Slashdot Mirror
WD Explains Its Windows-Only Software-Based SSHD Tech
crookedvulture writes "Seagate and Toshiba both offer hybrid hard drives that manage their built-in flash caches entirely in firmware. WD has taken a different approach with its Black SSHD, which instead uses driver software to govern its NAND cache. The driver works with the operating system to determine what to store in the flash. Unfortunately, it's Windows-only. You can choose between two drivers, though. WD has developed one of its own, and Intel will offer a separate driver attached to its upcoming Haswell platform. While WD remains tight-lipped on the speed of the Black's mechanical portion, it's confirmed that the flash is provided by a customized SanDisk iSSD embedded on the drive. The iSSD and mechanical drive connect to each other and to the host system through a Serial ATA bridge chip, making the SSHD look more like a highly integrated dual-drive solution than a single, standalone device. With Intel supporting this approach, the next generation of hybrid drives appears destined to be software-based."
Yeah, that was a nightmare!
Linux already supports SSD-HDD caching with normal drives so if anything, it will probably already work or work with little changes. Otherwise, just pick up a tiny SSD and ignore this solution. http://bcache.evilpiepirate.org/
Nice troll, btw.
It isn't clear, exactly, from TFA what the drive will look like when you plug it in. Both components(the HDD and the SSD) apparently can function as SATA peripherals; but they are both behind some sort of bridge chip, type unspecified.
If the 'bridge chip' is just a reasonably generic SATA port multiplier, then an unsupported OS, or Windows without the driver, will just see two drives, the larger mechanical one and the smaller flash one. This would leave the way open for any OS with SATA and AHCI support to do whatever it prefers to get the best performance(on Linux, I assume that'd be at the filesystem level, with something like btrfs)
If the 'bridge chip' is some sort of proprietary oddity, and the vendor driver is required to even communicate with the flash portion(presumably at least some part of the drive will be visible as a normal SATA device, or booting without specific BIOS support would be a problem...), then that's pretty much worthless.
The iSSD and mechanical drive connect to
I believe I speak for the majority here when I say.... D'ARGH! KILL IT WITH FIRE NOW! This is yet another pathetic attempt by WD to marry it's crappy line of mechanical drives to SSDs in order to stretch their relevance out a little bit longer and keep them from having to retool their assembly lines and such to produce SSDs exclusively. Weeeell, good for you guys. But as my father would say: "Shit or get off the pot." Either switch to SSDs, and eat the cost, or stick with mechanical drives because they're cheap. But don't waffle and try to do both; You're getting the worst of both worlds then.
#fuckbeta #iamslashdot #dicemustdie
Sounds like Apple's Fusion drive.
Perhaps WD is releasing this with Windows support and hoping people can upgrade to their drives in OSX?
... in all the big Fortune double-digits, that have their data centers overflowing with Linux servers.
The worst part of this, is that when WD goes bankrupt, as a result of this brilliant business strategy, there'll be even less competition in the HD market, which always means higher prices.
As an end-user, I'm NOT going to put up with a solution like this.
Even if it somehow performs better than current hybrid drives.
Even though most of my work is done on a Windows platform.
Hybrid drives are already a big compromise for minute gains.
Tying it to an OS choice?
NO FUCKING THANKS WESTERN DIGITAL!
In a budget situation I'd rather just put up with a competitor's hybrid or a plain old mechanical disk.
In a performance situation I'd rather just spring the extra cash for a real SSD. Better returns and more flexible.
Chas - The one, the only.
THANK GOD!!!
WD has taken a different approach with its Black SSHD
They'll have a lot more explaining to do, once some hacker, cracks the SSH password, starts pwning WD disk drives, and they begin to spew forth spam... :)
And its worth reminding people, that Windows already caches stuff in RAM, if you had 24GB of ram then it would be a lot faster cached, and the only gain with these drives is on startup and then not by much (since Windows arranges the disk so the common items are close together ready for boot).
So WD simply remind everyone why hard disk makers are struggling to remain relevant.
only in black?
There was an unknown error in the submission.
That's correct. Non-fw cache will be more effective.
