SSHDs Debut On the Desktop With Mixed Results 154
crookedvulture writes "Seagate's solid-state hybrid drives have finally made it to the desktop. The latest generation of SSHDs debuted with a 2.5" notebook model that was ultimately hampered by its slow 5,400-RPM spindle speed. The Desktop SSHD has the same 8GB flash payload and Adaptive Memory caching scheme. However, it's equipped with 2TB of much faster 7,200-RPM mechanical storage. The onboard flash produces boot and load times only a little bit slower than those of full-blown SSDs. It also delivers quicker response times than traditional hard drives. That said, the relatively small cache is overwhelmed by some benchmarks, and its mechanical sidekick isn't as fast as the best traditional hard drives. The price premium is a little high, too: an extra $30 for the 1TB model and $40 for the 2TB variant, which is nearly enough to buy a separate 32GB SSD. Seagate's software-independent caching system works with any operating system and hardware platform, so it definitely has some appeal. But dual-drive setups are probably the better solution for most desktop users."
oops (Score:4, Insightful)
So Seagate decides to take the biggest pitfall and hated feature and put it into a hybrid drive. All data written to the gigantic drive is passed through that 8GB buffer first. Flash memory that can put up with that amount of writes over the long term doesn't exist. These drives would maybe last a year or two if you're lucky.
Re: (Score:2, Insightful)
All data written to the gigantic drive is passed through that 8GB buffer first.
The article says,
the SLC zones store boot data and cache some incoming writes.
One of you is wrong.
Re: (Score:1)
I vote for all three being wrong.
Re: (Score:2)
Re:oops (Score:5, Informative)
It is not a write through cache. The drive firmware copies frequently read files to flash. I use a Momentus XT and can you can actually notice when it does this. Frequently used programs load quickly. If you update a program, it loads slowly a time or two, then suddenly switches to loading fast.
Bittorrents screw all of this up. Frequent reads lead to more and more programs being displaced. If I leave bittorrent running over night, it takes a day or two for the flash to repopulate with the OS and programs.
Re: (Score:2)
It is not a write through cache. The drive firmware copies frequently read files to flash.
The problem with that claim is that it doesn't jive with it being OS-agnostic. To know what a file are, you have to understand the file system. I can guarantee you that this drive does not understand XFS with external journal, which is what I use.
If you mean frequently read blocks, that's doable, but to have a counter for every block of a 2TB drive would take up far more memory than this device has.
What is feasible is a caching system which expires blocks that haven't been read in a certain amount of time
Re: (Score:2)
It is not a write through cache. The drive firmware copies frequently read files to flash.
The problem with that claim is that it doesn't jive with it being OS-agnostic. To know what a file are, you have to understand the file system. I can guarantee you that this drive does not understand XFS with external journal, which is what I use.
it doesnt care, it caches SECTORS, not files
Re: (Score:2)
It is not a write through cache. The drive firmware copies frequently read files to flash.
The problem with that claim is that it doesn't jive with it being OS-agnostic. To know what a file are, you have to understand the file system. I can guarantee you that this drive does not understand XFS with external journal, which is what I use.
If you mean frequently read blocks, that's doable, but to have a counter for every block of a 2TB drive would take up far more memory than this device has.
What is feasible is a caching system which expires blocks that haven't been read in a certain amount of time. But that would contradict the claim that it boosts boot speed, because those blocks are generally only read once, at boot time, and would get expunged.
Why wouldn't it just cache the first 8GB of blocks read after power on? That should cache the O/S startup files and whatever applications are autostarted after boot.
Re: (Score:2)
It is not a write through cache. The drive firmware copies frequently read files to flash.
The problem with that claim is that it doesn't jive with it being OS-agnostic. To know what a file are, you have to understand the file system. I can guarantee you that this drive does not understand XFS with external journal, which is what I use.
If you mean frequently read blocks, that's doable, but to have a counter for every block of a 2TB drive would take up far more memory than this device has.
What is feasible is a caching system which expires blocks that haven't been read in a certain amount of time. But that would contradict the claim that it boosts boot speed, because those blocks are generally only read once, at boot time, and would get expunged.
