Two guys above have already pointed out issues with TRIM but I will reiterate; supporting such by Apple would be a waste of time. As has been already mentioned TRIM support in SSD is a stop gap measure. Instead people should be demanding that manufactures fix their SSD.
It is unfortunate that this public and uninformed demand is likely pressuring Apple to support features not really needed in it's OS.
Dave
I think TRIM is great. There are only two options: TRIM, or extra space that can't be addressed on the SSD. You can use the extra space to instantly remap a block when it is written to, and the old block is then erased later. Or, you can have TRIM, be able to use 100% of your space, the the OS tells the SSD what blocks are no longer needed. Plus, you've got to have the write leveling, so that you don't repeatedly write the same block, so one of these two things has to happen.
If TRIM isn't the ideal solution, just what is? The only answer I can figure is SSD that doesn't wear out, and SSD that can read/erase/write just as fast as it reads.
Quote:
Originally Posted by cvaldes
SSD manufacturers should be implementing the garbage collection, not the operating system. TRIM isn't a universal standard, it's a hack.
The only way an SSD can garbage collect is to be able to read the file system. Otherwise, how can it tell when a block is no longer in use, without TRIM and without a reserved space for block remapping?
Quote:
Originally Posted by ontheinside
God do none of you appleinsiders look up facts before stating complete nonsense. TRIM is a hack? TRIM isn't standard? TRIM is software? lol. WTF!?!
Maybe you have a beef with your manufacturer, but my OWC Extreme SSD already handles the issue internally.
Perhaps you just picked the wrong SSD
Not me, because I don't really think SSD are ready for my needs though I'm very very tempted. Especially considering the massive strides made in thee last few months, the drives are really impressive of late but still lag space wise.
Quote:
indeed. TRIM is an ugly hack. What is really needed is a clean drive interface that has none of the ATA baggage that has been drug forward from the first PC days
Yeah like a PCI Express card format specific for storage.
What really bugs me is to see manufactures stress the design of the metal casing for their drives. Who really cares it is just a PC board.
Quote:
If anyone has the sales volume and influence to do it, its Apple. Hopefully it's cooking away in their labs. Hard drives and file system performance are the slowest link in the computer chain by a wide margin. Here's to hoping for a native designed for flash filesystem that goes beyond ZFS.
Yeah I was really hoping that Apple would have shown the world that it had some balls and defined a new SSD format for its laptops. Done right we could have some really thin laptops with multiple "slots" for SSD.
The threads above mine described garbage collection in terms that sounded like the SSD just "knows" what the user and file system have discarded. That doesn't happen with HFS.
It's more a function of OS support...
Quote:
My query was to ask them to think through how the SSD "knows". I don't want an SSD that has to have firmware revs every time an vendor revs/fixes the file system software.
Neither TRIM or disks with extra space for internal garbage collection introduces a firmware dependancy. Not sure where you are going with that...
Quote:
Reserve space is a nice temp fix because it'll be used as an alternative to writing over a used block. The SSD will eventually run out of that too.
If it's reserve space, by definition then it can't fill up, right? Otherwise, why would it be called reserve space?
Let's keep it simple. Let's leave out file systems and size and just talk about individual memory cells - because without TRIM that's all an SSD controller knows or cares about.
If I have a disk with an advertised capacity that just happens to work out to 100 memory cells, an I have another 20 cells in reserve, I will always have the aggregate equivalent of twenty free cells. Always - whatever OS and file system is running on the drive simply doesn't even know those cells exist.
So, when the number of totally empty cells passes some pre-defined point, the firmware analyzes all the cells, probably picks the cells with the least amount of data in them, consolidates them into one free cell and then erases the previously partially filled cells and boom - a net gain of free, ready to write at full speed cells.
Kind of a specialized de-frag, but to optimize free space.
With this model, the drive neither knows nor cares about what is being written to it - it just does it's thing in the background, ensuring as many totally empty cells are "on hand" as possible.
