Seagate May Sue if Solid State Disks Get Popular 242
tero writes "Even though Seagate has announced it will be offering SSD disks of its own in 2008, their CEO Bill Watkins seems to be sending out mixed signals in a recent Fortune interview 'He's convinced, he confides, that SSD makers like Samsung and Intel (INTC) are violating Seagate's patents. (An Intel spokeswoman says the company doesn't comment on speculation.) Seagate and Western Digital (WDC), two of the major hard drive makers, have patents that deal with many of the ways a storage device communicates with a computer, Watkins says. It stands to reason that sooner or later, Seagate will sue — particularly if it looks like SSDs could become a real threat.'"
holy cats! the world is changing! (Score:5, Funny)
Re:holy cats! the world is changing! (Score:5, Insightful)
Re:holy cats! the world is changing! (Score:5, Funny)
(Slowly I put the freshly printed page down...)
Re: (Score:2)
Re: (Score:2)
Re: (Score:3, Informative)
To be fair, it seems like their plan is to sue to get a piece of it, via technology that they really did create. It wouldn't be very profitable just to stop progress.
They, Seagate and Western Digital, can get a piece of it by releasing their own flash drives instead of suing others.
FalconRe:holy cats! the world is changing! (Score:5, Insightful)
Either they are violating your patents (sue), or they're not (don't sue).
You dont get to sit there and wait and wait until they make gobs of money in case 1.
So seagate, are they violating your patent? If so, proof please, if not, you yield all rights in case they are found to at a later date
Confusion (Score:5, Informative)
Re:Confusion (Score:5, Interesting)
Re: (Score:3, Insightful)
Re: (Score:2)
Re:Confusion (Score:5, Insightful)
If someone is violating your patent you need to sue now. Suing later just makes your job much, much harder. Especailly with the CEO of your company saying things like were said in the article.
Re: (Score:2)
One thing to consider if you become aware that someone is infringing your patent is the doctrine of laches. Under this doctrine, if you wait too long to sue for infringement after becoming aware of the infringement, and the accused infringer suffered material prejudice due to your delay, then the accused infringer may have a defense to the charges of infringement.
(Pienkos, JT. The Patent Guidebook p. 86)
Re: (Score:3, Interesting)
Re:holy cats! the world is changing! (Score:4, Interesting)
Re:holy cats! the world is changing! (Score:5, Insightful)
Damages start to accrue when the infringer is "put on notice" by the patent-holder. This can be done one of several ways, if you are the patent-holder:
1. Send a notice by mail (registered of course) to the manufacturer citing the patent(s) you hold, and claiming that they are infringing. At this point you could offer to license your patent, tell them to cease production and stop selling the item, or demand a settlement offer (and still shut them down or license the patent).
2. Mark your own products with the granted patent numbers. Usually something like "This product is covered by US patents 5,205,321 and 5,255,555." Although some manufacturers just put a list of patent numbers only on the product. Sometimes you will see a product that says "Patent pending." This does nothing. It might scare folks out of infringing, but I could write that and have no patent at all. No damages would accrue, and that is not "putting them on notice."
3. If the manufacturer inadvertently discovers that they are infringing the patent, then the damages begin accruing then. (Of course, proving that they knew when they knew is another matter entirely).
But you could really screw a manufacturer over by telling them about this article and showing them the patents. Essentially you would be doing the work of "putting them on notice" for Seagate.
Many manufacturers have several strategies for this problem:
1. On release of the product, they do a "patent scrub." This allows the inventors to file patents protecting this product. During the patent scrub, they are asked about any prior art or other patents that would limit the claims for this product. Often, though, during the prior art search, a patent claiming the exact product is found. Now the manufacturer must make a decision to either a) stop manufacturing/using/selling, b) risk infringement (with treble damages), c) contact the inventor for possible licensing.
2. Purposely don't do a prior art search. This way you won't turn up any patents accidentally. Sometimes it might be a good idea to save several dollars per sale in an escrow account in case of patent infringement (to easily pay off the patent holder). Then hope there is no product out there that has been patent-marked. If the manufacturer is contacted and "put on notice" by a patent holder, then the manufacturer can stop production immediately without any damages.
Often, however, a manufacturer could go a complete product cycle while unknowingly infringing a patent -- and never get a notice. If they make it all the way through, and then the patent-holder comes later claiming infringement, it is too-bad-so-sad. The manufacturer isn't making or selling that product anymore.
That's all I know. IAAPP.