It doesn't have to track every block on the drive, only those it has seen. And it wouldn't need a counter to each block, a simple seen/not-seen is enough. The controller could use an algorithm like:
1. Block first seen - Mark the block address in some list of seen blocks.
2. Block seen list full - Drop the LRU seen block from the list.
3. Block seen again - If the block is in the list of seen blocks, migrate the block contents to the cache.
4. Not enough cache space, evict the LRU block in the cache.
The block s
Re: (Score:2)
"Bittorrents screw all of this up. Frequent reads lead to more and more programs being displaced. If I leave bittorrent running over night, it takes a day or two for the flash to repopulate with the OS and programs."
This is an excellent ILLUSTRATION of how Seagate's design decisions were pretty obviously bonehead.
Their Flash research division convinces the bigwigs to put all their money behind the technology... but it's not really quite ready for prime time. Meanwhile, their HD operations -- which USED TO be just fine -- have shut down anything over 5,400 rpm.
Re: (Score:2)
It's not because the OS file-sectors get swapped out and get replaced by the torrent-files (that are obviously more often accessed and hence better served from the cache) that they are now SLOWER than as if there was no flash-cache at all. If some bytes are not cached they are simply read from disk like it would be with a 'normal' drive.
It probably would be possible to add something to the I/O protocol alike TRIM that would tell the SSHD to not-cache certain reads, but I don't know if they have enough momen
Re: (Score:2)
Re: (Score:2)
I thought they had these on the market for a while now though...
and yes, they'll fool boot time stats on sequential boots, you know, very useful for all those times you're doing sequential booting! otherwise they're no match for ssd and it's just seagate being cheatgate again.
Re: (Score:2)
By the way, I have 16Gb on my machine and I do a lot of content creation. I *easily* fill it up to 14Gb+, and I'm guessing that it only stops there bec
Just what problem is this trying to solve? (Score:2)
Flash solves the problem of waiting for verified writes. That's why putting database logs onto battery-backed high speed memory improves performance dramatically.
That said, do desktops really need that kind of performance boost. Unless you're doing some serious data creation, the only thing that's slow on today's desktops is booting Windows. Linux boots so fast, I wouldn't bother worrying about throwing hardware at that 'problem', and Windows isn't even that bad these days. Whether ChromeOS and it's lik
Re: (Score:2)
For the rest of us (developers, system administrators, content creators) SSD speed helps out immensely.
Re: (Score:2)
Assuming your OS caches data effectively, there's not much benefit to extra fast I/O access to what is essentially another cache. i suppose if frequently accessed data is kept in that cache across boots, that might speed certain things up. But then again, if that's really important, the OS could implement pre-fetch logic to re-load its cache with that same data in the background on boot. So, I repeat, non-volitile cache is a lifesaver when guaranteed immediate writes are required. But very little of wha
Windows already does that (Score:2)
Because Windows already takes care of this with SuperFetch. After you load the OS it immediately caches the applications you use most often into available ram, and removes them when you actually USE that ram. The cached applications load instantly, and the hard drive is none-the-wiser.
In fact, there used to be a grand push to put as much DDR cache in hard
Re: (Score:2)
It's apparently only 'some' writes cached, and one wonders how exactly that is done while still being OS agnostic. But not much.
This is just a crappy drive with an expensive cache, which may or may not die more quickly than normal case, whose only real advantage is persistence. Which means it will do little more than make your system boot faster. If you are booting often enough to worry about it you are doing it wrong IMHO. I'd rather spend the money on a faster spindle or a better conventional cache (or bo
Re: (Score:2)
"You write/read to sectors mapped as blocks, if you read/write to the same block more then a few times then it's worth caching. That's how block caches work."
And that is different from how a conventional cache works how?
Exactly how I said. Persistence. So it helps with boot-times where a standard cache is blank at startup. Once the machine has been up for a minute? No advantage. Probably a disadvantage compared to a cache where there is no need to worry about exceeding maximum writes.
you can't cache writes in RAM, locality of random (Score:2)
For READS, yes it's mainly about boot time and the first time you open a frequently used program. WRITES on the other hand can't be cached to RAM, not for more than a few seconds (and some not at all). Persistent cache makes all the difference for random writes.