We don't know (yet), how WD drives work - will we see one whole block device or maybe it will have two SATA ports with two separate drives. I'm not sure they will support access to SSD to the some encapsulated stream.
There are Linux solutions like bcache and flashcache that can deal with cache. So, maybe it's time to use it and include into kernel?
"It feels like I'm at the Zoo when reading this thread - I'm frightened, but it's interesting" (c)
Yep, and flashcache (facebook).
But none of these are included in the kernel.
Picking up tiny SSD is not a solution for notebook or for the desktop - you need more ports, more money, more energy etc
"It feels like I'm at the Zoo when reading this thread - I'm frightened, but it's interesting" (c)
Uh, okay, whatever.
Guess I won't be buying one. Best of luck to those that do.
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
they might call this the Western Digital 8.
WD just shot themselves in the foot .. many times over! People should boycott this company and their proprietary SSD technology.
At under $250 for a 256 Gig SSD, why would anyone buy a combination drive anymore? This would be like GM announcing that their next flagship green vehicle is powered by 150 horses.
The dm-cache device mapper target was added to the kernel in Linux 3.9. bcache is apparently on track for 3.10
Wonderful idea. I can't wait to run right out and buy that. /sarcasm
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
Think secure boot. If the 'bridge chip' has a key, that only the trusted driver can supply, then with UEFI and "secure boot", they have just locked down the machine to windows only.
You are being MICROattacked, from various angles, in a SOFT manner.
The nice thing about Seagate HHDDs or SSHDs or whatever the companies what to call them now is that they just work. You drop it in a system, it works like a normal drive but faster. The flash works like cache on a RAID controller or the like. It just speeds things up.
With this, there's mucking about. Even if you could use it as separate drives, why would you want to? If I want to to just some small SSD storage and larger magnetic, I can. In fact I do. In my laptop I have a SSD for OS and apps and an HDD (actually one of Seagate's hybrids) for media and samples. My desktop is the same but more and larger drives.
It just seems silly to me. An all hardware approach seems much better and clearly doesn't cost that much as Seagate's drives are not expensive.
Do it in OS-agnostic firmware, you lazy shits.
If I run sshd on my sshd? What if I have a virtual machine that runs sshd on a virtual sshd on a computer with an sshd that also runs sshd?
Also other stuff (*BSD, solaris, etc) has had support SSD caching with ZFS for a few years now. Linux ZFS exists too but not built into distros yet, still it's very easy to install on many linux distros. I don't know what the current state of ZFS is on macs but they may be able to do this too.
Like make the OS learn what stuff we usually read from the drive, and keep that stuff handy. That would be so cool!
It's probability because there is no set life just a chance of failure. For example, with tossing one coin you have a 50% chance of getting one head and with two coins you have a greater chance of getting at least one head. With two components the low risk of failure per year gives a very long time for each separately for the time when half are expected to fail. Combine them and the chance for either of them to fail is combined so the time when 50% of the complete systems are expected to have some sort of failure is reduced to a lower time than either of the parts. :)
That's what you get for saying "that's not how it works" without stopping to think about it
No, 24GB RAM is $60, not $200. I just bought 4GB sticks of DDR3 for 300 baht retail, = 6*$10 = $60.
Where the heck did you find that?
Newegg's cheapest price for 2x4GB is $53. ...if you have the RAM the HD with flash is cached in RAM regardless and the flash cache adds nothing.
Not true.
1) The SSD cache will persist through reboots. That helps not just boot time but also stuff that you do for the first time after booting. I keep my Windows computer hibernated when not using it, and see a noticeable speed boost from SSD.
2) I'm not sure if the hardware hybrid drives do this, but at least Intel's Smart Response claims to be intelligent about what it caches. For instance, if you do a long sequential read from the HDD, it won't cache the whole read, just the start. In some sense this makes the cache seem larger. I'm not sure if OS buffer caches do this, but I kind of doubt it because they are made for a separate purpose.
3) You're ignoring writes.
Twice the productivity!
sysadmins and parents of newborns get the same amount of sleep.
Maybe that phrase needs to be updated a bit. In more ways than one.