Now theoretically, if your drive isn't full and you write lots of big files, TRIM can allow the SSD to optimize more free cells, but as I said I would much rather see a new filesystem and interface developed that allows the OS t more effectively use "disk" - ESP. when the disk isn't disk at all but something totally different with totally different usage and optimization patterns!
I mean hard drives haven't mapped heads/cylinders 1:1 for over 20 years! Isn't it time to move on?
What 1TB internal are you looking into? I can't find a 7200 RPM version that's tested in a MacBook. I guess I could settle for the 5400RPM version since my system volume is an SSD.
Neither TRIM or disks with extra space for internal garbage collection introduces a firmware dependancy. Not sure where you are going with that...
If it's reserve space, by definition then it can't fill up, right? Otherwise, why would it be called reserve space?
Let's keep it simple. Let's leave out file systems and size and just talk about individual memory cells - because without TRIM that's all an SSD controller knows or cares about.
If I have a disk with an advertised capacity that just happens to work out to 100 memory cells, an I have another 20 cells in reserve, I will always have the aggregate equivalent of twenty free cells. Always - whatever OS and file system is running on the drive simply doesn't even know those cells exist.
So, when the number of totally empty cells passes some pre-defined point, the firmware analyzes all the cells, probably picks the cells with the least amount of data in them, consolidates them into one free cell and then erases the previously partially filled cells and boom - a net gain of free, ready to write at full speed cells.
Kind of a specialized de-frag, but to optimize free space.
With this model, the drive neither knows nor cares about what is being written to it - it just does it's thing in the background, ensuring as many totally empty cells are "on hand" as possible.
Now theoretically, if your drive isn't full and you write lots of big files, TRIM can allow the SSD to optimize more free cells, but as I said I would much rather see a new filesystem and interface developed that allows the OS t more effectively use "disk" - ESP. when the disk isn't disk at all but something totally different with totally different usage and optimization patterns!
I mean hard drives haven't mapped heads/cylinders 1:1 for over 20 years! Isn't it time to move on?
huh? Read from the beginning and don't compare magnetic to flash. I'm never saying they are the same.
I'm surprised Apple don't have Flash as drive-caches in all their mobile products. With all but heavy file-writing using only a few GBs I'd have though including 8 or 16GB flash as standard would have brought worthwhile performance & battery-life advantages.
Yaeh, except Apple already sucks up an ungodly amount of the worlds flash production - I seem to recall it being over 50%?
And anyway, if they did do that, they could just bake support for it into the OS. Kind of like MS did with Vista - and we know how successful that was
And TRIM has nothing to do with that anyway. TRIM is just a way for the OS to tell the drive what is really in use vs free to be deleted.
How about a new drive interface and filesystem optimized for flash, rather than pretending flash is normal rotating disk?
The whole SATA/TRIM thing is such a fiasco because they are trying to shoehorn flash into a fundamentally different model.
Oh well, as I said I can only hope Apple is cooking up something in their labs. They hired the guy from Be who's name I can't remember - he's pretty brilliant and it's obvious they aren't modernizing HFS, so what are they up to? I can only hope...
Two guys above have already pointed out issues with TRIM but I will reiterate; supporting such by Apple would be a waste of time. As has been already mentioned TRIM support in SSD is a stop gap measure. Instead people should be demanding that manufactures fix their SSD.
It is unfortunate that this public and uninformed demand is likely pressuring Apple to support features not really needed in it's OS.
I'm not sure how an SSD can do an optimal job of wear-leveling without the OS telling it when a block is to be considered freed.
What is the current favored system to replace the necessity of TRIM?
What is the current favored system to replace the necessity of TRIM?
Under the current masquerade of trying to making flash look like rotating disk, there isn't any.
We need a new model where the OS can be more aware of what is going on at a low level - and TRIM isn't the answer, it's just a hack to close the gap and let SSD be somewhat efficient in an ATA world.