The benchmarks understate the improvement because they generally lack locality and frequent rewrites. Benchmarks typically spread writes all over a 2 TB drive. In actual use, random writes aren't truly random. A database will write to the same xx
Re: (Score:2)
"For READS, yes it's mainly about boot time and the first time you open a frequently used program. WRITES on the other hand can't be cached to RAM, not for more than a few seconds (and some not at all). Persistent cache makes all the difference for random writes."
You generally only need to cache writes for a few fractions of a second, until the write head is in position to transfer them to disk. Dynamic RAM is fine for this, and using a persistent medium is not going to magically improve your results here.
"
Re: (Score:2)
The major difference is that the cache from the SSHD is persistent while your RAM isn't. For your OS to buffer a certain file it still needs to read it a first time from the disk. E.g. when starting up or rebooting this helps you nothing at all. The SSHD (might) have said file in cache and hence can serve it much faster.
I'm not sure how you come to the conclusion that your system cache "knows a lot more about access patterns and files than the on-drive logic can possibly gather.".
Given the way HD's work the
Re: (Score:2)
"The major difference is that the cache from the SSHD is persistent while your RAM isn't. For your OS to buffer a certain file it still needs to read it a first time from the disk. E.g. when starting up or rebooting this helps you nothing at all. The SSHD (might) have said file in cache and hence can serve it much faster."
That's what I have been saying over and over. This will give you persistence which should mean a faster boot, but is of little or no utility outside of that, when compared to the alternati
Re: (Score:2)
Off course these disks do not know anything about the filesystem at all; all they do is look at the raw throughput. There is something to be said about the OS knowing which files are worth caching and which not; but in the end the cache on the SSHD will come to the same conclusion as it simply keeps track about which blocks of data are most often requested. Plus, having an SSHD doesn't magically sabotage your system cache.
Like I said, you may find all sorts of reasons to not like it, in practice the thing d
not actually write caching blocks copied to flash (Score:5, Interesting)
I looked at buying one of these. Writes don't really go to flash. Selected blocks are asynchronously copied to flash.
There's cool way to avoid the cash over use you mention that I wish someone would make in an under $500 drive. Have 4GB of flash, 4GB of DRAM, and a capacitor. Random writes go to DRAM, making random io a thousand times faster. On power failure, the capacitor flashes the contents of the DRAM to the flash. You get the speed of DRAM, crash safety, and 3TB of capacity from the underlying spindle.
Re: (Score:2)
Well, in that case you could actually just skip the flash entirely.
The RAM could write directly to a dedicated area on the hard disk in the event of a power failure - sort of like how hibernate / suspend to disk works now.
you'd need a BIG capacitor if no Flash. (Score:5, Informative)
with actual real world write speeds of around 20 MB/s, that capacitor would need to spin the drive for three minutes. That would be one hell of a capacitor. Flash chips use less power and are faster, so they could run long enough on capacitors that actually exist.
.
I was wrong. This gen caches some writes to SLC (Score:2)
After reading TFA, I was wrong, that was the previous model. This model does cache some writes in an area of the flash operated as SLC.
Re: (Score:2)
its called OS cache, just allocate 4GB of ram to caching and be done with it
good luck with power outages tho
that's why the capacitor to copy ram to flash (Score:2)
Power loss is the reason for the capacitor to copy the RAM to flash, to basically make persistent DRAM cache.
test results? 512MB battery backed unimpressive (Score:2)
Do you happen to have any real world comparisons between these newer and some other options. The 512 MB of battery backed RAM in my 3ware 95xx SHOULD be awesome for some workloads where it's not. It's better than turning off the cache, but it's not what I was hoping for.
Since mdadm software raid is much faster than the 3ware for my raid 5, I'm hesitant to spend $$$ on a high end raid card that likely will be slower and less compatible than modern software raid. When bcache is in the vendor kernels THAT l
Re: (Score:2)
These drives would maybe last a year or two if you're lucky.
So, ten times longer than the average Seagate drive, then? That's actually pretty good for them! Eh, it's a joke, but based on reality. I've had so much bad luck with Seagate since they bought Maxtor, I don't buy their products anymore, ever.
Re: (Score:2)
I have plenty of Seagate drives.
I usually retire them for being too small before they actually fail. Although I do have a few older ones that I've kept around because they continue to chug along and simply haven't generated any SMART warnings yet.
