Slashdot Mirror
SSD-HDD Price Gap Won't Go Away Anytime Soon
storagedude (1517243) writes "Flash storage costs have been dropping rapidly for years, but those gains are about to slow, and a number of issues will keep flash from closing the cost gap with HDDs for some time, writes Henry Newman at Enterprise Storage Forum. As SSD density increases, reliability and performance decrease, creating a dilemma for manufacturers who must balance density, cost, reliability and performance. '[F]lash technology and SSDs cannot yet replace HDDs as primary storage for enterprise and HPC applications due to continued high prices for capacity, bandwidth and power, as well as issues with reliability that can only be addressed by increasing overall costs. At least for the foreseeable future, the cost of flash compared to hard drive storage is not going to change.'"
Doesn't creating a striped RAID make up most of the performance issues from using a HDD over a SSD? At that point, it's more the bus or CPU that's a limiting factor?
-SaNo
Captain obvious...
Fairly sure that increases in capacity usually means increases in performance as well. I have not seen any ssd on the market today that illustrates otherwise. .50$ a gig on ssds. Prices have been plummeting. You can get a 256 gig drive for ~100$ . 1TB drives have been almost hitting the $400 mark.
We're down to less than
When 2TB ssd come on the market, you'll see the rest drop in price as well. I'm not quite sure where the author is getting their information. Check the price drops over the last two years and you can see they haven't hit bottom yet.
WTF?
Reading through comments I start notice that some of them are cut off mid sentence. Investigating it a bit further, I notice that many of the comments are. Nothing to indicated this, other than poor grammar and no period (ie. - using ellipses '...' to indicate comment continues)
Anyone else notice this?
Anyone else think this is asinine?
Is there some behavio they are trying to encourage (ie - force)?
Today we can have an SSD for the price of $0.50 / GB. It is already good enough.
not buying into SSD bullshit
thats why its nto comin g down anytime soon
the article tells you why
its crappy technology
A 480GB Crucial M500 is slightly cheaper per GB than a 4TB spinning drive right now. I think the 960GB SSD is as well.
With enough numbers and pages tedious explanation, you can basically say anything you want.
I got all the way through the first page. The argument behind this seems to be that prices in the SSD world are higher than prices in the HDD world for technical reasons, and the technical reasons aren't going away so the price gap will remain. Until someone overcomes the technical problems, and the author, who is clearly a world leading expert in this field, doesn't know how the technical problems will be overcome. I think it'll take him by surprise when it happens.
"Is the Chief Priest an Offlian? Do dragons explode in the wood?"
Hard drives never completely obsoleted tape drives, and SSDs will probably never obsolete all hard drives, unless they can amazingly close the gap between SSDs ($0.50/GB) and tape drives ($0.01/GB).
the limitations are the physical limitations fundamental to flash storage itself, it is more likely that non flash based solid state drives will be required, so the argument is that this will take time (not that hard drives wont eventually get replaced)
Good thing there are other technologies to replace flash that will be around factors of 10 lower cost, and higher in speed.
Samsung, IBM and several Universities have new tech working out towards mass production levels as we sit and read this article.
Just a faulty prediction. They'll still be called SSDs or flash drives as the nomenclature no longer specifically refers to the tech inside, but the form factor.
Likewise, worker bee machines that are pretty much dumb terminals are not going to use SSD. But other machines that people actually do and store work on, that may be something different.
Look, tape is on the order of penny per gigabyte. Hard disks are somewhere between 5-10 cents a gigabyte. SSD is about 50 cents a gigabyte. Many people still back up onto hard disk even though tape is more reliable. We are going to use SSD because there are benefits that justify the order of magnitude increase.
"She's a scientist and a lesbian. She's not going to let it slide." Orphan Black
with spinning rust, you might re-engineer the bulk process that coats your disks, but the boost in recording density depends on changing the parameters of the head. bulk process and one device. compare to flash, where to boost density, you have to tweak each storage cell, controlling for defects and manufacturing flaws, where the yield of each cell multiplies, so defects are exponentially likely.
disks (and to some extent tape) will always have scaling advantages over litho-fabed storage.
you can certainly argue that latency and bandwidth also suffer by the same process - but for the most part, disk performance really is fast enough for most uses. it's a bit surprising that more disk vendors haven't tossed onboard a small flash chip (to all lines of HDs).
