Hey
Looking at getting my first SSD hard drive. Looking at either
OCZ Agility 3
http://www.scan.co.uk/products/60gb-...5mb-s-50k-iops
Or
Corsair Force series 3
http://www.scan.co.uk/products/60gb-...0mb-s-80k-iops
What does the IOPS figure mean?
Printable View
Hey
Looking at getting my first SSD hard drive. Looking at either
OCZ Agility 3
http://www.scan.co.uk/products/60gb-...5mb-s-50k-iops
Or
Corsair Force series 3
http://www.scan.co.uk/products/60gb-...0mb-s-80k-iops
What does the IOPS figure mean?
Input/Output Operations Per Second - it's the number of reads/writes the drive can do in a second. So 80k is better than 50k but I think you'd be hard pressed to notice the difference in real life usage - consider that 7200rpm mechanical hard drives top out at about 120iops :)
you may find this useful reading: http://www.anandtech.com/show/4604/t...-ssds-compared
http://en.wikipedia.org/wiki/IOPS
the bigger number the betterQuote:
IOPS (Input/Output Operations Per Second, pronounced i-ops) is a common performance measurement used to benchmark computer storage devices like hard disk drives (HDD), solid state drives (SSD), and storage area networks (SAN). As with any benchmark, IOPS numbers published by storage device manufacturers do not guarantee real-world application performance.[1][2]
IOPS can be measured with applications such as Iometer (originally developed by Intel), as well as IOzone and FIO[3] and is primarily used with servers to find the best storage configuration.
The specific number of IOPS possible in any system configuration will vary greatly depending upon the variables the tester enters into the program, including the balance of read and write operations, the mix of sequential and random access patterns, the number of worker threads and queue depth, as well as the data block sizes.[1] There are other factors which can also affect the IOPS results including the system setup, storage drivers, OS background operations, etc. Also, when testing SSDs in particular, there are preconditioning considerations that must be taken into account.[4]
i was actually thinking of getting one myself, from the same choice, but i think i was just going to get the cheaper one as i wasn't sure my laptop would make full use of the difference for 10% more cost
I am going to use the SSD drive with a new 990x motherboard, Asus M5A99x - which looking at the reviews can nearly exploit the SSD drives full potential.
Thanks for the input peeps.
It's just SSD manufacturers adding another metric to make their drive seem better, and to disguise their real world performance.
I think they call it "marketing," but with a pseudo scientific origin. It's a measure of the performance, but it gives you no indication of how fast that drive actually performance. That is only done via a suite of benchmarks measuring the speed reading and writing differing sized files, hence ATTO, AS SSD etc...
I suggest that before buying you identify what your common uses are, and then look for some benchmarks on that to understand how that drive will affect you.
Failing that, if you know you'll be spending £180, then just get the one that has the highest read & write speed. Avoid the OCZ Agility though - It seems as though it is the poorer cousin to the Vertex.
A couple of reviews of recent releases to help you...
http://www.bit-tech.net/hardware/sto...240gb-review/1
http://www.hexus.net/content/item.php?item=31209
Whether your laptop uses it or not is almost irrelevant, as you are highly unlikely to notice any difference in day-to-day responsiveness. As with many benchmarks - whether artificial or "real world" - the fact that there is a measurable difference between two components rarely means that there will be a noticable difference between them in day-to-day usage.Quote:
Originally Posted by uni
IOPS is actually a useful metric, if your buying something for a specific task and understand what your doing.
So for an example, back when SSDs were still cool and new I was using a FusionIO device for a project, we required very fast random reads, never writes, of 32kb, syncronous reads of 64kb chunks too.
As such these metrics can make a difference with how you choose the device and even influence the design of the software a bit.
Given that both are Sandforce, the IOPS measurement doesn't really matter as the drive will throttle due to 'Durawrite' and give you seriously decreased performance compared to the advertised specs.
Defo have a search and read about Durawrite before buying a Sandforce, as it changes a lot of peoples minds. If you want a consistent drive in terms of speed, go with the M4 or similar.
Quote:
Originally Posted by Agent
It's not necessarily that bad - Anandtech's comments are:
taken from http://www.anandtech.com/show/3656/c...0-is-very-goodQuote:
Originally Posted by anandtech
M4 has a real low write speed compared to the other drives. Won't that make it alot slower.Quote:
Originally Posted by Agent
Yup. Durawrite hits you when the data is heavily compressed or incredibly random (presumably this includes encrypted?)Quote:
Originally Posted by Brewster0101
A slow write speed on the M4 will always affect you
That's from 4/14/2010 and doesn't seem to take into account the newer 25nm NAND chips on smaller processes which are more prone to degrading.Quote:
Originally Posted by ik9000
There have been reports of people having very nasty slowdowns on the 25nm drives due to their less lifespan.
It may be a necessary evil for Sandforce drives - No solution out there is ideal, but I'd rather people know about it and make an informed decision. Although granted, home desktop users are unlikely to know / care :)
The latest Anandtech articles on Sandforce were on 11th August 2011, I gave the link above and here it is again. http://www.anandtech.com/show/4604/t...sds-compared/8Quote:
Originally Posted by Agent
It concludes with:
Quote:
Originally Posted by Anandtech
It just depends on what is important to you.Quote:
Originally Posted by Brewster0101
Some of the slow downs reported on the Sandforce drivers are insane (although there is a lot of debate on if this is actually a bug or if people are abusing them). Defo have a read over at the OCZ forums. Quite often the solution is to 'secure erase' the drive and reimage your date back.
Others OCZ are saying are 'faulty' if the speeds drop too much without the correct amount of data being written.
Either way, in my mind, it's something that would bug me.
Sure, the M4 speed might be slower write speed wise, but it's consistent. Your write speed isn't going to plummet more because you throw data at it is doesn't like. You also need to keep in mind that the vast majority of drive activity for a home machine is reading data, not writing.
Again, there is no perfect solution to NAND degradation currently, but just be aware of how each drive deals with it and make a decision based off what you prefer and how you're going to use the drive.
I missed your first link, sorry :)Quote:
Originally Posted by ik9000
As per my last post - I don't disagree that Sandforce, as a whole, is a good solution for a lot of users. I'm just saying that it's best to be aware of how it works and see if it suits your usage style, nothing more - I don't want to seem like I'm taking sides, as I'm not :)