In my computer, I have a Crucial M4 SSD for the boot drive and the more speed fasterness "crucial" apps. Then I have a WD Black terabyte drive for all the shit that doesn't need to be maximum possible speed.
Sometimes, I change my mind what needs to be faster than what is not at the forefront of my mission anymore. That's when I move my files around manually. Mind you, these are usually 8 - 20 GB of files or whatnot. This type of operation I do not want an un-brained background process to be performing at random times. If it picked the wrong time, I might drop FPS in an online match that was worth so many imaginary dollars to nobody at that precise moment...
Remember those great Intel software-based modems back in 1993 that Intel had for Windows 3.1 that weren't supported by any operating system afterwards?
Remember how Intel sold that line of business and left us all hanging? Yeah.
Software based hardware like this is destined to be a one-trick pony. Use it in one system, and then it's stuck on that operating system for the duration. You'll be left in the lurch when the next version of you OS is released.
Go ahead and line-up to get screwed by WD and Intel. I'll skip this round of fleecing.
in your Ultrabook? Is there a research/poll out there that shows people buying Ultrabook and they want 500GB to 1TB of storage? Unless you have a lot of videos, I think 256GB is sufficient. Are there many people carry around their movies library?
With how much straight SSD prices have dropped over the past few years, I don't even really see much need for a hybrid drive. In 2011 I bought a 60gb ssd for $95 ($1.58/gb). Today, I can buy a better performing 500gb ssd for $350 ($0.7/gb).
Wow.. thanks FYI.
Since OSX and Linux already do this, maybe only Windows needed a driver.
I'm not talking about suspend data.
I'm talking about after the system boots, and I start Opera, or I start HL2. That data isn't going to be cached. It's not even going to be cached if it was in the buffer cache before shutting down.
(I use HL2 because it, and especially the episodes, benefit tremendously from being located on an SSD. The "loading" times are cut tremendously, which helps a lot with one of that series's big immersion breakers. (I am comparing HDD to full SSD -- not hybrid/cached -- but benchmarks for hybrid setups show a lot of the benefits.))
First, that long read will take a lot longer to go from HDD -> RAM than from SSD -> RAM.
But mainly, you missed my point. If your working set (defined loosely) is more than RAM size, it won't all fit into RAM. If I cycle through a couple different data-heavy games then go back to the first one, that will no longer be in the buffer cache because it will have been flushed. It will, however, likely see a speed boost from part of it being in the SSD cache.
Lots of writes are synchronous, you know.
Heck, there's even a substantial difference on the Smart Response benchmarks (that's Intel's northbridge-based, hybrid-hardware/software SSD caching scheme) between using the SSD as a write-through and write-back cache.
Anyway, I'm not saying all workloads will benefit. But I've found that the SSD has helped, and many others (e.g. Linus) have said that going from a HDD to SSD is by far the most cost-effective upgrade for many people. Saying that it only helps boot times is just flat out wrong.
Having installed an SSD for the purpose of caching my HDD, I'm afraid I must disagree. The performance gain really has been pretty amazing, and it's really great having a 2TB drive that is automatically cached to whatever I happen to be using the most. SSDs cost way more per GB, so I could buy a smaller one. HDDs are way slower, but I don't commonly hit that data. I really can't identify any disadvantages with what I did, except that installation took a bit to figure out. A hybrid drive would make installation much easier. I will concede, you are correct that once SSD cost/GB is the same or less than HDD, there may not be a point for HDD. But, this looks to be quite some time from now. I want more performance now.
No, 24GB RAM is $60, not $200. I just bought 4GB sticks of DDR3 for 300 baht retail, = 6*$10 = $60.
Where the heck did you find that?
Since he paid in baht, I'm guessing somewhere in Thailand.
When our name is on the back of your car, we're behind you all the way!
There are different grades of flash. SD cards and USB sticks are bad. Actual SSD drives that you use as a substitute for HDDs write faster than HDDs.
Anandtech.
"While the Vertex 3 [pure SSD, no caching] is still a bit faster, you can't argue that Intel's SRT doesn't deliver most of the SSD experience at a fraction of the costâ"at least when it comes to individual application performance."
"Performance keeps going up. The maximized SRT system is now virtually indistinguishable from the standalone SSD system."