Under the current masquerade of trying to making flash look like rotating disk, there isn't any.
We need a new model where the OS can be more aware of what is going on at a low level - and TRIM isn't the answer, it's just a hack to close the gap and let SSD be somewhat efficient in an ATA world.
Then to restate my question so that maybe it gets answered, exactly what is that model? Is there no standard being worked on that does what you describe? I can't find one. If there isn't, then the wait is going to take years, because making standards takes years, so the short term is either have TRIM or not. Something better probably isn't an option for a few years, so you'll have to deal with the indignity of having a hack for a while.
Yes it fits! But it does not work in the just released models! I am not sure if it is something with power or firmware but the newest i7 unibodies can not boot from this drive. Disk Utility can see it in the installer, but will not install to it. Installing onto it as an external from another computer works, but then it still can not boot in the i7 machine
I had the drive on order and someone else I know during the OptiBay found an article stating that there is an issue with the 1TB drives and SATA controllers in the mid-2010 MBPs. However, the 15" and 17" use a new intel SATA controller and the 13" uses a new Nvidia controller and of ource are Core-i systems. It sounds like this may be an issue with those machines, but I have the mid-2010 13" MBP so it's likely I'd be fine.
Fortunately, this space crunch allowed me to clean up my data files drastically (and I still haven't to iTunes yet) so the 500GB @7200RPM drive now has 60GB remaining and should suit me for several months. I'll likely not need to upgrade the drive before I buy a new MBP next year.
What 1TB internal are you looking into? I can't find a 7200 RPM version that's tested in a MacBook. I guess I could settle for the 5400RPM version since my system volume is an SSD.
I had ordered the WD 1TB HDD which is only 5400RPM, as far as I know.
Wait....the worlds most advanced OS doesn't support this already?
SSD manufacturer is ultimately responsible for the TRIM implementations and it only being applied many months after the first batch of SSD drives came out i.e. just recently (<2 years). I doubt the first generation of Macs that optioned SSD have the built-in support for TRIM at driver level hence no point for the OS supporting it. Once the driver is there at hardware level, only then the OS can support it. Much like only recently they updated the controller with the much better SandForce rather than the old Intel or Marvel controller.
Seems 10.6.4 shows the TRIM section in System Profiler on my Mac Pro 1,1. Was never there before, however my OCZ Vertex Turbo SSD (which does have TRIM support) is listed as "TRIM Support: No"...
With the garbage collection built into later model drives/firmware, TRIM isn't necessary for great SSD performance in real-world usage.
If you're waiting for TRIM support in Mac OS X before upgrading to SSD, you're wasting your time.
+1
TRIM support in the OS doesn't mean much when you have good GC/TRIM right on the SSD itself and because the SSD knows the controller it would probably be superior to TRIM in an OS.
TRIM support in the OS doesn't mean much when you have good GC/TRIM right on the SSD itself and because the SSD knows the controller it would probably be superior to TRIM in an OS.
TRIM support in the OS doesn't mean much when you have good GC/TRIM right on the SSD itself and because the SSD knows the controller it would probably be superior to TRIM in an OS.
Read the thread. Garbage collection doesn't work unless the controller on the SSD understands the filesystem. The controller can't reset blocks when it doesn't understand which are in use and which aren't.
Read the thread. Garbage collection doesn't work unless the controller on the SSD understands the filesystem. The controller can't reset blocks when it doesn't understand which are in use and which aren't.
If the SSD has extra capacity to perform garbage collection in, knowledge of what is going on at the file system level isn't required.
That's what OWC's Mercury Extreme drive does, and I can verify that it works as I have one in my Mac Pro.
Comments
Two guys above have already pointed out issues with TRIM but I will reiterate; supporting such by Apple would be a waste of time. As has been already mentioned TRIM support in SSD is a stop gap measure. Instead people should be demanding that manufactures fix their SSD.