It will be about 5 years before any of my SSDs or Hybrid drives have been in service as long. It will be awhile before ANYONE can say that actually.
That's the problem with the newest shiny shiny. No track record yet.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Forget endurance failures, think more about firmware blowouts. The latter are far more likely and will take all your data with it, while leaving the drive for warranty purposes and statistics functional. Premature SMART errors are on thing on traditional hard drives due to bad firmware, but this trend with SSDs of firmware issues hosing all data is the real concern.
Jason.
Re: (Score:2)
"So Seagate decides to take the biggest pitfall and hated feature and put it into a hybrid drive. All data written to the gigantic drive is passed through that 8GB buffer first. Flash memory that can put up with that amount of writes over the long term doesn't exist. These drives would maybe last a year or two if you're lucky."
That's only half the problem. Seagate made a SECOND really poor decision, when it decided to dump the manufacture of spinning platters over 5400 RPM. They were TOLD that was a bad idea, yet they did it anyway, and look at the results: their very FiRST generation of new drives can't keep up. And what about the future?
Come on, Seagate. That's TWO MAJOR dumbass moves in a row.
Re: (Score:2)
When I Was in the market for and SSD, as I recall they had a really rocky start, but have of late gotten their shit together and put out decent quality products.
Re: (Score:2)
Yeah, I have a 128G Vertex 4 and have been abusing it (recording 4/8 channel streams with ardour, running three or four pretty active VMs, using it for the swap partition, ...) for about a year and the wear indicator is at 97% life remaining... AFAICT, the main issues were horrible firmware bugs that were resolved eventually, but not until a whole lot of people experienced data loss. It might help that I have a daily cron job to run TRIM (shame it can't be enabled by default for online use since the command
Re: (Score:2)
"note: "more write cycles" not "faster""
Doesn't matter. MLC only reduced the # of write cycles by a few times, not "orders of magnitude". That's still an exaggeration.
Huh? (Score:5, Informative)
I've had one in my desktop for a couple years now.
Re: (Score:3)
These aren't yer momma's Momentus drives. These are NEW and ADAPTIVE!
Re: (Score:2)
Re: (Score:3)
What? I still use my first generation Momentus XT (500GB with the 4GB of SLC), never had any BSOD after firmware SD23. Of course, the damn thing doesn't spin down on my laptop. Or if it did, it would lag as it spins up.
Re: (Score:2)
I belive you're confused because you use the normal definitions of "debut" and "news", not the slashdot ones.
Even though people have been using this for over a year, it's still called "debut", because it's a slow news day.
Amen (Score:2)
Re:Amen (Score:5, Funny)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
> Definitely not. 32GB cache drives force you to buy an overpriced motherboard
No. They just require that you have a decent amount of SATA ports.
Considering that one of the cheapest motherboards I could lay my hands on 6 years ago had 6 SATA ports on it, this should not be a problem.
Although you can never trust the name brand pre-packaged kit. Some of those machines are like oversized Mac Minis. Then again, SATA cards are cheap too making expansion possible even on some lame-*ss Compaq.
Re: (Score:2)
Definitely not. 32GB cache drives force you to buy an overpriced motherboard with a rather odd Intel chipset (maybe AMD has one too). Then if power is lost basically ever, you're screwed.
Perhaps for the Intel caching application, but not for just having a big disc for data and an SSD for boot/applications. For that, anything would do, and many laptops will handle second drives in the optical drive slot, as the optical drives are increasingly SATA. My 1+ year old Lenovo E530 will take 3 HDs (one must be the small format SSD, one in the optical slot, and one "regular"). Sure, I can't set up the intel proprietary SSD caching for HD application, but I don't need to. And power outages on lap
Re: (Score:3)
No no, don't worry about the intel one. For Linux there's bcache and flashcache. Like with any multi-drive RAID-ish system, better to go with a plain software solution than a software solution hidden in a proprietary driver. There is also ZFS with support for read cache devices, but you won't get great boot speed because it's not persistent across reboots (there is a patch in one version of ZFS). Basically a bigger RAM cache that also doesn't get wiped out by applications
We use a mix of SSD and and HDD (Score:1)
The SSHDs are useful, but we tend to combine them on our multicore machines.