"I worried that Intel's SRT would only cache the most frequently used level and not improve performance across the board. I was wrong."
The story isn't entirely good -- after all, it's not a full-blown SSD -- but like I said, it's absolutely false to say that SSDs only help with boot time.
Congrats on your ass-talking, considering that you just admitted that you have no information to support your claims.
shut the fuck up i'm not a troll, you're a fucking troll.
that software you linked to is not FREE as in speech.
Even Speech comes with a price.
Yes, I know. I'm relying on extrapolating from anandtech's statement that game loading speeds are helped a great deal.
So the WD's algorithm is perhaps not as good as Intel's Smart Response. Benchmarks would tell. That doesn't torpedo the idea of an SSD or hybrid drive, however.
That's unlikely. Typical buffer cache replacement algorithms are approximating LRU. If it were perfect LRU, then cycling 24GB of data plus 1 page would be sufficient. Since it's only approximate it would be rather less, but there are also "inefficiencies" in that figure that would raise it. In any case, you would be highly unlikely to require anything remotely close to 48 GB.
You think it's DX drivers that cause game load times to be long?! No, it's level data and textures and models and such.
Write speed for HDD: "The range can be anywhere from 50 â" 120MB / s"
Write speed for SSD: "Generally above 200 MB/s and up to 500 MB/s for cutting edge drives"
source
No, they don't. They use better flash and they bank them.
And by the way, you're continually ignoring my statement that the SSD cache persists across boots. With an SSD cache, playing HL2 today helps it go fast tomorrow. With a buffer cache, it does not.
Even if I ran windows I can't see having a driver for my hard drive. It should just work no matter what OS I am running. Sad and stupid. That's okay though as for me I don't think the hybrid is the way to go. SSD for the OS and external platter type for storage. Like most compromises this seems like the worst of both worlds.
Sounds like WD knows they have to write the drivers for WIndows, but for Linux they will just let the community do it. Saves them money.
Heaps of people seem to think this is the same thing... but it is an extremely inefficient way of utilizing your space - if you have a 2GB file with only say, 100 MB of which is "hot" then you are moving the entire 2GB file.
A controller / driver / intelligent operating system can determine which BLOCKS in the file need to be cached, so in the above example, instead of consuming 2GB of SSD, the caching would consume 100 MB for that file - leaving the other 1.9 GB free to cache something else.
Also, you are not able to re-shuffle what is cached, on the fly while using the system as performance requirements change. A controller/driver or intelligent OS can do this. So, say you're in a game that wouldn't fit on your SSD due to you putting other stuff there, and it needs to frequently read/write from part of a data file as you traverse a level. The controller/OS can cache that hot part of the file while you're in the game, automatically.
If you were shuffling the data around yourself, manually you just couldn't get that optimisation.
I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
That speed is after the RAM cache is through.
Don't believe me? Another benchmark. That's a random drive that I arrived at by entering "anandtech ssd review" into Google and picking the first link. It's not even the fastest drive in the charts on that page, by far.
The write speed they measured was 317 MB/s. The total amount of data written to the disk was over 100 GB -- so plenty to get past the RAM cache and to the actual SSD. If the 317 MB/s is limited by the RAM and not Flash speed, that means the Flash is even faster.
Don't like that benchmark or Anandtech? OK, here's one from Tom's Hardware. Measured the time do decompress a 50 GB folder. (Again, the RAM cache on the drive and your OS buffer cache (and benchmark scenarios that don't happen in the real world.
Fine, I'll give you a real-world scenario as best as I can. In a moment, I'll switch over to my Windows box and compile something on SSD and then on HDD. We'll see who wins.
Meanwhile, I've been happily using my 750 GB Momentus XT at home (I have 256GB SSD at work), quite happily as a trade off between capacity and performance. To get the 750 GB i want in my laptop would cost me about 6x the price and would certainly not give me 6x the performance. It's also 1/4 of a mortgage payment back on my mortgage.
I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
Bungled my HTML. Here we go again:
That speed is after the RAM cache is through.