Re: (Score:2)
Indeed. I would think they could apply to concept of laches [wikipedia.org] to defeat any claims of infringement. Of course, IANAL, so I don't know why this is not invoked.
Who was first? (Score:4, Interesting)
Is it not possible that someone at Samsung came up with the idea before Seagate, but just didn't patent it? Or we could go by the saying: "Ideas are cheap." Just because you dreamed up an invention, why should you get some of the money for all the work put into implementation, marketing, manufacturing, etc?
To answer my own question, I suppose it's because otherwise, no one would report their ideas without a working model and/or contract with a production company in place. They'd never be able to make any money off it as it would be used by someone else if made known. I won't go on about how I feel about the mighty dollar/euro/rupee and how it stifles innovation...
So what about solid state disks 20 yrs ago? (Score:4, Informative)
patents but by now these will be expired. (They used the rejects from memory fabs, which they
called "the skim milk of the crop", and worked around all the bad bits to get usable memory that
was cheap enough to use.) Certainly one can use similar techniques to theirs (likely today with
better memory) and make solid state disks. No way Seagate or anyone else could patent that (once the
old technology was pointed out).
Re:Who was first? (Score:4, Insightful)
Patents used to be for specific implementations, not the ideas behind it wholesale as no one back then seriously thought you should have a monopoly or could even own an entire idea.
Re: (Score:2, Informative)
If I remember correctly, under the US patent system, the first one to come up with the idea gets the patent. In some other places, first one to file gets the patent.
So, I believe if you keep something a trade secret and someone else tries to patent that technology, you can acquire the patent by demonstrating you had the idea first.
Re: (Score:3)
Inventors (aka company engineers) typically ARE NOT supposed to research patents when inventing. That generally creates too much liability for the company if they were to find something interesting. Most companies have a legal department designed to handle patent searching anyway. They are usually technically-minded people who are also in the legal field. As a former IP paralegal, I really would rather the engineer NOT do patent searches unless asked.
6 Billion people, but how
Re: (Score:3, Interesting)
Is it not possible that someone at Samsung came up with the idea before Seagate, but just didn't patent it?
If that's true, and Samsung could prove it, then Samsung could and would be awarded the patent instead. US law awards patents to "first to invent", not "first to file".
This is incidentally in the works to be changed. The rest of the world uses "first to file" and it is claimed that the change would reduce the paperwork required to approve patents. (Like we need more patents, sheesh.)
Doesn't even cover what they could sue over (Score:5, Insightful)
Yeah, personally I'd like to see some actual specific patents rather than a CEO full of hot air making baseless threats. I'm sure Seagate has patents on storage device communication, but this article offers no insight on how SSD makers could be infringing. This is like the crazy patent claims Microsoft made against Linux (what was that? 184 alleged patents? More?) Examples would be nice.
Anyhow, flash prices may be dropping, but I don't see SSDs gaining majority marketshare within the next 5 years. Developers get lazy, cameras get more mega pixels, more people need digital video. Spinning disks are still massively cheaper per GB than SSDs, and unless the price were to drop dramatically, hard disks will still have the edge to keep the throne. Laptops may see SSDs sooner due to power, but I'd imagine that one way to forestall the inevitable victory of SSD would be more intelligent caching and a larger onboard cache for hard drives.
Anyhow, Seagate is worrying about market dominance, and the Seagate CEO makes vague threats that the lawyers at Intel and Samsung probably laughed off. Not that newsworthy in my opinion. Specific patents or litigation would be very notable though.
Re:Doesn't even cover what they could sue over (Score:5, Interesting)
Re:Doesn't even cover what they could sue over (Score:4, Insightful)
Re: (Score:2)
Re:Doesn't even cover what they could sue over (Score:4, Insightful)
Re: (Score:2)
Regular hard drives are magnetic in how they store their data.
Re:Doesn't even cover what they could sue over (Score:5, Insightful)
Wouldn't it look good on the quarterly if the stock price spiked toward the end?
And yes, likely Seagate and WD have a lot of IP that they have inter licensed or at least have an informal non-aggression pact about suing each other over (kinda like amd and intel), whether any of it is current in these days of industry wide standards like SATA, SAS, SMART etc is another thing entirely.
When it gets to the courts, THATS when it will be worth reporting on.
Re: (Score:2)
Actually I'd suggest they play their strengths rather than their weakness. Disks are never going to compete with flash on seek times. They'd be better off dumping the entire 'speed' thing to flash and moving backwards to slower rotational speeds and vastly larger platter area. Can you imagine 5 1/4 inch disks with todays data density?