It is unfortunate that this public and uninformed demand is likely pressuring Apple to support features not really needed in it's OS.
Dave
I think TRIM is great. There are only two options: TRIM, or extra space that can't be addressed on the SSD. You can use the extra space to instantly remap a block when it is written to, and the old block is then erased later. Or, you can have TRIM, be able to use 100% of your space, the the OS tells the SSD what blocks are no longer needed. Plus, you've got to have the write leveling, so that you don't repeatedly write the same block, so one of these two things has to happen.
If TRIM isn't the ideal solution, just what is? The only answer I can figure is SSD that doesn't wear out, and SSD that can read/erase/write just as fast as it reads.
SSD manufacturers should be implementing the garbage collection, not the operating system. TRIM isn't a universal standard, it's a hack.
The only way an SSD can garbage collect is to be able to read the file system. Otherwise, how can it tell when a block is no longer in use, without TRIM and without a reserved space for block remapping?
God do none of you appleinsiders look up facts before stating complete nonsense. TRIM is a hack? TRIM isn't standard? TRIM is software? lol. WTF!?!
You must be new to Apple.
Maybe you have a beef with your manufacturer, but my OWC Extreme SSD already handles the issue internally.
Perhaps you just picked the wrong SSD
Not me, because I don't really think SSD are ready for my needs though I'm very very tempted. Especially considering the massive strides made in thee last few months, the drives are really impressive of late but still lag space wise.
indeed. TRIM is an ugly hack. What is really needed is a clean drive interface that has none of the ATA baggage that has been drug forward from the first PC days
Yeah like a PCI Express card format specific for storage.
What really bugs me is to see manufactures stress the design of the metal casing for their drives. Who really cares it is just a PC board.
If anyone has the sales volume and influence to do it, its Apple. Hopefully it's cooking away in their labs. Hard drives and file system performance are the slowest link in the computer chain by a wide margin. Here's to hoping for a native designed for flash filesystem that goes beyond ZFS.
Yeah I was really hoping that Apple would have shown the world that it had some balls and defined a new SSD format for its laptops. Done right we could have some really thin laptops with multiple "slots" for SSD.
Maybe they will wise up with the AIR revision.
Dave
The threads above mine described garbage collection in terms that sounded like the SSD just "knows" what the user and file system have discarded. That doesn't happen with HFS.
It's more a function of OS support...
My query was to ask them to think through how the SSD "knows". I don't want an SSD that has to have firmware revs every time an vendor revs/fixes the file system software.
Neither TRIM or disks with extra space for internal garbage collection introduces a firmware dependancy. Not sure where you are going with that...
Reserve space is a nice temp fix because it'll be used as an alternative to writing over a used block. The SSD will eventually run out of that too.
If it's reserve space, by definition then it can't fill up, right? Otherwise, why would it be called reserve space?
Let's keep it simple. Let's leave out file systems and size and just talk about individual memory cells - because without TRIM that's all an SSD controller knows or cares about.
If I have a disk with an advertised capacity that just happens to work out to 100 memory cells, an I have another 20 cells in reserve, I will always have the aggregate equivalent of twenty free cells. Always - whatever OS and file system is running on the drive simply doesn't even know those cells exist.
So, when the number of totally empty cells passes some pre-defined point, the firmware analyzes all the cells, probably picks the cells with the least amount of data in them, consolidates them into one free cell and then erases the previously partially filled cells and boom - a net gain of free, ready to write at full speed cells.
Kind of a specialized de-frag, but to optimize free space.
With this model, the drive neither knows nor cares about what is being written to it - it just does it's thing in the background, ensuring as many totally empty cells are "on hand" as possible.
Now theoretically, if your drive isn't full and you write lots of big files, TRIM can allow the SSD to optimize more free cells, but as I said I would much rather see a new filesystem and interface developed that allows the OS t more effectively use "disk" - ESP. when the disk isn't disk at all but something totally different with totally different usage and optimization patterns!