Different stripes for different tykes.
Secure Shell Daemon (Score:2)
bcache (Score:2, Interesting)
I must say I find Linux's bcache much more appealing than hardware hybrids. I'm also not sure how this new hybrid drive would cope with software full disk encryption.
Had one in a laptop (Score:5, Interesting)
I had one of the laptop versions for about a year and a half now, and it's definitely an improvement over a traditional drive and considerably cheaper per GB than an SSD.
I'm not sure why the majority of the population wouldn't opt for these as they still give you decent capacity and speed over dedicated HD or SSD drives.
Sure they're not as good as a dedicated setup with a SSD and a HD, but then again, the average user can still install everything on their C: drive without making any changes from the default installation.
Re: (Score:3)
The cache, whether it's flash or ram should be 'close to' the processor that benefits from rapid access to the medium.
Putting the cache on the other side of the interface always was and still is stupid. We don't need SSHDs. We need motherboard makers to stick flash close to the processor (in terms of access latency and throughput) on the motherboard.
Re: (Score:3)
Re: (Score:3)
>Your argument makes more sense for RAM cache
No it doesn't. Ram cache doesn't give you the boot time benefit of a non volatile cache. It would still be better if that was hanging of a direct memory bus to the processor, rather than on the other side of a SATA interface. SSDs are only fast compared to rotating disks.
Re: (Score:2)
Maybe call it something like, oh I don't know, Robson.
That's PCIe flash, it's crazy fast (Score:2)
PCIe is very "close to the processor", and PCIe flash cards are available. They are awesome, but require software such as bcache.
Re: (Score:2)
Yes. PCIe flash is good juju.
Re: (Score:2)
the average user can still install everything on their C: drive without making any changes from the default installation.
\
This one is important. I purchased the largest SSD I could reasonably afford(fiscally conservative), and found that messing with manually shifting programs around(lots of games) was costing me more time than what the SSD saved in quicker response times.
I'd like to buy a combo with a larger SSD, or see an automatic program capable of managing the caching, but these drives address a potentially huge market - almost as cheap as a HD solution while retaining ~90% of the 'real world' performance of an SSD.
I'd
Re: (Score:2)
I'm pretty sure windows can do it for you. I know it can at least do automated caching via USB sticks (for what that's worth). You can also buy external cards for pretty cheap that will make a hybrid automatically: http://www.amazon.com/gp/product/B009LIPHNC [amazon.com]
Personally, I've got a 128GB SSD which only really fills when I get lazy about deleting games from my steam library.
Re: (Score:2)
2) If Seagate's SSHD never performed worse than normal desktop drives I'd be tempted to buy the SSHD AND an SSD. However for some reason it's actually slower in many cases (look at the real world copy speeds for exam
Re: (Score:2)
> but then again, the average user can still install everything on their C: drive without making any changes from the default installation.
It's 2013 and this is still a consideration?
That's just plain sad.
Re: (Score:2)
I did have one in my previous laptop (one of the first "Momentus" drives). It's *great* to have 500 GB but even that got too tight. This was four years ago, flash only wasn't an option for that capacity. The hard drive itself was also reasonably fast, even without the flash.
There are some clear disadvantages:
- It's always spinning and making as much noise and vibrations as a HDD
- Uses more power than a HDD or SSD
- Should be as careful with the laptop as for HDD when shutting down or hibernating
- Significant
Re: (Score:2)
Seriously if I am having my Io cached *before* it is on oxide, I damn sure want to make sure it is redundant. This is the worst of all worlds frankly.
Maybe you should read up on the specs for these cache. The drives do not acknowledge the writes to the OS until the write has been committed to the persistent and non-volatile storage.
Apple fusion drive (Score:3)
I've been using a conventional hard drive paired with a SSD in Apple's Fusion drive configuration. This is only for Macs, but it makes it possible to use whatever size SSD you want, and the system automatically keeps the most recently written data on the SSD, saving the user the trouble of having to decide what to keep on the SSD and what to keep on the HD. In practice, the speedup is quite dramatic.