Re: (Score:3, Insightful)
This sounds like the mythical "near line storage" that tapes ceased to be years ago... A 5.25 full hight 3200rpm drive would be reasonably cheap, very big, quiet and power
Re: (Score:2)
Re: (Score:3, Insightful)
I'd have to argue against your statement though for NEAR LINE storage. For real time DB/webserver/boot drive no. For holding an archive of my gigaton of DVDs? I'd go apeshit for a 5TB drive that's quiet, inexpensive and could transfer on the order or 30-40MB/sec. Raid 1 for safety and sanity and you're golden. I don't need 100+MB/sec for watching movies. Vaugely intelligent caching will easily allow multiple read/write streams for HD content within that rough data bandwid
Re:Doesn't even cover what they could sue over (Score:4, Interesting)
Exactly the kind of use I had in mind. The huge but very linear datasets where track jumping is minimal that tend to make up the vast bulk of personal storage these days and that will only increase.
OS, software, databases, etc, need low latency to be 'fast'. Bulk multimedia storage doesn't.
I'd go apeshit for a 5TB drive that's quiet, inexpensive and could transfer on the order or 30-40MB/sec.
A 5-10 TB drive would probably be doable today with existing surface density. I dont think you'd lose that much on the transfer rate tho; data rate depends on the surface velocity which remains almost as high due to the increased circumference.
And, yes, I had a bigfoot too. It's not something you want to put your OS on; then again, as noted, flash might be the best thing to put that on whatever you do to disk speed. Personally I have most my OS cached in actual RAM on my workstations in combination with the RAM on my iSCSI storage servers. But it's not that part that causes most issues; it's finding actual space for storing the ever increasing amount of mythtv recordings...
Re: (Score:3, Interesting)
Re: (Score:2)
I know this is SORT OF off topic, and I am by no means on Microsoft's side in their Patent-War-On-Linux, but I CAN give you an example (albeit stupid):
Microsoft's patent on Long File Names on FAT/FAT32. Like it or not, it is there. To make matters worse, the Official Microsoft spec [washington.edu] has bunches of code snippets that it seems a lot of developers never care to rewrite. They ev
Re: (Score:2, Interesting)
... I'd imagine that one way to forestall the inevitable victory of SSD would be more intelligent caching and a larger onboard cache for hard drives...
Oh, so the way to fight these IC devices is by adding more IC devices to the spinning disks? Oh, I know! We can slowly phase out the platters, and go to a fully IC device! That'll keep us from going to a fully IC device!
Sorry, that fails the logic test. Seems to me that spinning platters are on their way out. Welcome to the solid state world.
Re: (Score:2)
SSDs are fast already, and getting faster quicker. Except for the unfortunate problem with wear that wear-leveling isn't really solving as well as some would have you believe, an SSD with suitable capacity for a majority of users - especially business users - at an acceptable price with much higher performance may easily be here within a year.
And if they could solve the write wear problem wit
Re: (Score:3, Informative)
Wear leveling works. Period. Do your homework and check back. FYI, a SSD with an acceptible price point (for business/enthusiast/high-end users) and high performance IS HERE NOW.
I suggest you look at the access times and data transfer rates (especially for small, random reads) on some of the new 64GB-256MG drives floating around. They're a bit shy of the *top end* magnetic disks in linear read but completly BURY them in random read (which constitutes the vast majori
Re:Doesn't even cover what they could sue over (Score:4, Informative)
Not really true. Seagate's Cheetah 15K.5 300GB, being (last I checked) the fastest magnetic media you can buy, can easily beat that. It peaks at 135MB/sec. Some other 15,000 rpm drives can post comparable numbers.
I still agree with your post in general, but that specific statement is untrue.
Re:Doesn't even cover what they could sue over (Score:5, Interesting)
I think you miss the best reason for SSD on laptops. That means with the exception of my fans, the laptop is all solid state. Which means I don't worry as much about moving it around while reading data, or gyroscopic forces. I worry about these things because I don't know excatly how they work, but I 'm sure shaking a spinning disk based drive is bad. And if I don't know, than I will arrogantly (but accurately) say that most people don't know. Hence, selling point.
