Not even close to practical. The magnetic disk manufacturers implemented wear leveling back when the drives were in the 200MB-range. Before that disks wore out even quicker than flash disks and I didn't even use swap-files then. There is a huge difference between unlimited number of writes and undefined number of writes. In critical applications, a bad number is better than an undefined one. At least you can calculate a life-time and design after that.
No, sir. HDDs (at least up until I stopped writing FW for them in 1999) did not have any wear leveling algorithms. In other words, the translation of LBA to physical location on the media (sometimes called Physical Block Address or PBA) is fixed, other than for defective sectors which have been remapped. So if an O.S. wrote to a specific LBA or range of LBAs repeatedly (think paging/swap file or hibernate file), those PBAs would be written to more frequently (or at least at a different rate) than other PBAs across the drive.
especially once USB gets its shit together and provides enough power to run HDDs without an extra power supply.
It already does... for 2.5" drives.
No it doesn't. At least not spinning rust - those things easily require 2+A to start up. On USB alone, that would prevent them from even starting, nevermind running.
I've been using external USB-powered 2.5" HDDs for years, connecting to laptops, desktops and USB hubs. Works just fine in all situations, which is why companies such as WD are still selling these things like hotcakes.
My Micronta alarm clock has worked great for 30 years.
Meanwhile, my kids are on their third set of alarm clocks, due to various failures (in the clocks!). Looks like they don't design clocks to last, any more.
Let's just ban cars. And scissors. How many people died from choking last year? Surely there's some way to prevent those.
And if they don't ban cars, they should at least ban alcohol, because that would save thousands of auto fatalities per year. And guns, too....lots of lives to be saved there. And mosquitos, those little malaria-carrying bastards.
I received the message via Sprint, despite being 400 miles from the affected area. I guess this is one way to make sure people start ignoring these messages.
Yes - I have been, a dozen times or more in actual respected WRITTEN PUBLICATIONS in the art & science of computing (would you like a partial list?)...
I enjoy a good read. Please enlighten me with some of your published works.
It's not the truth or value (or lack of) in your post that gets it modded into oblivion, it's the fucking insane length. In addition to TL;DR (which goes without saying for a post of such length), how about irritating readers by requiring them to scroll through 20+ screenfuls just to get to the next post.
If you want to publish a short story like this, please do everyone a favor and blog it somewhere, then provide a brief summary and link to your blog. Readers intrigued by your summary will go read your blog, and everyone else will just move along at normal/. speed.
Whether or not there is a shortage of native IT workers in Australia, companies could potentially switch to off shoring the jobs if the government prevents importing of workers.
I used to live in a townhouse about 50ft from a 6-lane highway, with no protective soundwall. The noise from the highway was a constant buzz coming into the house, with occasional spikes (big rig trucks and straight-pipe Harleys).
Installing dual-pane windows solved the general problem of the constant higher noise level. Only the truck/Harley noise came through. Note that after a rain, car tires make more noise, and THAT still came through the windows a bit.
In your case, you need more help. Maximize your dead air space. Dual-pane windows will have have a smaller air gap than your walls, so start with the windows. If you need further noise abatement, hit the exterior walls, next.
I later bought a 60-year-old house (which already had dual-pane windows), and during a subsequent remodel we replaced all the half-inch drywall with 5/8th-inch, and added better insulation in the walls.It did a great job of reducing the "garage noise" (power tools, racing engines, etc.) from my neighbor's house.
We did the same for the interior walls, and replaced all the hollow-core interior doors with solid-core doors. Now any teen-chatter or stereo noise from the kids' rooms is effectively muffled into oblivion.
This is due to the fact that when a cell wears out the data doesn't go away. You can still read the data, you just can't write to that cell again.
That is incorrect. A flash cell can outright fail or can simply return an excessive number of bit errors. The SSD firmware decides when to stop using a block based on the likelihood of unsuccessfully retrieving the data in the future.