The sizes of SSDs which are readily available and affordable are big enough not to leave most consumers wanting for more. In the 90s I used to pay more per HDD than I do now for SSDs (unit price, nevermind the price per GB). Half of all drives I bought in the last two years are SSDs. I haven't bought a maximum capacity HDD in five years. HDDs are running into serious technological problems: The next step is shingled recording, which will bring the problem of write amplification to HDDs, slowing them down even more in return for higher storage density. HDDs are rapidly becoming a specialty product for businesses with unusually high capacity storage requirements.
We need reliable hybrid drives with 120-160+ GBs of flash memory, instead of the ridiculously worthless 4-8 GB ones we have now.
A hybrid with a 1:30 or 1:20 ratio of flash to platter (200 GB for 4 TB for instance) would pretty much be perfect for anyone, even enterprise applications if RAID controllers cooperated with the hybrid caching properly.
We do not need 100% flash, just give us a practical median.
In fact, I guarantee if someone made a hard drive with a controller with an mSATA slot for adding a SSD and offered the controller to be setup as pass-through (act as two drives) or caching (SSD keeps a cache of platter), it would sell like crazy.
An mSATA would fit easily beneath a standard 3.5 inch platter hard drive.
http://www.notebookreview.com/...
Damn, why didn't I think of this?
People, is it true? Would the market bear a "Republican Technology News" site?
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
The article is a bit weird. It keeps saying to ignore consumer: low price, cheap parts, focus on mobility as inapplicable to enterprise. But then it focus on enterprises disks that aren't far removed from consumer models rather than enterprise models like IBM's flash solutions (ex 840: 33T per U so more than 1P per rack). If we are going to look at enterprise flash I don't understand why you would focus on smaller solutions. Obviously the $8-14g price is even higher but it is at those price points that flash makes sense because it is allowing virtualization to replace multiple servers and thus replacing real estate, network and energy costs not raw HDD costs.
I guess if the article is just saying that mid sized server solutions (say $5-100k) are unlikely to go all SSD before 2020 I have to agree. But I think it should have been more specific. Once you exclude the consumer space and the mass storage space we are probably talking a tiny fraction of the whole HDD market.
I have a 120 gig Sandisk Extreme 2 SSD and as a performance upgrade, you really can't do better than an SSD, assuming a minimum of 4 GB of ram. I was a little skeptical of claims when I bought it, but I can vouch that people aren't messing around when they talk about instant boot and zero-second loads times for applications. Mileage may vary depending on the brand and model, research and watch the specs closely. A paltry 120 gigs by itself is not enough for me or most people these days so I balance things out by installing the OS and applications on the SSD, while most files go onto a hard drive. This means a slight change in workflow, but it is entirely worth it.
Brought to you by Carl's Junior.
The article is mainly talking about from now will 2020. I'd have to agree I don't think it will happen by 2020. I suspect even by 2030 mass storage solutions on HDD will still be around. It took a very long time for HDD to kill off tape entirely.
Flash storage is more than .50$ a dedicated person to use cases. For most people (and manufacturers) are pretty much dumb terminals are talking about to hard disk even though tape is no where the $400 mark. When 2TB ssd on the rest drop in there are pretty much dumb terminals are talking about 50 cents a smaller number of magnitude increase. "Flash storage costs have not in capacity people who must balance density, cost, reliability and store work on, that increases in price as well. I'm not seen any ssd come on these machine, the order of people bought computers with drives have both. a 256 gig on the biggest drives possible at least, and 1TB drives that far exceeded their information. Check the data on ssds. Prices have an external 2TB ssd on the market, you'll see the last two years and you are somewhere between 5-10 cents a bigger number of the big bonuses of the combined servers the same system broadly, but not going to continued high prices for my use SSD. But other uses. Personally for critical files, although I've been almost hitting the rest drop in there was a place that may be addressed by increasing overall costs. At least for the cheapest product in performance as well. I'm not going to slow, and HPC applications due to swap them out are somewhere between 5-10 cents a separate BDR box. Fairly sure where near enough for the biggest drives have an external 2TB ssd.
due to continued high prices for capacity, bandwidth and power
How the hell is power an issue? SSD's consume something around 1/100th of the power that a hard drive does.