Don't believe me? Another benchmark. That's a random drive that I arrived at by entering "anandtech ssd review" into Google and picking the first link. It's not even the fastest drive in the charts on that page, by far.
The write speed they measured was 317 MB/s. The total amount of data written to the disk was over 100 GB -- so plenty to get past the RAM cache and to the actual SSD. If the 317 MB/s is limited by the RAM and not Flash speed, that means the Flash is even faster.
Don't like that benchmark or Anandtech? OK, here's one from Tom's Hardware. Measured the time do decompress a 50 GB folder. (Again, the RAM cache on the drive and your OS buffer cache (<8GB) isn't gonna save you.) The throughput during that test was ~300 MB/s. You can't get that if you're not writing that fast to the flash.
Fine, I'll give you a real-world scenario as best as I can. In a moment, I'll switch over to my Windows box and compile something on SSD and then on HDD. We'll see who wins.
Yeah, a drive that depends on any kind of OS besides the bare metal firmware on the board to which it is attached makes me uneasy. There are just so many more answers to the "what could possibly go wrong" question.
... whatever
OK. So the difference was actually very small, only 1-2%, though it did hold up across several runs and configurations. I also tried HL2 level loading times, and saw about a 4% improvement there.
In all cases I explicitly warmed the cache. And for the compiler scenario especially, that's a very realistic scenario. (Though I'd be very interested in trying it on my work computer, which has far more cores -- that would put more stress on the I/O system. Too bad I don't have an SSD in that machine.)
However, I still am going to stand by my statement that an SSD will often help, because I'm not convinced that having the cache warmed is the common scenario, or would be the common scenario even if you got a very large amount of RAM. (This is also in response to part of one of your comments that I didn't address earlier, where you mentioned preloading again.)
Take that 24GB number. In a typical month, I will (1) play multiple large games, often which I haven't played for a while (e.g. I go back to Portal 2 every few weeks and play some more PTI levels), use a couple programs like Latex with an absolute ton of tiny files (an enormous hit if they are not in cache), open up my photo library in lightroom (where just the catalog of previews is over 6 GB and I can easily hit ten gigs of photos just by flipping through things), and often record a video with Fraps (~2 GB/minute) and then do some light video editing. Now, I can't put all of that onto an SSD. But I can come a hell of a lot closer then getting it into RAM.
The problem with caches in general is that they can only react. If what you do changes around between different data-intensive tasks fairly rapidly and your cache isn't big enough, there's only so much it can do. If in my real usage the level load times dropped -- and they definitely have -- that's what matters.
Sadly, we can rarely afford their silence... .
Um, no, not really. While Seagate doesn't explain their caching strategy, my guess is that unlike things like bcache, which cache at the block level, the seagate probably caches at the logical file level, which would explain why it's Windows only, they only implemented their algorithm for NTFS. My guess is they broke with tradition and implemented a disk that is aware of the logical layout of the files on it, instead of one that simply manages blocks(with file metadata just being another one of those blocks)
File level caching has a lot of advantages, for starters there tend to be a lot fewer blocks than their are files, meaning that the overhead imposed by the caching algorithm is a lot lower(flashcache for instance uses 500 megs of memory per TB of storage IIRC). Furthermore, by caching at the file level you can apply heuristic rules like not caching movie or music files, where performance is very unlikely to be critical etc.
Of course this all comes at a cost, having the disk actually know about the logical layout of the file system does break a lot of conventions and introduces a large # of potential issues....Trying to upgrade the file system on one of these things is probably not very pretty.
Monstar L
Nice post, but the article is about WD, not Seagate....
and don't forget the excellent EnhanceIO, https://github.com/stec-inc/EnhanceIO. This allows you to add cache to any device on the fly without and pre-formatting. For the storage tiering inclined people, similar performance gains can be made with btier, http://www.lessfs.com/. Mark from lessfs also has a great deduplication project as well called...well... lessfs.
Except these disks are more standard. They're basically an SSD and a HDD hooked to a SATA multiplexer (that lets you connect more than one SATA device to a SATA port. NOTE: Note all controllers support MUXes. Also, both drives share the bandwidth of the upstream port).