Re:Doesn't even cover what they could sue over (Score:5, Interesting)
I think its all hot air or at least is trying to gauge Intel's and Samsung's reaction because he's not threatening some small time business here. Intel probaly had a large team of persons compiling new patents on a daily basis and if push came to shove in court Intel would counter with their own set of patents along with Samsung's team and then Toshiba and IBM might jump in and the proverbial MAD (Mutual Assured Destruction) ICBM's filled with patent experts will be landing lawsuits left and right at each others door until only the lawyers are left standing.
So no... Seagate would never want to actually go through with it because they have no idea what patents Intel might have that they could claim that Seagate is currently violating. My hunch is Seagate wants to calm investor fears in the current technology.
After all, even if Seagate won the war, Intel might just start making mother boards or CPU that have lines of code that say:
if $HDD_Manufacture = Seagate { do not boot & give error message "Faulty HDD! contact OEM" }
Hell no! (Score:4, Insightful)
I am dying for an SSD. It is silent, and rigid, and uses less power. And newer drives are guaranteed to be up to 4 times faster than the current "standard" SSD drives that Dell and Apple put in their laptops (200Mb/s versus 50Mb/s read). If you had the money, you would have no reason NOT to move to SSD, especially if you have a laptop, and more so if you use your computer for work.
I agree that the low-end computer market will consist of HDDs for a long time to come. HDDs will not go away as long as they provide cost effectiveness. However, once the next generation drives are out and hit the 5USD/GB mark everyone with a buck will want one especially when their IT friends will be all over them.
Re: (Score:3, Interesting)
I wouldn't be so sure with that. Harddrives are mechanically complicated devices, and that establishes a floor for harddrive prices, as it still costs a certain amount to create all the mechanic
patents and copyrights (Score:5, Interesting)
and now they are used as perverse tools to squash progress, stifle innovation, and make us culturally impoverished
not that any of this means there will be a social revolution, but i see the real possibility of a legal revolution. that is, the public simply ignoring the bullshit intellectual property lawyers invent in order to justify their existence
dear intellectual property lawyers: you suck. your entire field is becoming a farce. you write and interpret and enforce law that does not serve society, it only serves your field. i propose a mutiny and jettison of the whole lot of you useless parasites
Can you sue about a "No-duh" idea? (Score:3, Interesting)
Re: (Score:2, Interesting)
Re: (Score:2)
Re:Can you sue about a "No-duh" idea? (Score:5, Interesting)
Have you ever noticed how a lot of new cars look pretty much the same? Hard drive interfaces are a lot like that if you allow anyone to patent minor differences. Oh, but hard drive interface ABC uses blue connectors instead of grey. Minor differences and logical extensions of existing patented inventions are things that make for bs patents. SSD is an enhancement to existing technology, NOT some "OMG, how did they think of that" technology. Back in '92 people were talking about that.
Logical extension of existing things: email on wireless devices, methods to store data on a computer, using RAM to replace the mechanical magnetic materials in hard drives, and on and on
Innovation... (Score:5, Insightful)
Re:Innovation... (Score:4, Insightful)
Give me the f*cking address (Score:5, Funny)
Seagate sold me 4 of their stupid Maxtorgates (drives from the former maxtor factory) that failed immediately - and most of the replacements ALSO failed immediately.
The won't give me the address to send them the legal notice so I can sue their sorry asses off.
Their Indian tech support said "we can't do that!"
Get me the address - I want to sue on behalf of everyone who bought brand-new drives and had to pay the shipping to get replacements.
Re:Give me the f*cking address (Score:5, Informative)
Anyhow, the address is: 920 Disc Dr
Scotts Valley, CA 95066-4544
Disc Drive. Ugh.
Re:Give me the f*cking address (Score:5, Funny)
Re: (Score:3, Informative)
Re:Give me the f*cking address (Score:5, Insightful)
And if not giving you my address is all it took to keep you from suing me, then I wouldn't give you my address either.
Re: (Score:3, Informative)
C T CORPORATION SYSTEM
818 WEST SEVENTH ST
LOS ANGELES, CA 90017
Looking up CT's record shows this as *their* agent for service:
JERE KEPRIOS
C/O C T CORPORAITON SYSTEM
818 W. SEVENTH STREET
LOS ANGELES, CA 90017
I'm pretty sure every state keeps their corporate records online, and these records will always have the designated agent for service given. I don't usually even bother with a company's web site anymore when
Re: (Score:2)
Look elsewhere... (Score:2)
Thermal issues may come into play here.
If the firmware/serial numbers were near the same than it was probably a batch/fab issue.
