely the point. What matters is which system they used first. A significant percentage of computer users used a Mac before Windows.
And if it's a Windowsism, why do Mac apps have context menus at all?
Because alot of mac software is written by microsoft, and the context menu was brought to the mac with it. Pre OSX, context menus were an OS extension. They weren't part of the system. Also, apple is trying to get people to switch, so they want windows users to be comfortable on the new platform.
I don't know what your issue is. Most mac users use and enjoy the one button mouse. For everybody else, you can spend an additional $5 for another button.
EVERY Mac user I know does the same thing upon un-boxing their new Macs
All you've demonstarated is that you know a very small portion of the mac user community. Most mac users use the included mouse. It's no harder to hold down the option key then it is to right click, and there's no memorization invloved. Did you have to memorize the right mouse button?
OS X is even written with a two button scrollwheel mouse in mind!
Really? I don't see how you mean. I have a mac with a three button scrollwheel mouse (I *HATE* scrollwheels, though), and the only time I use a button other then the left one is in Microsoft software.
That's an overstatement. ATA/IDE/whatever storage is pretty cheap, but SCSI and Fibre Channel disks are still pricey. In order to protect a 1 TB filesystem with RAID 0+1, you'd have to have 2 TB worth of (let's say) Fibre Channel drives. That extra terabyte would cost you many thousands of dollars. But to protect the same filesystem with RAID 3 or RAID 5, you only have to have (at least) one spare drive. That's a lot cheaper than the 6 or 8 or 16 or whatever drives you'd have to buy to mirror the whole filesystem.
Basically, you've got to spend 33% more on your storage to use 0+1 instead of RAID 5. For smaller arrays, that's not a lot of actual dollars. The only real place I see for RAID 5 anymore is for low-end, medium-reliability configurations with more than 500Gb of storage. If you're a low end user with a 1TB array, then you'll go for the 33% savings. If you're not backing up your array, you'll probably not be spending the extra 33% (even though you probably should!). If you're backing up your terabyte array, and you are in an enterprise environment, the cost of your backup solution and your dedicated redundant caching controllers dwarfs the 33% increase in cost to go from raid 5 to raid 0+1. Also, the cost increase is typically justified by the increased performance of the 0+1 solution.
You're right, RIAD 5 isn't dead, but it's niche is shrinking.
It took me a minute to find the button that was going to give me a context menu. Sigh. I always thought that it was just an old joke/troll but seriously, why?
Because the right-click context menu is a windows-ism, and as such, people who have never used windows don't care. In fact, if you gave them another button they wouldn't use it, much like how windows users don't care they are missing the ever-so-useful middle button.
People who do care plug any old multi-button USB mouse into their mac and forget about it.
MacOS X version 2
MacOS X version 3
.
.
.
MacOS X version 8
And so on. Wasn't so hard, now was it? It won't annoy me as much as advertising that uses three single word "sentences" that are also supposed to go together, either. Man, those piss me off.
I was pretty upset when I discovered that they only came with raid1 and raid0.
Standard practice nowadays is to use RAID 0 and RAID 1 together instead of using RAID 5. The data protection is better, and the performance is too. You should make mirror sets, and then stripe them rather than the other way around. That way your system can keep running at full speed if any single disk breaks. The other thing you may consider (I don't know if this is possible under OSX, but it should be) Is to RAID your partitions instead of partitioning your RAID. This should overcome your filesystem issue.
With the low cost of storage these days, RAID 5 is basically obsolete. Spend the extra few gigabytes, and use RAID 0+1
My TiVo generally has more accurate guide data than the DirecTV box. Don't ask me how they manage that, but it's true. Anyway, the TiVo is connected to the box with a serial cable. No IR necissary. You can also get the DirecTivo integrated with the sattelite box, and a similar model for AT&T digital cable. You just chose the wrong provider. Being TiVo friendly was the primary reason I chose DirecTV over dish network.
Sorry, buddy. HDD MTBF is ~5 years. These things are going to be broken before you.
If it doesn't break, though, then theoretically the subscription will last longer than your lifetime.
Just think of the cost of the device as being higher.
TiVo does what you want without a subscription. You just don't subscribe and you have to enter what you want recorded by time and channel.
If you're going to do that you may as just use a VCR though.
Wow.
I don't even know where to start. You'd rather pay more per page to avoid the initail toner charge? My ink jet printer costs less than the cartridge. That's why it's in the box unused... Anyway, back on topic: It was alright that you would pay half the cost of the device just to use it as long as they folded that ammount into the overall price, but now that they've seperated it out, and the total price is still the same, it's unacceptable? No wonder so much stuff is $0.99, and we have a $0.009 at the end of the gas price. That stupid person marketing stuff really works.