Chas - The one, the only.
THANK GOD!!!
SSD will have an artificially inflated cost for a while because if they went and dropped the price of SSD to hard drive prices then no one would really buy that many HD's especially in SAN's and this would cause such an uproar in the industry that create the platter drives. It would practically put most of those businesses under.
It's try HDD will likely continue to get cheaper too. But, the question is not what is the absolute cheapest/GB. The question is, at what point cheaper space no longer a relevant metric?
We've already seen CPU speed increases essentially become irrelevant. We're pretty much there already with disk space as well (for typical consumers). Unless you're ripping or pirating Blue-Rays, running a PVR, or other tasks that include storing uncompressed HD video, what exactly are you going to fill up multiple TB with? Millions of songs? Billions of word processor documents? Thousands of games?
HDD will be around a LOOONG time for tasks that need huge storage (enterprise, video processing, etc), just like tape was around for a LOOONG time before HDD killed it. But, the masses have no real need for that kind of space. The masses get by quite nicely with tablets that have ~16GB of space.
I disagree. Sure, most people don't need an SSD. However, they are really nice. I picked up my first one a few years ago and use it for OS and frequently used software only, my primary storage HDD holds everything else. Having your OS on SSD really does make everything feel much snappier; it isn't just about boot time.
I had the option of going with hybrid. They are great for average consumers but most of us technically inclined people would greatly benefit from being able to control what is on flash and what isn't, so SSD+HDD, or at least a user-configurable hybrid, is a better solution.
Platter drives have been artificially held high for the past few years... and it will burn them unless they start budging on capacity and price, as SSDs will continue to drop.
With 5TB and 6TB drives finally making it out into the consumer space, platter drive pricing may finally start dropping, but will it be too little too late? Will there be enough of a market now in the consumer space to support the larger drives? I suspect the average user has plenty of storage already - perhaps to the point of full porn saturation - but more seriously... how much drive space does Grandma need for her cat pictures and baby videos? 2TB is probably more than enough, and within the year, she'll be able to save all that to a 2TB SSD that boots her e-mail/web browsing machine in an eyeblink.
Of course, the platter drive makers have brought on this trouble themselves... like the DRAM price gouging back in 1994 (The Sumitomo explosion supposedly endangered epoxy resin supplies, prices of RAM tripled overnight), platter drive makers have taken the same opportunity to create a scarcity to drive prices up and keep them up. They also delayed higher capacity platter drives, giving the SSD makers an opportunity to catch up. They can't keep this up and stay in business.
HDD killed tape?
I thought it just pushed it into long term storage.
We need reliable hybrid drives with 120-160+ GBs of flash memory, instead of the ridiculously worthless 4-8 GB ones we have now.
Not really.
ZFS' caching mechanism uses a maximum of half or RAM by default (memory pressure from apps causes it to decrease of course). Given that most servers comes with 32-128 GB of memory, that means Sun/Oracle/ZFS Team generally recommended a maximum of 16-64 GB of a SSD write caches ("ZIL"). And even with those small numbers the boosts you get in performance are quite remarkable. Of course you can also add read cache ("L2ARC") SSDs as well, which will also help.
I think the key though is that the logic is in the file system, and the programmers can use the extra levels of storage more intelligently with the file system strucutures.
The "problem" with hybrid drives is that the caching logic is in the firmware, and there's only so much you can optimize I/O at the block level (as opposed to the object/file//inode level).
What we really need is for hybrid drives to expose the HDD and SSD components separately so that the OS and FS folks can get at them.
I don't see tape being killed off until magnetic density in HDDs hits major diminishing returns. Even though there is only one tape drive maker these days (Quantum with the LTO line), they can keep advancing tape because the media has a lot more area than a HDD platter (or a stack of platters.) An average LTO-6 tape is 846 meters long, and that is a lot of space, even with factoring in the physical contact that the media has to go through.
It would be nice to see a consumer grade tape drive that can run from USB 3 or 3.1, especially if WORM cartridges were available, with media about 1TB native in capacity. Couple this with some decent backup software, and it would come in handy to mitigate data loss. Tape's advantage is that it is inexpensive, easily stored (drop a cartridge, and if there is no physical damage, it will still work), and can be set read-only in hardware.