Something else to keep in mind is that cells degrade even when not being accessed, when neighboring cells are read -- an effect called "read disturb".
Bullshit. They can state all they want. But until those 5 years have passed, and we have actual data on a significant amount of SSDs, it’s all just wild guesswork right out of the marketing department.
Oh, and others are not pretty similar, but much much worse. You know that, because you deliberately picked Intel. The only manufacturer to have the balls to make up numbers that are in the acceptable range. (But they are still made up. Unless they got a time machine.)
There are many types of technologies that are put through accelerated life testing *precisely* for the reason of having engineering test data and not "marketing data" that form the basis of what manufacturers claim for their products. Now of course that doesn't mean that marketing folks won't bend the truth or play with words, but any manufacturer that simply makes up numbers won't be around very long.
Of course accelerated life testing isn't perfect, but this is how manufacturers are able to make reasonable predictions to the useful lifetimes of your CDs, DVDs, Blu-rays, the platters in your hard drives, etc.
Reading has no wear effect on SSDs. Writing does, but it's a very high limit.
Unfortunately reading NAND flash is not "free", i.e. a cell's lifetime is reduced simply by reading *neighboring* cells. The effect is called "read disturb". I'm not an EE, but the explanation makes sense.
"Disturb testing is designed to study the robustness of the data storage of the flash cells when the
state of a nearby cell is being changed, either through programming or reading. A disturb failure
means that the initial (and expected) state of the cell has been changed (disturbed) to the opposite
state as a result of programming or reading the nearby cells. Disturb failures are usually soft
failures that require additional device commands to repair.
Flash manufacturers acknowledge disturb failures can occur on their devices and try to provide
users with guidance on how to address them."
i think you will find them being pcie cards, not 3.5" sata drives, as the drive shape limits chip numbers.
All of the SSD drives that I've seen are 2.5" drives. I've often wondered why no one uses some simple heat sinks and packs the chips as dense as can be into a 3.5" form factor.
I think the limiting factor is the SATA bus. Pack a 3.5-inch drive with NAND and a good controller and you will absolutely swamp a 3Gb SATA bus. When 6Gb SATA becomes mainstream, then you may start seeing 3.5-inch SSDs.
I'm kind of surprised that something as precision-oriented as (what are now) super-capacity hard drives don't have any accounting for this, when if that's the solution, they should have been right up there with fan controls as a reasonable tweak for overclockers.
So you want acceleration and braking controls for your HDD like you have on your car? Well, the best you are likely to get is the manufacturer's dumb tool that lets you config the HDD for quiet vs. fast operation.
The head does not move to a position based on field strength (open loop control). It is free to move on low friction bearings, the applied field strength accelerates the head. Closed loop control is needed to make it stop at the correct position.
The servo firmware engineers I worked with went to great lengths to define and maintain (in real-time) the acceleration profiles for the actuator arm. The point is that the seek algorithm is fully closed loop.
Originally they walked the fine line between all-out performance and reliability, but later they started slowing down the seeks (on 7200 RPM drives) to make them quieter.
Re:google reader is also down for me
on
YouTube Is Down
·
· Score: 2, Funny
Quit routing through servers in China. Or move to California, where YouTube and other google things are still running.
For me, it comes down to the annoyance factor. If the ads on a site are cleanly organized in a way that won't distract me while reading the article, then I'm okay with it. But lots of sites display those seizure-inducing, bright-blinking-scrolling ads. THEY get black-listed.
Slashdot Mirror
Not even close to practical. The magnetic disk manufacturers implemented wear leveling back when the drives were in the 200MB-range. Before that disks wore out even quicker than flash disks and I didn't even use swap-files then.
There is a huge difference between unlimited number of writes and undefined number of writes.
In critical applications, a bad number is better than an undefined one. At least you can calculate a life-time and design after that.