(Former HD recover tech that had to deal with pallets of failed Maxtor drives.)
Re: (Score:2)
Is our patent system now unconstitutional? (Score:5, Interesting)
"To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries;"
Since we're now getting companies suing to prevent advances in the useful arts using powers granted through patent legislation, can we now find that contemporary patent law is a violation of both the letter and spirit of the Constitution?
And, yes, I'm aware that there's a lot of other stuff going on our federal govt. that's probably a violation of the letter and spirit of the constitution.
Re: (Score:3, Interesting)
It's not unconstitutional because patents are granted to the inventors. You'll never see a business name in the inventor field of a patent--as, compared to say, the rest of the world where entities are regularly listed as the inventor.
Instead, patents, and in some senses a patent application, are treated like any other piece of personal property. That means that the owner can sell all or some part of the rights in that patent. Companies end up with ownership
Re: (Score:2)
Since we're now getting companies suing to prevent advances in the useful arts using powers granted through patent legislation, can we now find that contemporary patent law is a violation of both the letter and spirit of the Constitution?
They are not threatening to sue to prevent advances. They are threatening to sue those who would commercialize on advances documented in their patents without paying them the royalties on those patents. Without patents all hardware would be commoditized. And that would mean the lowest quality players (the cut-throats) would always be the standard setters. Don't confuse business-method patents with patents on actual engineering innovations.
Re: (Score:2)
What seems to be riling people up is the legal concept of reduction to practice, i.e., the point at which an invention has been deemed sufficiently completed, constructed, or created to warrant patent protection. Currently
Poor USAians... (Score:4, Interesting)
Now I am pro-patents, but software does not belong in the patent world. This could be the ultimate example of how patents stop innovation and technical progress.
By the way, I wonder how a SSD hard disk is really different from a standard memory card.
They won't of course (Score:3, Insightful)
For Seagate, the issue is to a certain extent
Bad moderation (Score:2)
"A patent is really a license to be sued" (Score:2, Insightful)
Make love, not war (Score:5, Interesting)
Re: (Score:2, Funny)
Re: (Score:2)
Comment removed (Score:4, Insightful)
Defend it or lose it ... (Score:2)
But, seriously
Re: (Score:2)
Eh? You're thinking of trademarks.
Re: (Score:2)
Do patents not have any such obligation? Surely there can't be a blanket opportunity to allow company to produce a product, wait until it gets successful, and then sue to say that someone is infringing.
Well, maybe not surely with the state of IP laws nowadays.
The sheer idea that they can basically say that if someone else's product becomes successful we will then start suing is quite offensive to the sensibilities -- something about that just seems plain wrong.
Cheers
Innovation (Score:2, Funny)
How much could they get? (Score:5, Funny)
Re: (Score:2)
Is that how the IRAM-2 died? (Score:5, Interesting)
The next generation of IRAM [techpowerup.com] fixed my major pain point - allowing dirt cheap DDR2 RAM and allowing 8G max storage per drive.
This is what irks me... (Score:5, Insightful)
The "practice" has been perfectly legal and... (Score:2, Informative)
And BTW, Gibson's patent from 1999 does actually seem to cover the Guitar Hero game's "system and method of a simulated musical performance". Blame the USPTO and the current patent laws, not Gibson here, because Gibson is following the patent law pretty
I *DARE* them to sue Intel or Samsung (Score:5, Insightful)
I imagine that Seagate is violating some Intel and/or Samsung patents, in one obscure and stupid way or another. Seriously, Seagate doesn't have the juice to take on those 2 companies. Never mind that if Seagate really decides to start some shit with their hard-drive patents, I imagine that IBM will get involved, since they own most of the patents on the basic technology of hard drives. And we all know how IBM deals with people that sue them- they take no prisoners.
Re:I *DARE* them to sue Intel or Samsung (Score:5, Insightful)
And, therein lies the rub with patents.
The big players can basically play "Mutually Assured Destruction" and reach agreements where they don't sue one another. There can be no little players, and no new entrants without a huge barrier.
How does this foster innovation and moving the state of the art forward?
Cheers
Re: (Score:3, Informative)
Yes, they might try to do what they can to stop the little guy, but not by legal-flooding him. That is mythological.
Usually a big company would rather just settle and make the little guy go away. If it's a reasonable enough request, they usually do it. If not, then they will litigate. Lit
Where is Intel? (Score:5, Insightful)
Re: (Score:2)
Because USB drive interfaces are just SATA or IDE interfaces with a USB bridge.