I've wondered if a HDD maker could make archival grade hard disks, with media that can last 25 years or so. This might require multiple sets of read/write heads (similar to a drive that had two sets and could access different data sets at the same time independantly.) Couple that with a form factor that is easily grippable/manipulable by a robot, and that would replace both VTLs and real tape libraries.
RAID does impact random IOPS quite severly - IOPS are additive. It's just that the difference of the IOPS from one SSD and one HDD is so huge, that it is financially ineffective to try to achieve IOPS of a single SSD with a huge array of HDDs.
Then we need to start to talk about the capacity. A huge array of HDDs will handle much more data than a single SSD.
Enterprise grade HDD can survive many years of enterprise use. SSD drives that can do so are crazy expensive.
In the end it's a mixture of IOPS, seek time, throughput, capacity, MTBF and cost.
Lets be honest here - outside of a small percentage of users doing raw uncompressed video operations HDD are more than fast enough. Drives and OS both offer large caching of high use objects which reduces seek/startup time differences to a very small amount. The biggest difference is on start up and even there.. do those 5, 10, 15 secons extra really matter that much? How often are you booting? Or even resuming from hibernation if thats your thing?
As to power, idle is now around 5 or 6 watts and standby around 1. Even in a laptop the difference in power use between hdd/sdd is not going to make or break the deal. Your screen, however, another story.
I agree that hybrid storage is great, but it can "easily" be done in software (there's a couple of projects for Linux, like bcache, as well as ZFS, and there's an Intel driver in Windows). Then you can pick the size of the SSD and HDD at will, and optionally make a RAIDs of the HDDs and SSDs to mitigate against the increased failure probability.
When multiple drives aren't an option, in laptops, the problem with hybrids is that you lose out on the non-performance advantages of SSD: low power usage and durability. The controllers could improve on this, by shutting down the hard drive and doing more writeback caching, but current hybrids lose on these points. (my laptop has a 256GB SSD, which I find about a factor 2 too small. I can't sync my /home there so everything on it becomes temporary and a syncing chore. Still I wouldn't change it for a bigger hybrid of the current generation, even if there was space for one, due to the decreased mobility).
We already have almost every version of this, hybrid drives for laptops, software techniques that mimic this but they're all fairly stupid and unpredictable, training it to cache the right things take time and suddenly what used to have SSD performance might have been evicted. If you're the kind of user who needs >100GB you probably know what it is taking space. Put your big media (video, photos, music) on D:, everything else like applications and documents stay on C:. The only really tough call is games which often have a huge install size but also app code that benefits from being on an SSD, Steam lets you define multiple library folders so you can have one on C:, one on D: but no easy way to swap them in and out, for now the only supported way is uninstall and reinstall on the other. There are workarounds for that though.
Live today, because you never know what tomorrow brings
The Samsung 640 EVO 240GB has been $138 recently at Amazon ... check the price history at camelcamelcamel: http://camelcamelcamel.com/Sam...
...a more interesting article summary would read "a number of issues will keep HDDs from closing the performance gap with flash for some time". I'm not willing to give a vendor money for more space: I already have too much - and new PC's ship with more GB than an average user will ever fill.
But as a consumer I will pay for more speed: especially since switching from HDD to SSD is a material improvement (as opposed to spending the same upgrade money on a CPU that's 15% faster). And businesses definitely see the benefits, especially in common virtualized environments: where IOPs are precious and expensive to obtain with traditional HDD arrays. We could run a lot more VMs on the same RAM and CPU at work, once we slapped a couple SSDs in the drive trays.
I have a fileserver that will be fed replacement HDDs as they die, but that's it. Every other computer I touch gets nothing but SSDs.
I don't have 2013 figures but 2012 tape was down to $1.06 billion (drives and media) falling at a rate of 30% annually. I'd say the media is dead. The problem is the price isn't going down much. We are still near $.01/g which gets you in the range of bad HDD. BTW most commercial RAID allow for read only.