Personally, I would love to see a USA federal tax on Gas. That would rise by 5 cents /year for the next 5 years and then rise by 10 / year.
What exactly would that acomplish? Without suitable public transportation, people will be forced to pay the higher prices just to get to work. Salaries will have to increase, and we'll have massive inflation. Oil consumption will not change.
Solve the problem (poor transportation), not the symptom (high oil consumption).
The user's private key is just for authentication. The data keys are seperate. To decrypt the outgoing data, the sysadmin would require the private key that was generated by the remote server. At best he could hope to pull the local key out of the clients system memory and decrypt the incoming stream, and if he can pull that off without the user noticing, he could probably get a job that pays much better than sysadmin.
Just curious, how do you propose that you use ngrep to detect http traffic tunneled through an encrypted ssh tunnel?
Multiply that value by 512 bytes if you are using one address per hard drive sector.
Aside from the bad math, the issue here was that they memory mapped the disk. That means you have to byte address the disk instead of sector address the disk.
I'm guessing that red LEDs being the cheapest might have had something to do with the decision.
But how soon will IBM's pixie dust make possible disks with partitions larger than can be addressed in 64-bits?
Considering that you can byte address billions of gigabytes with a 64-bit address space, I'm going to go out on a limb and say that there will never be an IBM pixie dust hard drive that can't be mapped with 64 bits.
The processor with the highest number of valid virtual address bits currently available uses 54 bits of the available 64. Even this machine is not in danger of being unable to address all of a disk that is made with currently available technology.
hell, microsoft will be in the black shortly.
4 1ME
Um, no. I hate to burst your "I bought an Xbox, so it damn well better succeed" bubble, but Microsoft is demonstrateably nowhere near having brought in $1 billion in revenue from the Xbox. Look at their last few quarterly reports. You'll see two things. First, they group Xbox revenue in with revenue from MSN subscriptions, PC games sales, and consumer software. Basically everything microsoft makes that's not Office, Windows, or a server product, and all the revenue (not profits, revenue) from MSN. Second you'll see that they bring in under $2 billion a year with all those things. There is NO WAY that Xbox accounts for half of Microsoft's non Windows/Office sales, especially since those numbers aren't significantly increased from the previous year when Xbox didn't exist. Not only that, but revenue figures don't take into account the expenditure for building each device.
Estimates I've seen tend to agree that Microsoft must sell between 10 and 20 games to break even on an Xbox sale. How many games do you have for your Xbox?
The xbox is selling very well
Again, bullshit. There's loads of market research that shows Xbox in an uphill battle for second place. http://www.instat.com/press.asp?ID=390&sku=IN0200
You need to shell out $3k for the numbers, but it's not to hard to figure it out from the abstract. If sony sold >20 million consoles in the last 12 months, 31 million consoles were sold overall, and Nintendo and Microsoft have sold roughly equal numbers of consoles, you can see that the Xbox is not selling very well compared to PS2. That means there are 40 million PS2 in people's houses, and ~5 million Xboxes out there. Also, if you head over to NPDFunworld, you'll see that for the last 6 months Xbox has had on average 1 game in the monthly top ten based on sales. That's not anything to write home about, especially when there are typically 6 PS2 games and 3 GBA games on the list.
Maybe someday there will be enough good Xbox exclusive titles out there to get more people to shell out the $200 for an Xbox, but with microsoft already planning on releasing Xbox 2 in just over 2 years, and 90% of the good xbox games being available on other platforms, Xbox sales may not be picking up anytime soon.
I'm assuming you have a fast net connection. If you don't, please disregard...
Download the debian boot-floppies. You can do a full net instalation, and you'll only install what you need instead of downloading the entire contents of some distro's ISOs.
If you're not afraid of trying a development kernel with a beta filesystem patch, then the debian installation process should be simple.
Seriously, though. It looks like this new filesystem patch isn't quite ready. They're still finding leaked blocks and other corruption. You should run a "stable" kernel on your primary box, or you should do frequent backups! My primary workstation is running 2.4.17 (Hasn't had any downtime since 2.4.17 was released, so I haven't upgraded.), and I reserve the 2.5 series kernels for the rack of test machines behind me. If they break it's OK, but if I loose my emacs session I get seriously pissed off.
-- No need for flameproof armor. My home box runs windows most of the time. I hack Linux all day at work, so I want to use my home box for gaming. No better OS for that right now than windows. --
I honestly want to know if it is a huge increase, or a small increase.
As in all things benchmarking related, the answer is: It depends. This will be significant for certain uses of your system, but unimportant for others. If you've got a very busy file server, a news server, or a build machine where you do alot of compilation, this will be very significant. For other tasks you might not even notice.
This is a kernel change, so there won't be any ISOs. Why not just try it now?