So plug this into a Windows PC and install the drivers, and two drives become one. Plug it into a Linux PC and you see two drives. Plug it into a Windows PC without drivers and again, you get two drives.
I would be concerned about how accessible my data was without the drivers. So you're using Windows and your data is partly on the platter and partly on the SSD; you reboot to an OS without the driver (i.e. the driver breaks when you upgrade Windows, you boot into Linux, whatever) - can you still get at your data. My guess would be that whilst the contents of the drives will be accessible as two independent drives, they will be in some undocumented format and therefore irrecoverable.
they'll have some tools or another for it. of course nobody is forcing you to buy these.
HOWEVER! current hybrid drives that figure on their own what to keep on the ssd are total bullshit that only work fooling repeated read speed tests. this actually makes a lot of sense, that the os can hint to keep certain files on the ssd.
world was created 5 seconds before this post as it is.
Similar question: Which would you rather have: software RAID or hardware RAID? On Linux, software RAID is usually faster, cheaper, more reliable, less buggy & fuller featured.
So yes, I'd prefer the drivers in my operating system rather than buried in some inaccessible firmware somewhere.
I suspect the 'disk' is still just a block manager (well, disk(s)) that doesn't have any awareness of the FS, just the driver half that makes it 'look' like one volume 'transparently'.
This is of course similar to 'fakeraid' with all the potential trappings thereof. If it presents as two disks otherwise or something similar in a standard way (and the driver is just to deliver product ahead of the standardized implemention being available) that might not be too bad.
XML is like violence. If it doesn't solve the problem, use more.
File-level caching also has a lot of disadvantages. First and foremost, as you mentioned, it has to be tailored to specific file systems, like NTFS. For Linux and Unix OSs, there are so many file systems that porting file-level caching software to every one of them would be a nightmare. Linux/Unix has a generic block layer, so block-level caching software written for Linux/Unix is file system independent.
Second, the list of heuristics can easily spiral out of control. You might not want to cache a video file of a movie you're watching, but you probably would want to cache a video file of a movie you are editing. You have to know where the file came from, where it's going, how fast you want it to get there (i.e., priority), what programs are using it, etc. There's just too much contextual metadata that needs to be tracked; the more complicated the contextual metadata is, the more complicated the heuristics need to be. Also, statically defined heuristics may be inappropriate for dynamic environments.
Third, why cache an entire file if you're only working on a very small percentage of it? I have to admit, some file-level caches will only cache the "popular" portions of files, but in that case, why not just use a block-level cache and enjoy its better performance?
Also, I'm confused about "there are a lot fewer blocks than files," did you mean that the other way around? If you have a single 100 MB file, and the system's block size is 4K, you have a lot more blocks than files. There may be a smaller memory overhead with a file-level cache, but there's a greater CPU overhead due to all the contextual metadata and heuristic processing.
bio->bi_end_io(bio, error);
TFA states: "With Intel supporting this approach, the next generation of hybrid drives appears destined to be software-based."
Intel supported 'Windows only' Itanium processors as well and how did that work out for them?
WINDOWS GOOD LINUX BETTER.
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
Picking up tiny SSD is not a solution for notebook or for the desktop - you need more ports, more money, more energy etc
A great many laptops have mini PCIe slots. I have a Dell Inspiron 1720 (core 2 duo from May 2008) that has THREE of these (two of which are open; one holds a wifi card which I recently upgraded). You could pop an mSATA in either of the other two. BIOS might not see it correctly (one of the slots in the 1720 is labeled to support mSATA, and docs say the others won't work, but I don't believe it... I had to put the new wifi in the mSATA slot because my card was half length and that's the only place I could get antanna's to stretch to - wifi catcher switch doesn't work now, but linux sees and uses the new card just fine).
For the desktop... I don't know what you're talking about. You can fit two 2.5" drives in a 3.5" drive slot (or in the external 3.5" spot), or more in a 5.25" slot, or use a PCIx card solution, etc.
More energy? Really? The SSHD's have a flash drive and spinning platters in them. Having those separate *might* use a little more power, but that just depends on what drives you buy. The difference will be negligible at best.
Several things people:
1) These are Hybrid drives, NOT SSD.
2) Worst idea ever, wtf, lol!
