Intel 160 GB ssd drive now $425
in [Storage]
|
From: Mark F on 12 Aug 2010 10:34 On Wed, 11 Aug 2010 11:29:59 -0700, DevilsPGD <Still-Just-A-Rat-In-A-Cage(a)crazyhat.net> wrote: > In message <i3uci3$fdt$1(a)news.eternal-september.org> "Percival P. > Cassidy" <Nobody(a)NotMyISP.com> was claimed to have wrote: > > >On 08/05/10 01:32 pm, Timothy Daniels wrote: > > > >>> Intel 160 GB ssd drive now $425: > >>> http://www.newegg.com/Product/Product.aspx?Item=N82E16820167017 > >>> > >>> OK, this is getting very close to my buy point. Very close. > > > >> Personally, I'd get two (one for the page file). :-) > > > >It was suggested to me recently that, because of the finite number of > >writes that SSDs will survive, the only thing for which they are really > >suitable is something that will seldom be written but frequently read -- > >i.e., the OS itself. > > You probably should stop listening to whoever made that suggestion. > Modern SSDs have write cycles measured in the millions of writes. > > http://www.storagesearch.com/ssdmyths-endurance.html gives a good > breakdown, but to save you some reading: > > | 2 million (write endurance) x 64G (capacity) divided by 80M bytes / sec gives the endurance limited life in seconds. > > Or to phrase that in a form that Google Calculator or Wolfram|Alpha can > understand: > > 2 million * 64GB / (80MB/s) I think this number is off by a factor of 20 for SLC, by 200 or more for MLC - see my entries at the end. > > <http://www.google.ca/#hl=en&safe=off&q=2+million+*+64GB+%2F+(80MB%2Fs)> > <http://www.wolframalpha.com/input/?i=2+million+*+64GB+%2F+%2880MB%2Fs%29> > > Assuming you write to your SSD 24/7 continually, without pause for any > reason, and without taking time to wipe into consideration, you'll want > to start stressing out in 40 years or so. 2 years for SLC, a month or so for MLC. However, for home computers it is unlikely that the disk will be written to continuously, so perhaps 20 years for SLC, a couple of years for MLC, even if heavily used. Thus, I agree that in real life the limited write lifetime is not a problem with the current generation of SSDs. > > In practice, most SSDs have some form of wear leveling in play too, > either out of a desire to wear level or alternatively just a desire to > optimize writes to avoid having to flush blocks prematurely due to the > fact that blocks can only be flushed in larger chunks. > > SSDs have their issues (mainly in the $/GB area, although the sector > alignment issue is a bit annoying on out of date operating systems, but > you can work around it), but write cycles isn't one of them unless you > have very early generation gear, or very very cheap parts. > > In practice, most SSDs that I've seriously considered have had 3-5 year > warranties, which probably covers their useful life anyway. From: http://www.sand{remove this}isk.com/business-solutions/ssd/faqs: Q: Can I trust my data to SanD isk� SSD G4? A: SanD isk SSD G4 has no moving parts, thus it is engineered to endure shocks and vibration adding to its reliability in stressful operating conditions so your files are less likely to be lost due to disk drive failure. With an MTBF of 2 million hours, SanD isk� SSD G4 is rugged and reliable. In addition, SanD isk SSD G4 can support/withstand 160 terabytes written. The SanD isk SSD G4 has 64GB, 128GB, and 256GB models. So, they say: (160x1E12 bytes written)/(64x1E9 bytes capacity) = 160/64 x1E3 or about 3000 cycles Did I do something wrong? NOTES: 0. I picked San Disk because they had their numbers in an easy to find location. In tel and several others gave the numbers in Note 1 below, but I couldn't find the references today. I have put spaces in the company names so that this article won't be used to say that the devices of the companies that I mentioned have lower performance endurance that those of companies I don't mention. 1. I did some web searches around 2010 January. From those I thought that the lifetimes were about 100K cycles for SLC. This was about 1/10 of what the lifetimes were around 2000. From the searches at the same time I thought that the MLC lifetimes were about 10K. This would be consistent with the SanD isk statement that you can expect 3000 cycles in a real environment. (Although 700 cycles for the 256GB model seems conservative.) 2. Remember that 2000000 hours MTBF doesn't mean you can expect any given part to last 200 years even under the best power-on conditions. 3. Remember that the data fades with time, so, perhaps 5 years is how long you can expect data to be maintained in a powered off device, not 100 years. 4. I don't know where the 2000000 cycle number came from. The Wikipedia article http://en.wikipedia.org/wiki/Solid_state_disk says: "Flash-memory drives have limited lifetimes and will often wear out after 1,000,000 to 2,000,000 P/E cycles (1,000 to 10,000 per cell) for MLC, and up to 5,000,000 P/E cycles (100,000 per cell) for SLC" I assume "P/E" means "Program/Erase". I don't understand why this is different than the per cell P/E number. The per cell numbers match the numbers I found in Note 1 and are consist ant with the SanD isk reference that I give.