Re: (Score:2)
Smart Response Technology is a bit different, as it is a caching system. You generally find it used with small SSDs, because it is expensive to buy a big SSD to use it only as a cache. The Fusion Drive is a different type of system, because the SSD is not used as a cache--the SSD plus HD are combined into a virtual drive that is larger than both. The most recently written data is stored on the SSD. For example, I have a 240 GB SSD paired with a 1 TB HD. It appears to me as a single 1.23 TB drive. 240 GB is
Re: (Score:2)
I just bought a new desktop and opted for SRT rather than the hybrids.
OS drive: High performance (so quite expensive) dedicated SSD
Data drive: 2TB 7200rpm with a rather cheaper 60GB SSD as a cache
NAS : Purely spindle, it's there for infrequently accessed files and backup
The biggest benefit over the hybrid is that 60GB not 8GB. If I import 10GB of photos from my camera I don't want them getting flushed off the cache. If I want to play a 15GB install-size game I don't want to lose the benefits of caching. Rig
Re: (Score:2)
Funny thing is that Readyboost won't work on a drive through the SATA controller. Yes, I tried. You can hook up your SSD through a USB to SATA adapter and it'll work with that.... but that's just stupid.
Not necessiarly (Score:2)
For one thing, it is annoying to have to separate the OS and whatever apps you want to launch fast on to a tiny drive from everything else. So it is of use to people that need cheap space, but wish to have convenience.
However another use is for people like me: Who have SSDs, but can't afford them for all their storage. I have a 512GB SSD for my boot and program drive, and another 256GB SSD for my samples. However I then have 2x2TB HDDs for storing data, particularly bounced audio tracks. I can't afford that
Re: (Score:3)
For one thing, it is annoying to have to separate the OS and whatever apps you want to launch fast on to a tiny drive from everything else.
I just put / on the SSD and /home and /var (if it has to hold a lot of data) on the hard drive.
Oh, sorry, are you running one of those weird old operating systems that have that drive letter nonsense?
Re: (Score:2)
Re: (Score:2)
Even on Linux that's just a load of nonsense.
Yeah, because I totally buy expensive SSDs so I can open text files fast rather than boot and start applications quickly.
Re: (Score:2)
RTFS. This is a hardware based solution, not two separate logical drives, so you can just create on single root partition and stuff everything into it.
SSHD vs HDD + SSD + caching (Score:2)
In general, I don't see a lot of use of an SSHD on the desktop, at least not with only 8GB of NAND. There are significant advantages for a system (such as a notebook) where there is only a single available storage option.
However, if you have the capability to have both an SSD and an HDD you have a couple of much better options (e.g. on a notebook with an mSATA port or any desktop).
1. Install OS to SSD, manually manage installing things to HDD.
This will generally give you the fastest performance for the thin
Re: (Score:2)
I know that SSHDs work. They work pretty well for what they are. But they don't work as well as an HDD + SSD cache drive when that is an option (which it isn't always).
And guess what - it's possible to have a check box "Add SSD cache drive (faster performance)" as well. As much as a year ago I started seeing Ultrabooks configured with HDD + mSATA SSD cache drive out of the box (in fact, my current work machine is one of them - although I replaced both SSD and HDD ...). The initial boot image configures the
I've had SSHd running on all mu computers for 15y+ (Score:5, Insightful)
I really do hate overloading acronyms. SSH / SSHD is pretty well known already. It's what most unix folks (and I really do hope that that is the majority of slashdot readership) use to log on to servers every single day.
C'mon.
Re: (Score:2)
Shhhhhhhhhhhd
No... (Score:2)
A hybrid drive is better for MOST users. very few users can figure out how to configure windows to put /users elsewhere from the main drive. It's a HUGE design flaw that they dont let you pick it at install time like every other freaking OS on the planet.
True test (Score:2)
$40 SSD (Score:2)
The idea of buying a $40 32GB SSD and using it as a cache instead of a hybrid drive is silly - those cheap SSD's wear out very quickly with sustained writes.
I don't know if it's SLC, or what, on the drives, but you can push a lot of data to disk and not break the SSD on these things, which is fairly remarkable for NAND flash.