The issue of longevity is a very big advantage of tape. The problem is that the overwhelming majority of data doesn't need decades of retention. It is just a small fraction that does, in which case that small fraction can just be redundant and duplicated in with whatever other long term storage solution is needed. It is possible that tape might find some tiny niche and stabilize but that's going to be close enough to dead.
Heck, when it comes to long term storage I still think microfiche is a good solution for long term storage. 200 years from now I have no idea if a tape is readable or the data format will be usable but I'm pretty sure people will still have magnifying glasses.
I think the theory behind caching is that what *should* work best is just keeping a list of the most frequently accessed blocks on flash, since, well, that's what you access most frequently. I would be nice to have a config tool that would be able to flag file(s) or directories as "always-cache".
I think the parent is mostly right in that most of the hybrid drives just have too little flash to really provide a lot of meaningful acceleration. 8 GB just doesn't cut it against 750 GB of platter. More flash capacity would also allow you to reserve some meaningful space to cache disk writes.
Mmmmno. Hybrid drives are convenient, I give you that, but they are very limited in what sort of information they have about the data and its uses and if/when a new filesystem format comes around which the drive's firmware doesn't understand the drive falls back to dumb block-based caching. Cache done on OS-side of things have access to things like frequency of use, what sort of situations are the files used, expected ranges of reads and writes in the various aforementioned situations, new, improved filesystems, actual content-type, which user or users are logged on and so on. The hybrid drive, for example, cannot know who is logged in or that the user likes to e.g. listen to certain playlist while doing image-manipulation -- it doesn't know how to predict these situations and preload/cache things accordingly.
No we don't. Hybrid drives are stupid. The added software complexity alone makes them a non-starter for anyone who wants reliability. The disparate failure modes make it a non-starter. The SSD portion of the hybrid drive is way, WAY too small to be useful.
If you care enough to want the performance benefit you either go with a pure SSD (which is what most people do these days), or you have a separate discrete SSD for booting, performace-oriented data, your swap store, and your HDD caching software.
-Matt
Tape is far from dead...
I've got better things to do tonight than die.
Perhaps the biggest use case would be for installing and updating Windows 7, one hell of a pig of an OS. I did it recently on an old laptop, the damn thing took four hours! though much of it was download on slow wifi, also the laptop "only" had 1GB ram but why should I swap just for a desktop, background updates and browsing for drivers. (OS needed to be installed anyway, memory upgrade coming later)
I was pissed to babysit that shit instead of going outside.
I think SSD would be useful just for users who have Windows 7 updates installing, Windows prompts for reboot, need to reboot and let updates install when rebooting. One security risk is users pissed off by updates and disabling them or not letting them through.
No I don't want hybrid drives. I'm on my third year on my OCZ Agility2. In the same time 3 1TB spinning drives have developed bad sectors. I'll keep them separate...
(I like your controller idea, kinda similar to what Intel does on newer chipsets)
I've got better things to do tonight than die.
OK what data do you have. I have for 2012 it was down to $1.06 billion (drives and media) falling at a rate of 30% annually. What are your figures?
In fact, I guarantee if someone made a hard drive with a controller with an mSATA slot for adding a SSD and offered the controller to be setup as pass-through (act as two drives) or caching (SSD keeps a cache of platter), it would sell like crazy.
This already exists. It's called Intel Smart Response Technology and it's been available on motherboards since it was release three years ago in 2011. It works with both mSATA and standard SATA attached drives.
:-(
Most people just haven't paid the extra $100-200 to buy an SSD to cache their HD not to mention the technical know-how required to install the drive and setup the BIOS and software correctly.
What we need is systems that are sold with this already built in so users don't have to do it manually. But that adds $100-200 cost to the system and many shoppers look at the price tag first when they're buying a new computer
1st) Raid is no backup. Many people seem to forget this.
2nd) Tapes rock for linear writing.
3rd) Even if there is a new LTO type you wont be buying new drives for some time as the version you have will last long enough to skip one or two gens nowadays.
4th) Tapes are more resilient than ever and will be reused over and over till one fails or there is need for more space in the rotation.
5th) Data that is worth backing up doesn't increase that fast as some people believe.
The media isn't dead. It is just too freaking reliable.
This is actually a very interesting proposal. While I imagine the engineering and programming would be a relative nightmare*, it would provide a number of options for hard drives.