I wish I could boycott Seagate for this (Score:2)
... but I am already boycotting them for selling products that fail to communicate properly with the computer. Well, it's more like buying something else (WDC) that works better (especially in Linux). It's too bad that this abuse of the patent system leads to high costs of improving on the mistakes, screw ups, and often incompetency of others.
Re: (Score:2)
WD in my opinion has always been "bottom of the barrel". I don't know if it's just a problem when they get imported into Canada or what, but I always have them die much to often. I've since switched to Seagate exclusively as well. No problems. Last computer was a 120GB drive, lasted 6 years no problems ever. Passed it on to a community house in my area for free after I bought myself a new system.
I feel that if this Seagate patent thing goes too far, I am going to switch to Samsung. I have heard gr
Toshiba preps 128GB solid-state notebook drive (Score:4, Interesting)
So...
There's also an issue related to ROI.
Fun with Laws (Score:2)
Nothing to see here... At least not yet... (Score:2)
Re: (Score:2)
And what if it turns out that they do have a legitimate claim on patent infringement?
The odds of this are exactly 0. Zero. There is nothing that could be done in a SSD that would be unique enough and have cost Seagate enough money to research that would justify a patent. If the patent is something ridiculously obvious, then indeed I'm sure there is something in every storage device that uses it. Big deal, unless you think "legitimate" means "whatever the lazy patent examiners allowed to be patented".
Sad (Score:3, Interesting)
If we fail to keep our heads above the water by making good products, we'll sue.
And this time, it's not just some slashdotter seeing ghosts where there may not be any: this is straight from the horse's mouth.
If they think their competitors should not be using certain technologies, they should negotiate with them to come to acceptable agreements. If that fails, they can sue. Threatening to sue if their competitors' products become successful is...evil.
Re: (Score:2)
Now, some other manufacturers who haven't participated in this development are able to swoop in at the last minute and take advantage of the latest developments. No R&D spending was necessary. Someone else did the work
Patent law needs a trademark-like defense addendum (Score:2)
The enemy of my enemy... (Score:5, Insightful)
The enemy of my enemy ... is my friend.
If Seagate pushes hard enough, they may find this out the hard way. They may be an 800lb gorilla in the storage market. But, even a large ape does not like getting stung by a thousand bees, and Seagate is waving a stick around a number of bee hives, in my view of things.
What if, faced with a potential lawsuit from Seagate, we were to see Intel, Samsung, TI, etc., get together and develop a new standard that bypasses Seagate's IP. They could license it to each other for next to nothing... except to Seagate... no soup for you. Sure, they'd like to be safe from a backlash in the spinning media world. But, given the rapid price drops on SSD storage, at some point the SSD media will be "cheap enough" for primary storage and spinning media would be relegated to 2nd tier, archival storage. Intel certainly has the smarts and the fabrication facilities to develop a competitor to anything Seagate might come up with.
Here's an honest question I've been wondering about for a while. Why don't we use GigE or 10GigE to communicate with storage? I imagine there's more overhead than with the currently used protocols, but how much are we talking about here? I'm more of a software than hardware guy, though I know a little about the different layers in the ISO model. *waves hands*. Build in a router on the motherboard, have a port for talking to the outside world and a few ports for talking to storage. Economy of scale and the hardware would be dirt cheap... right? Since it seems like an obvious idea, I'm sure I'm missing something. Would someone who knows these things care to elaborate? Tnx!
Re: (Score:3, Informative)
Here's an honest question I've been wondering about for a while. Why don't we use GigE or 10GigE to communicate with storage? I imagine there's more overhead than with the currently used protocols, but how much are we talking about here? I'm more of a software than hardware guy, though I know a little about the different layers in the ISO model. *waves hands*. Build in a router on the motherboard, have a port for talking to the outside world and a few ports for talking to storage. Economy of scale and the hardware would be dirt cheap... right? Since it seems like an obvious idea, I'm sure I'm missing something. Would someone who knows these things care to elaborate? Tnx!
Because the hard drives (and SSDs) throughput is already dwarfed by current wire speeds. The best hard drive you can get can push 1 Gbps (that's bits per second) while SATA and SAS can deliver 3 Gbps. Parallel SCSI can deliver 3.6 Gbps. And when you really need a SAN, there are already iSCSI and fibre channel.
Is This Legal? (Score:3, Insightful)
Comment removed (Score:4, Interesting)