I mean hard drives haven't mapped heads/cylinders 1:1 for over 20 years! Isn't it time to move on?
What 1TB internal are you looking into? I can't find a 7200 RPM version that's tested in a MacBook. I guess I could settle for the 5400RPM version since my system volume is an SSD.
It's more a function of OS support...
Neither TRIM or disks with extra space for internal garbage collection introduces a firmware dependancy. Not sure where you are going with that...
If it's reserve space, by definition then it can't fill up, right? Otherwise, why would it be called reserve space?
Let's keep it simple. Let's leave out file systems and size and just talk about individual memory cells - because without TRIM that's all an SSD controller knows or cares about.
If I have a disk with an advertised capacity that just happens to work out to 100 memory cells, an I have another 20 cells in reserve, I will always have the aggregate equivalent of twenty free cells. Always - whatever OS and file system is running on the drive simply doesn't even know those cells exist.
So, when the number of totally empty cells passes some pre-defined point, the firmware analyzes all the cells, probably picks the cells with the least amount of data in them, consolidates them into one free cell and then erases the previously partially filled cells and boom - a net gain of free, ready to write at full speed cells.
Kind of a specialized de-frag, but to optimize free space.
With this model, the drive neither knows nor cares about what is being written to it - it just does it's thing in the background, ensuring as many totally empty cells are "on hand" as possible.
Now theoretically, if your drive isn't full and you write lots of big files, TRIM can allow the SSD to optimize more free cells, but as I said I would much rather see a new filesystem and interface developed that allows the OS t more effectively use "disk" - ESP. when the disk isn't disk at all but something totally different with totally different usage and optimization patterns!
I mean hard drives haven't mapped heads/cylinders 1:1 for over 20 years! Isn't it time to move on?
huh? Read from the beginning and don't compare magnetic to flash. I'm never saying they are the same.
I'm surprised Apple don't have Flash as drive-caches in all their mobile products. With all but heavy file-writing using only a few GBs I'd have though including 8 or 16GB flash as standard would have brought worthwhile performance & battery-life advantages.
Yaeh, except Apple already sucks up an ungodly amount of the worlds flash production - I seem to recall it being over 50%?
And anyway, if they did do that, they could just bake support for it into the OS. Kind of like MS did with Vista - and we know how successful that was
And TRIM has nothing to do with that anyway. TRIM is just a way for the OS to tell the drive what is really in use vs free to be deleted.
If TRIM isn't the ideal solution, just what is?
How about a new drive interface and filesystem optimized for flash, rather than pretending flash is normal rotating disk?
The whole SATA/TRIM thing is such a fiasco because they are trying to shoehorn flash into a fundamentally different model.
Oh well, as I said I can only hope Apple is cooking up something in their labs. They hired the guy from Be who's name I can't remember - he's pretty brilliant and it's obvious they aren't modernizing HFS, so what are they up to? I can only hope...
Read from the beginning and don't compare magnetic to flash. I'm never saying they are the same.
Neither did I. Ignore the last line and re-read.
huh?
Huh indeed...
Maybe they will wise up with the AIR revision.
I don't know about wising up, but the air would be a perfect machine to do something new with. You might be on to something.
Two guys above have already pointed out issues with TRIM but I will reiterate; supporting such by Apple would be a waste of time. As has been already mentioned TRIM support in SSD is a stop gap measure. Instead people should be demanding that manufactures fix their SSD.
It is unfortunate that this public and uninformed demand is likely pressuring Apple to support features not really needed in it's OS.
I'm not sure how an SSD can do an optimal job of wear-leveling without the OS telling it when a block is to be considered freed.
What is the current favored system to replace the necessity of TRIM?
What is the current favored system to replace the necessity of TRIM?
Under the current masquerade of trying to making flash look like rotating disk, there isn't any.
We need a new model where the OS can be more aware of what is going on at a low level - and TRIM isn't the answer, it's just a hack to close the gap and let SSD be somewhat efficient in an ATA world.