While it wouldn't double performance in most cases, especially not sequential operations, for random operations it'd be almost as good as two drives. Maybe better if the access is typically really random and one head can 'field' mostly the outer disc calls while the other catches the inner disk ones.
*Just look at the difference between programming a single thread application and multi-threading!
I don't read AC A human right
Having followed this for some time now, one thing I don't get - why do people go for SATA SDDs instead of all the way for PCIe ones? Cost can't be the reason, b'cos the only reason to prefer flash memory to the usual hard disk media is performance. It wouldn't make sense to put a PCIe interface on an HDD, since there is no way the HDD could provide the data at that speed. But flash is different, and can. So it only makes sense to go w/ SATA/PATA HDDs if cost is the issue, and PCIe SDDs if performance is. But I just don't see the point in going w/ SATA SDDs, where one gets all the disadvantages of both SATA and SDDs, and the only advantages of SATA - cost - is probably more than offset by the disadvantage of SDD - which is again cost.
So someone explain to me again - what's the case for going w/ SATA SDDs at all?
Other reason for having the OS on SSD is that one doesn't frequently change the OS data - it's mainly read, except for times one does an OS update. Given that the endurance of SSDs are lower than that of HDDs, it makes sense that something that's less frequently updated would sit on an SSD, while user data, which is frequently updated, would sit on an HDD. Also, the OS is a fixed size, and would typically be something like 16GB. So one could get an SLC SSD - one w/ the highest performance, w/ a PCIe interface, w/ a low enough density to avoid blowing up the cost of the configuration, and just use it for the OS.
Also - how big is the OS itself - in terms of Mb/Gb? Windows 8.1 is overall some 16GB, from what I understand, so how big would the kernel be? Reason I ask is that so far, PCs have had a NOR flash of 4Mb for their BIOS. Given how memory densities have grown, there is the scope here to grow the BIOS flash and contain within it the entire kernel of Linux/*BSD/Windows8.*, and then let the rest of the OS reside on the storage.
That way, the system is more secure, since there are ways a BIOS flash can be protected by hardware (Write Protection states, for one) and other things that fall in Userland can go into an SSD or an HDD, depending on the system configuration. User data, such as movies, can go into the HDDs.
The paper is from Steve Swanson's group at UCSD, *not* Microsoft Research.
And the reasons for slowdown with more bits per cell: (a) writing is done in incremental steps, which have to be smaller for the more precise levels needed for 8 or 16 levels per bit, requiring more steps, and (b) the charge on a flash cell can't be measured directly; instead the chip can measure which cells in a page are above (or below) a threshold voltage, so sensing 16 levels requires 15 separate read operations.
Does anyone else remember when bubble memory was supposed to replace hard drives? There's a long road between the current state of post-NAND technologies (Phase Change Memory, spin-torque-transfer magnetic RAM, Resistive RAM, and a few others) and mass-market high-volume chips. If one of them becomes good enough for someone to risk a $5 billion fab on, and it gives more bits per dollar than flash, then it will probably replace flash almost instantly. If no one bets a cutting edge fab, however, it doesn't matter how promising the technology is. (in particular, the "10x better" is based on assumptions that e.g. PCM can be built in sizes vastly smaller than today's flash - of course we don't know how to build the fabrication plants to do that yet. No one has a story for something 10x better at the same feature size.)
The author of the study makes a lot of arguments based on factors that are easily changed, like the configuration of an SSD. However there are a few basic technological trends:
1. Disks and NAND flash are both getting more dense at fairly comparable speeds - disk has been getting cheap faster than flash lately, but may have a hiccup in the next few years. Where flash has conclusively replaced disk is in applications like iPods and mobile where "enough" storage is cheaper than a single disk. (the iPod went flash when 2GB of flash reached $50, which is the price of a micro-disk) It's not going to replace disk for high volume data storage anytime soon.
2. With today's disks and chips, a hard disk drive has a relatively fixed cost (the cost of the factory amortized over the number of drives produced) and similarly flash has a relatively fixed cost (cost of fabrication plant over the number of chips produced in its useful lifespan). The number of bits on each doesn't really matter - that's why packing them more tightly makes the bits cheaper.