3) Who actually buys a Hybrid drive anyway?
They are usually advertised as the "best of both worlds", but I (and I think most, which is why they are not popular) think they are more like the worst of both worlds.
Don't compare them to SSD, they are not SSD. Don't compare them to HDD, they are not a HDD. For the love of god don't buy them. It makes much more sense to buy an SSD for your system drive, and a HDD for your storage drive, like everyone that knows anything will tell you. Mashing the two together, will just give you a shitty compromise, that you can likely find for better/cheaper in a higher end traditional fast RPM HDD.
Not embedding your cache controller in firmware which is independent of OS, and not only making it dependent, but dependent on Windows, well that is just stupid. Not only are you limiting your market, but you are going to have to try and keep up with every stupid change that they do. I have no idea why they would do this, what would be a single benefit. Perhaps they save a dollar in not having to put the firmware memory on board? Who knows, crazy.
Does anyone know anyone who has bought one? Can they formulate a cogent argument as to why they did?
To Summarize: Stupid, but Irrelevant anyway.
That's good unless your operating system is inaccessible.
That's a much bigger problem with Windows than with Linux. You can easily mount md-raid drives from linux booted from USB or CD. I don't believe Windows supports booting off of anything other than a hard drive.
I would not want a windows-only software hard drive.
… don't buy WD drives!
Similar question: Which would you rather have: software RAID or hardware RAID? On Linux, software RAID is usually faster, cheaper, more reliable, less buggy & fuller featured.
You're joking, right? Last time I checked, the MD system wasn't big on adding drives / swapping drives for larger drives / replacing bad drives online with no disruption, and offers no way of visually indicating bad disks for local hands to remove. I also couldn't find any agreement on how the system should discover MD volumes, and there seemed to be some bizarre recommendation of having an extra device for metadata or something. In the continued absence of ZFS, I'll stick with HP's HBA RAID for now.
Because a lot of people remember WinModems, and others are remembering Microsoft holding the keys to the current boot system. And Microsoft suing Android vendors...extortion is what I call it, even if lawyers use a different term. And others are remembering Noika...to mention only one recent example.
I, personally, am upset because I've frequently bought equipement that I couldn't get to work on Linux, which makes me very reluctant to trust any company that's providing "windows only" drivers.
This is abetted because one can never buy in stores the models that have been tested and found to work with Linux, so one needs to buy something that's "nearly the same" and which "will probably work". I will definitely pay extra to avoid this model, but I will, at the same time, hate WD for making me pay extra. (If they only offer this model at a premium price, and continue to offer their standard models, this final comment won't apply. But I won't know that for a year or so, when they've had a chance to rev all their models.)
I think we've pushed this "anyone can grow up to be president" thing too far.
Yeah, I meant the other way around, nice catch.
Monstar L
totally agreed, i rn a 60 GB SSD with two 2TB data disks (one is rsynced to the other so I have an immediate backup of everything all the time). I built the box in 2007 with a core2 and have upped the RAM to 8GB, which helped a little, but that SSD, oooh that made a huge difference
Subversion of spatial scale luxury decoration ideas.
I think you misunderstand that such a thing is an unlikely use case outside of embedded systems deliberately designed to save power by not logging stuff to disk at short intervals. There's some linux laptop stuff aimed at the same thing and from what I read it was not a trivial exercise to cut down on all that disk activity.
SSD failure rates indicate far less reliability over time than RAM. In fact some mechanical disks appear to have lower failure rates than SSDs at this time which is why I spoke up against the mistaken idea of the above poster of some vast difference. Tom's hardware or similar will help if you want some up to date numbers.
It's a version string, it's not a floating point number.
If could be written as "Version 3, Release 10" in which case you would have no problem recognizing that "Version 3, Release 9" was the prior one. But remember, devs are lazy, and keystrokes are expensive, so shorthand notation just gives you the important (numeric) pieces, with a visually minimalistic delimiter.
This is even more apparent for projects with Major/Minor/Patch level designations. They are, once again, just using periods as a simple delimiter in a shorthand notation. Surely you don't complain that 2.3.32 is less than 2.3.6 do you?