Under the current masquerade of trying to making flash look like rotating disk, there isn't any.
We need a new model where the OS can be more aware of what is going on at a low level - and TRIM isn't the answer, it's just a hack to close the gap and let SSD be somewhat efficient in an ATA world.
Then to restate my question so that maybe it gets answered, exactly what is that model? Is there no standard being worked on that does what you describe? I can't find one. If there isn't, then the wait is going to take years, because making standards takes years, so the short term is either have TRIM or not. Something better probably isn't an option for a few years, so you'll have to deal with the indignity of having a hack for a while.
What is this incompatibility you speak of??
Working fine here (2x1tb WD drives in a macbook unibody
I'm having some trouble finding the exact article that made me hesitant but here is a post from Apple's forums that is a clue to the potential issue.
Posted: Apr 23, 2010 11:22 AM\t \t
Yes it fits! But it does not work in the just released models! I am not sure if it is something with power or firmware but the newest i7 unibodies can not boot from this drive. Disk Utility can see it in the installer, but will not install to it. Installing onto it as an external from another computer works, but then it still can not boot in the i7 machine
I had the drive on order and someone else I know during the OptiBay found an article stating that there is an issue with the 1TB drives and SATA controllers in the mid-2010 MBPs. However, the 15" and 17" use a new intel SATA controller and the 13" uses a new Nvidia controller and of ource are Core-i systems. It sounds like this may be an issue with those machines, but I have the mid-2010 13" MBP so it's likely I'd be fine.
Fortunately, this space crunch allowed me to clean up my data files drastically (and I still haven't to iTunes yet) so the 500GB @7200RPM drive now has 60GB remaining and should suit me for several months. I'll likely not need to upgrade the drive before I buy a new MBP next year.
@solipsism
What 1TB internal are you looking into? I can't find a 7200 RPM version that's tested in a MacBook. I guess I could settle for the 5400RPM version since my system volume is an SSD.
I had ordered the WD 1TB HDD which is only 5400RPM, as far as I know.
Wait....the worlds most advanced OS doesn't support this already?
SSD manufacturer is ultimately responsible for the TRIM implementations and it only being applied many months after the first batch of SSD drives came out i.e. just recently (<2 years). I doubt the first generation of Macs that optioned SSD have the built-in support for TRIM at driver level hence no point for the OS supporting it. Once the driver is there at hardware level, only then the OS can support it. Much like only recently they updated the controller with the much better SandForce rather than the old Intel or Marvel controller.
With the garbage collection built into later model drives/firmware, TRIM isn't necessary for great SSD performance in real-world usage.
If you're waiting for TRIM support in Mac OS X before upgrading to SSD, you're wasting your time.
+1
TRIM support in the OS doesn't mean much when you have good GC/TRIM right on the SSD itself and because the SSD knows the controller it would probably be superior to TRIM in an OS.
TRIM support in the OS doesn't mean much when you have good GC/TRIM right on the SSD itself and because the SSD knows the controller it would probably be superior to TRIM in an OS.
I'll take TRIM wherever I can get it.
+1
TRIM support in the OS doesn't mean much when you have good GC/TRIM right on the SSD itself and because the SSD knows the controller it would probably be superior to TRIM in an OS.
Read the thread. Garbage collection doesn't work unless the controller on the SSD understands the filesystem. The controller can't reset blocks when it doesn't understand which are in use and which aren't.
Read the thread. Garbage collection doesn't work unless the controller on the SSD understands the filesystem. The controller can't reset blocks when it doesn't understand which are in use and which aren't.
If the SSD has extra capacity to perform garbage collection in, knowledge of what is going on at the file system level isn't required.
That's what OWC's Mercury Extreme drive does, and I can verify that it works as I have one in my Mac Pro.
But don't take my word for it:
http://macperformanceguide.com/SSD-R...evereDuty.html
No TRIM needed!