I've used one for over a year (Score:3)
one of the 2.5" drives. Benchmarks are a dubious thing - sure you can overwhelm the cache but in real world use that is not going to be an issue. Most users are doing the same thing(s) over and over and the drive optimizes to keep the necessary files in the ssd for rapid access. Likewise, modern drives, even the 5400 variants, are fast enough to keep up with video recording. So this really boils down to - are you willing to take a performance hit on the odd times you actually read or write a gigantic file in exchange for near equal everyday performance and a huge capacity at a very cheap price.
In my caae, the answer is a firm yes.
Re: (Score:2)
sure you can overwhelm the cache but in real world use that is not going to be an issue
L.A. Noire has a single file that's over 5GB in size. The full game is 12GB in size.
That's just one application. I multitask. 8GB of cache just doesn't cut it.
Re: (Score:2)
you aren't using 5GB at one time are you? Given the game reqs are 2 to 8GB I think not. As to multi-tasking - that is what it is good at: keeping the most frequently used files in the cache. But hey, go ahead and spend $600 and up for what can be had for $100. Or $1,200 and up if you want 2T vs oh.. $140.
Re: (Score:2)
The game has a 5GB file. I have no idea how much of it gets used at once, but it's not unreasonable to assume that across a 12GB game there's a fair bit of data being used.
I'm also certain that I'm using 8GB of storage when I'm processing 8GB of photographs.
I'm also certain that I'm using more than 8GB of storage over the course of a week. A $140 hybrid drive just isn't going to give me the performance I'd like.
So I spent $180 and got a 60gb cache not an 8gb one. But hey, go ahead and save $40 if you don't
Re: (Score:2)
None of your examples are using 8GB at one time. Are you really processing a single 8GB photo? Highly doubtful. Do you have 500 16MB photos? Possible. Once you start batch processing the drive logic is going to be precaching the next likely hits whether its an SSD or HDD. And if game performance was going to be affected by disk drive speed in the way you imply its a first for me - I've yet to see one spec out with "SSD required".
As to cost - no, you are paying $180 more than I am to get a bigger buffer
Re: (Score:2)
None of your examples are using 8GB at one time. Are you really processing a single 8GB photo? Highly doubtful. Do you have 500 16MB photos? Possible. Once you start batch processing the drive logic is going to be precaching the next likely hits whether its an SSD or HDD. And if game performance was going to be affected by disk drive speed in the way you imply its a first for me - I've yet to see one spec out with "SSD required".
Game performance as a 'frames per second' isn't a storage IO issue. Game performance as a 'why is this game taking 40 seconds to load a level' is. That's where an SSD helps. Trust me, there's a very discernable difference.
And yes, I quite often have 500 16MB photos. Forget precaching, there's "write to hard disk, read from hard disk, generated cached image, write to hard disk, read from hard disk" even before I do the actual photo editing. Being able to do all of that on an SSD cache (and again when I'm rev
Re:Does it require windows only software? (Score:5, Informative)
Re: (Score:2)
Re: (Score:2)
I'm going to voice the opinion that software-independent caching system is definitely NOT the way to go: the firmware has no way of knowing how often the files are going to be accessed in the future and if they should be kept in cache for a longer time, the firmware doesn't know what type of a file it is, the firmware doesn't know the size of it and so on and so forth -- basically, it knows none of the important details and will end up caching stuff it doesn't need to cache, will end up not caching things that should be kept in cache and since it doesn't know actual filesystem details it can't even optimize its own structures to match those of the filesystem in use. It's just a really basic block cache.
Apple's Fusion drive works quite well that way. They don't care about files - everything is based on 128 KByte blocks; probably the write page size of the SSD drive. In a typical configuration (128 + 1000 GB) you have about 9 million blocks of which the one million that is used most often is located on the SSD drive. It even works if you use a VM to run Windows.
If you have a large music library, typically only the metadata from each music file will be on the SSD drive. If you have apps that are large bec
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
How the Fusion drive works (Score:2)
The Fusion drive does not use the SSD as a cache. Rather, it merges the SSD and the HD into one large virtual drive, keeping the most recently written data on the SSD and the old stuff on the HD. The advantage, of course, is that you can pair a reasonably sized SSD, which will hold most of the data that you or currently using, with a big HD, and get much of the speed benefit of a SSD at much lower cost than a SSD big enough to hold all your data, and without the need to "triage" your data to decide what sho
Re: (Score:2)
OpenBSD should sue them.