No, sir. HDDs (at least up until I stopped writing FW for them in 1999) did not have any wear leveling algorithms. In other words, the translation of LBA to physical location on the media (sometimes called Physical Block Address or PBA) is fixed, other than for defective sectors which have been remapped. So if an O.S. wrote to a specific LBA or range of LBAs repeatedly (think paging/swap file or hibernate file), those PBAs would be written to more frequently (or at least at a different rate) than other PBAs across the drive.
No it doesn't. At least not spinning rust - those things easily require 2+A to start up. On USB alone, that would prevent them from even starting, nevermind running.
I've been using external USB-powered 2.5" HDDs for years, connecting to laptops, desktops and USB hubs. Works just fine in all situations, which is why companies such as WD are still selling these things like hotcakes.
My Micronta alarm clock has worked great for 30 years.
Meanwhile, my kids are on their third set of alarm clocks, due to various failures (in the clocks!). Looks like they don't design clocks to last, any more.
Let's just ban cars. And scissors. How many people died from choking last year? Surely there's some way to prevent those.
And if they don't ban cars, they should at least ban alcohol, because that would save thousands of auto fatalities per year.
And guns, too....lots of lives to be saved there.
And mosquitos, those little malaria-carrying bastards.
These busy bodies are just thinking too small.
I received the message via Sprint, despite being 400 miles from the affected area. I guess this is one way to make sure people start ignoring these messages.
NOW - As to myself on that very note??
Yes - I have been, a dozen times or more in actual respected WRITTEN PUBLICATIONS in the art & science of computing (would you like a partial list?)...
I enjoy a good read. Please enlighten me with some of your published works.
Good point! :-)
Hey APK,
Protip:
It's not the truth or value (or lack of) in your post that gets it modded into oblivion, it's the fucking insane length. In addition to TL;DR (which goes without saying for a post of such length), how about irritating readers by requiring them to scroll through 20+ screenfuls just to get to the next post.
If you want to publish a short story like this, please do everyone a favor and blog it somewhere, then provide a brief summary and link to your blog. Readers intrigued by your summary will go read your blog, and everyone else will just move along at normal /. speed.
Whether or not there is a shortage of native IT workers in Australia, companies could potentially switch to off shoring the jobs if the government prevents importing of workers.
Well put, sir.
I used to live in a townhouse about 50ft from a 6-lane highway, with no protective soundwall. The noise from the highway was a constant buzz coming into the house, with occasional spikes (big rig trucks and straight-pipe Harleys).
Installing dual-pane windows solved the general problem of the constant higher noise level. Only the truck/Harley noise came through. Note that after a rain, car tires make more noise, and THAT still came through the windows a bit.
In your case, you need more help. Maximize your dead air space. Dual-pane windows will have have a smaller air gap than your walls, so start with the windows. If you need further noise abatement, hit the exterior walls, next.
For windows, something like this:
http://www.soundproofwindows.com/
For walls:
http://www.google.com/search?q=sound+absorbing+insulation/
I later bought a 60-year-old house (which already had dual-pane windows), and during a subsequent remodel we replaced all the half-inch drywall with 5/8th-inch, and added better insulation in the walls.It did a great job of reducing the "garage noise" (power tools, racing engines, etc.) from my neighbor's house.
We did the same for the interior walls, and replaced all the hollow-core interior doors with solid-core doors. Now any teen-chatter or stereo noise from the kids' rooms is effectively muffled into oblivion.
I went with Appleseed.
My routers are Deunan Knute and Briareos.
My servers are Athena, Hitomi, Niki, Tereus and Yoshi
That made me smile. Love that movie!
> Agreed. And, I believe that 34nm is near the best they can do today, in any kind of production.
There are at least two companies manufacturing 25nm parts right now.
Posting to undo fat-fingered mod. Sorry.
This is due to the fact that when a cell wears out the data doesn't go away. You can still read the data, you just can't write to that cell again.