3. Disk bandwidth for 7200K drives isn't going to go over say 300MB/s anytime soon with today's perpendicular recording technology - if the disk is moving past the head at a constant speed, the only way to get more bits through per second is to pack them more closely on the platter. And the best you can do by spinning faster is a factor of 2, at 15K. (and those are very low capacity and very expensive)
2 and 3 mean that flash can easily supply cheaper bandwidth than disk - it's the SSD maker's choice how widely they want to stripe data over the chips in the drive. (64 ways isn't unreasonable) There's a huge advantage today, and it will stay the same (see #2) if flash chips don't get faster, and get bigger if they do. (at some point getting that speed may require paying for more flash than you need, but at that point a single disk will be bigger than you need, too)
For years flash was getting slower and less reliable (requiring more complex error correcting codes) as it got denser - that's partly why it got cheap so much faster than e.g. RAM, where you can't cut those corners. The next generation of flash (3D NAND) may reverse that for a while; in addition SSDs are finally a noticeable fraction of the market so there's an incentive for vendors to make faster flash. (3 years ago SSDs were 3% of the flash market, and the rest went into iPods, phones, and removable drives and cards - SSD vendors had to make do with flash that was designed for systems where you don't care about performance)
1) Of course a RAID can be used as a backup. RAID doesn't back itself up itself up but that's a different question.
2) How are tapes any better for linear writing than HDD?
3) Certainly tape drives can go a long time.
4) HDD can be used over and over as well.
5) If the volume of data is small the cost factor goes down which kills the only advantage of tape (other than longevity).
What a load of lies that article is. And by the way, I am not giving away my new flash technology!
I've been doing backups for around a quarter century. So I know the issues is all I'm saying, not trying to put on airs.
Certainly the gold standard for backup protection is a separation between the backup and the thing being backed up. However to hit the gold standard you need off-site storage of your backup. And why do you really need the gold standard? Well as a practical matter, it's mainly:
1). Complete destruction of the source system, due to fire/flood/earthquake/whatever;
2). Theft.
And traditionally, off-site backup is a pain in the consumer space. Cloud backup solutions are pretty good answer here.
Are these really the primary reasons for backup though? Ever had a drive fail? Ever accidentally delete some data and be unable to get it back except via the backup? Those sorts of things are far more common in practice.
So what does in-system backup buy you?
1). Blazing speed, especially if you can stay on the SATA (or equivalent) bus;
2). Killer reliability of the backup resource, which means it can easily be automated. And automation is the key to always having a current backup.
So disk-to-disk backup in a single system should be thought of, not as a replacement for off-site backup, but as additional to it. That's what I recommend.
Hybrid drives are stupid
Apple seems to disagree with you.
8 of 13 people found this answer helpful. Did you?
I can purchase a 500 Gig SSD for less than a 500 Gig Hd as of 4 or 5 years ago. That's plenty of high speed room for the OS, Program, and mail.
Well, if we followed Apple's lead we'd all have one button on our mice.
Am I then only one that sees that price gouging is going on with SSDs?
I bought my Corsiar 115gb SSD for $69.99 in July of 2012. That's almost 2 years ago. Today a generic 120gb SSD costs $69.99 from newegg. 5 more gb, same price 2 years later. If I want a Corsiar SSD again, a 120gb one costs $109, $40 *more* than what I paid 2 years ago!
More manufactures make SSDs now and more devices use SSD now, but you're going to tell me years later the prices haven't dropped?
Come on, give me a break, obvious price fixing is going on with SSD prices, I can't be the only one that sees this, when the was last time that prices on computer hardware went up years later? Never? This is just like the one billion dollar LCD price fixing scheme a few years ago, I'm sure this will be in the news a few years from now.
my karma will be here long after I'm gone
When multiple drives aren't an option, in laptops
I've just recently decided to upgrade to an SSD in my laptop, but I need quite a bit of space and didn't want to break the bank on a large drive.
There's the option to do away with the DVD drive, which frankly I've used a total of about 3 times in thid laptop's 3-year life, and get an adapter for a 2.5'' drive.
I see this as a win-win: you get a fast drive for your system and a large one for bulk storage while also getting to remove a relatively heavy component from your laptop.
Look for "drive caddy" on Amazon or so. They're usually around the 20$ mark.