From: DevilsPGD on 12 Aug 2010 16:00 In message <8chfroF6evU1(a)mid.individual.net> "Rod Speed" <rod.speed.aaa(a)gmail.com> was claimed to have wrote: >DevilsPGD wrote >> Rod Speed <rod.speed.aaa(a)gmail.com> wrote >>> Timothy Daniels wrote >>>> Lynn McGuire wrote > >>>>> Intel 160 GB ssd drive now $425: >>>>> http://www.newegg.com/Product/Product.aspx?Item=N82E16820167017 > >>>>> OK, this is getting very close to my buy point. Very close. > >>>> Personally, I'd get two (one for the page file). :-) > >>> Thats the last thing you should put on one unless you plan on replacing it often. > >> Why? > >Because its used more than the rest of the drive for writes, in most real world situations. It is? I thought you're the one arguing for sufficient RAM so that the page file is rarely used? >> The pagefile is a perfect example of something >> that can and should be placed on a modern SSD. > >Wrong, it makes a lot more sense to have more physical system ram instead >so the page file doesnt get used at all. MUCH cheaper and much faster too. It generally makes financial sense to buy an appropriate amount of RAM for normal operating conditions plus a small buffer, and to use a page file to avoid out-of-memory crashes when exceptional circumstances occur. Buying additional memory for infrequent events probably isn't worth the money, especially if we end up needing more than just RAM (more on this below) >> Pagefile performance is critical when you're in a low-memory situation, > >So it makes a lot more sense to not have a low memory situation instead. Agreed -- In which case you don't have to worry about writes to flash since the pagefile isn't constantly being written. >> and an SSD will help speed things up sigificantly. > >Enough system ram so the page file doesnt get used will speed it up >MUCH more for a lot less money. At a significantly higher cost, yes. >>> And it makes a lot more sense to replace the use >>> of the page file with more system ram anyway. > >> True, to a point. > >Its always true unless it isnt possible to add more system ram for some reason. That's the "to a point" right there, yup. >> If you've maxed out your hardware's capabilities > >You hardly ever are in that situation and it makes a lot more >sense to replace the hardware so it doesnt have that situation >than to go for an expensive SSD for the page file anyway, because >enough system ram so the page file doesnt get used costs a lot >less even if the motherboard has to be changed to allow that. I'd question your definition of "Hardly ever" In my world, systems with chipset driven 2GB limitations aren't all that uncommon, and CPU/OS architecture limits at ~3.2-3.5GB are prevalent. When it comes down to a case of replacing an entire system (power supply(1), motherboard, CPU, RAM, and possibly video card(2), printers and scanners(3)) vs a hard drive for a system that is otherwise adequate, the choice should be a no-brainer in most cases. 1) Looking around the office, I've got plenty of systems with power supplies inadequate for a new build, usually a lack of additional connectors. I've still got a few systems in service that have 20-pin PSUs, predating both the 24-pin connector and P4 connector. 2) AGP video cards will generally need to be replaced. 3) Many perfectly serviceable printers and scanners have no x64 drivers at all, requiring these otherwise functional devices get replaced when moving to a x64 environment. >> or only need to catch rare/occasional edge cases, > >Then the speed isnt a problem, so there is no point in an expensive >SSD for just the page file. Nice try at a strawman argument: No one suggested an expensive SSD just for the page file. However, since you brought it up, Intel's 40GB SSD starts at $109. Can you quote me a x64 capable system (or parts) with 4GB of RAM for under $109 to replace my mother's existing Dell SFF system with a 2GB RAM chipset based limit? The system is a little dated, I believe it's a 2.8GHz P4 or so, so something that meets that level of performance would be required. She's typically using around 1.2-1.5GB of memory, but she's got one particular application that she runs once a month (a monthly report, oddly enough) that needs a good 1GB of RAM to get it's job done. If you can suggest a replacement system that would get her up to 4GB of RAM for under $109, I'm sure she'd be grateful. Anyway, your strawman attempt aside, the cost of putting a page file on a SSD *you already own* is $0 unless you're already desperately low on space on the SSD. In other words, I'm suggesting that if you have an existing SSD, take advantage of it to improve system performance in low memory conditions by putting the page file on the SSD.