That is incorrect. A flash cell can outright fail or can simply return an excessive number of bit errors. The SSD firmware decides when to stop using a block based on the likelihood of unsuccessfully retrieving the data in the future.
Something else to keep in mind is that cells degrade even when not being accessed, when neighboring cells are read -- an effect called "read disturb".
Bullshit. They can state all they want. But until those 5 years have passed, and we have actual data on a significant amount of SSDs, it’s all just wild guesswork right out of the marketing department.
Oh, and others are not pretty similar, but much much worse. You know that, because you deliberately picked Intel. The only manufacturer to have the balls to make up numbers that are in the acceptable range. (But they are still made up. Unless they got a time machine.)
There are many types of technologies that are put through accelerated life testing *precisely* for the reason of having engineering test data and not "marketing data" that form the basis of what manufacturers claim for their products. Now of course that doesn't mean that marketing folks won't bend the truth or play with words, but any manufacturer that simply makes up numbers won't be around very long.
Of course accelerated life testing isn't perfect, but this is how manufacturers are able to make reasonable predictions to the useful lifetimes of your CDs, DVDs, Blu-rays, the platters in your hard drives, etc.
Reading has no wear effect on SSDs. Writing does, but it's a very high limit.
Unfortunately reading NAND flash is not "free", i.e. a cell's lifetime is reduced simply by reading *neighboring* cells. The effect is called "read disturb". I'm not an EE, but the explanation makes sense.
From a JPL/NASA document: http://trs-new.jpl.nasa.gov/dspace/bitstream/2014/40761/1/08-07.pdf
"Disturb testing is designed to study the robustness of the data storage of the flash cells when the state of a nearby cell is being changed, either through programming or reading. A disturb failure means that the initial (and expected) state of the cell has been changed (disturbed) to the opposite state as a result of programming or reading the nearby cells. Disturb failures are usually soft failures that require additional device commands to repair. Flash manufacturers acknowledge disturb failures can occur on their devices and try to provide users with guidance on how to address them."
When chicks fight on the Internet...
i think you will find them being pcie cards, not 3.5" sata drives, as the drive shape limits chip numbers.
All of the SSD drives that I've seen are 2.5" drives. I've often wondered why no one uses some simple heat sinks and packs the chips as dense as can be into a 3.5" form factor.
I think the limiting factor is the SATA bus. Pack a 3.5-inch drive with NAND and a good controller and you will absolutely swamp a 3Gb SATA bus. When 6Gb SATA becomes mainstream, then you may start seeing 3.5-inch SSDs.
I'm kind of surprised that something as precision-oriented as (what are now) super-capacity hard drives don't have any accounting for this, when if that's the solution, they should have been right up there with fan controls as a reasonable tweak for overclockers.
So you want acceleration and braking controls for your HDD like you have on your car? Well, the best you are likely to get is the manufacturer's dumb tool that lets you config the HDD for quiet vs. fast operation.
The head does not move to a position based on field strength (open loop control). It is free to move on low friction bearings, the applied field strength accelerates the head. Closed loop control is needed to make it stop at the correct position.
The servo firmware engineers I worked with went to great lengths to define and maintain (in real-time) the acceleration profiles for the actuator arm. The point is that the seek algorithm is fully closed loop.
Originally they walked the fine line between all-out performance and reliability, but later they started slowing down the seeks (on 7200 RPM drives) to make them quieter.
Quit routing through servers in China. Or move to California, where YouTube and other google things are still running.
For me, it comes down to the annoyance factor. If the ads on a site are cleanly organized in a way that won't distract me while reading the article, then I'm okay with it. But lots of sites display those seizure-inducing, bright-blinking-scrolling ads. THEY get black-listed.
When I left Quantum a decade ago, they had already begun developing the technology for hard drives to write their own servo tracks.
Of course porn drives tech; how many slashdotters would be outside, away from their computers were it not for porn?