From: JW on 13 Aug 2010 05:01 On Fri, 13 Aug 2010 09:42:13 +0200 David Brown <david(a)westcontrol.removethisbit.com> wrote in Message id: <4c64f788$0$32152$8404b019(a)news.wineasy.se>: >On 12/08/2010 20:34, Rod Speed wrote: >> David Brown wrote >>> Rod Speed wrote [...] >>> As has already been noted, modern SSDs have effectively unlimited >>> write lifetime, >> >> Thats a lie with a system that is using the page file a lot because >> it doesnt have enough system ram. >> > >Swap files don't usually involve a great deal of writing. When you have >memory pressure, little-used pages are written out to the swap file, >while you make use of the pages that are still in memory. It's not as >if pages jump back and forth between swap and main memory continuously. > >Regarding write endurance, have a look at this link: > ><http://www.eetimes.com/electronics-news/4206123/Micron-rolls-enterprise-SSD-with-SATA-6-Gbps-interface> > >To be fair, it's a top quality SSD that's just come out, and thus it's >got higher ratings than others you'll find in the shops at the moment. >But it's rated for 1.9 TB per day, every day, for 5 years. You'd have >to be doing some pretty impressive abuse of your swap to wear out that disk. You know you're wasting your time, don't you? Rod is one of the most clue-resistant trolls on Usenet. Might as well teach a pig to fly.
From: David Brown on 13 Aug 2010 06:20 On 13/08/2010 11:01, JW wrote: > On Fri, 13 Aug 2010 09:42:13 +0200 David Brown > <david(a)westcontrol.removethisbit.com> wrote in Message id: > <4c64f788$0$32152$8404b019(a)news.wineasy.se>: > >> On 12/08/2010 20:34, Rod Speed wrote: >>> David Brown wrote >>>> Rod Speed wrote > > [...] > >>>> As has already been noted, modern SSDs have effectively unlimited >>>> write lifetime, >>> >>> Thats a lie with a system that is using the page file a lot because >>> it doesnt have enough system ram. >>> >> >> Swap files don't usually involve a great deal of writing. When you have >> memory pressure, little-used pages are written out to the swap file, >> while you make use of the pages that are still in memory. It's not as >> if pages jump back and forth between swap and main memory continuously. >> >> Regarding write endurance, have a look at this link: >> >> <http://www.eetimes.com/electronics-news/4206123/Micron-rolls-enterprise-SSD-with-SATA-6-Gbps-interface> >> >> To be fair, it's a top quality SSD that's just come out, and thus it's >> got higher ratings than others you'll find in the shops at the moment. >> But it's rated for 1.9 TB per day, every day, for 5 years. You'd have >> to be doing some pretty impressive abuse of your swap to wear out that disk. > > You know you're wasting your time, don't you? Rod is one of the most > clue-resistant trolls on Usenet. > Yes, I know about Rod. I have had interesting discussions with him on several occasions - I've even learned a few things from him. But once something triggers his rodbot mode, there's no going back - it's the end of that branch of the newsgroup thread. > Might as well teach a pig to fly.
From: JW on 13 Aug 2010 07:39
On Fri, 13 Aug 2010 12:20:16 +0200 David Brown <david(a)westcontrol.removethisbit.com> wrote in Message id: <4c651c95$0$14477$8404b019(a)news.wineasy.se>: >On 13/08/2010 11:01, JW wrote: [...] >Yes, I know about Rod. I have had interesting discussions with him on >several occasions - I've even learned a few things from him. But once >something triggers his rodbot mode, there's no going back - it's the end >of that branch of the newsgroup thread. Yup. On a side note, I've been running a write endurance test on a 8GB compact flash plugged into a compact flash to PATA adapter. I'm using Passmark BurnInTest on the disk drive only at 100% utilization since July of 2009. So far, I've written and verified 142 terabytes to the drive with no errors which would be about 17,750 over-writes. Pretty impressive... The media is Transcend 100X industrial and is spec'd at two million over-writes, so I guess I can expect a write failure sometime after 2100 or so. A pity they don't make power supplies and motherboards that can last that long. |