The OCZ Octane Review (512GB)
by Anand Lal Shimpi on November 23, 2011 12:00 PM ESTPerformance Over Time & TRIM
Testing TRIM functionality is important because it gives us insight into the drive's garbage collection algorithms. OCZ insists the Octane has idle time garbage collection, a remnant of the original Indilinx drives, however in my testing I could not get the idle GC to do anything once I put the drive into a highly fragmented state. Let's start at the beginning though. The easiest way to ensure real time garbage collection is working is to fill the drive with data and then write sequentially across the drive. All LBAs will have data in them and any additional writes will force the controller to allocate from the drive's pool of spare area. This path shouldn't have any bottlenecks in it; the process should be seamless. As we've already seen from our Iometer numbers, sequential write performance at low queue depths is around 280MB/s. A quick HD Tach pass of a completely full drive gives us the same result:
The Octane works as expected here, but now what happens if we subject the drive to a ton of 4KB random writes? Unfortunately this is where the Octane falls short. If we just throw a few minutes of random writes, constrained to a small LBA range, at the Octane its performance hardly varies:
However once the Octane passes a threshold of fragmentation, the performance drop is considerable. Our standard test involves a 20 minute, 4KB random write across all LBAs at a queue depth of 32. A sequential write pass across the drive afterwards took place at between 2 and 7MB/s. Since our test drive was a 512GB model, there simply wasn't enough time to conduct a full pass in the course of preparing this review. Instead I did a shorter test with HD Tach to give you some indication of what happens to the Octane under a highy random load without TRIM:
Performance drops considerably. A single TRIM pass restores performance to new. I did have one TRIM test where only the latter half of the drive seemed to TRIM but I couldn't get the same result more than once. Now the question is, what does all of this mean?
If you have TRIM enabled on a desktop platform with a client (read: non-server) workload, none of this should matter to you. TRIM works and there doesn't appear to be any weird lag or bottlenecks in the GC path. If you don't have TRIM enabled (read: OS X) with a client workload, this could warrant a pass. The only reason I'm hesitant to recommend the Octane for use with a TRIM-less OS X installation is because I'm not entirely sure the drive will recover from this ultra low performance state without TRIM. Sequential writing alone may not be enough to adequately restore the Octane's performance. Normally idle GC would be enough, but it seems as if things get slow enough the drive's idle GC can't do much. I suspect all of this is stuff that OCZ can tweak via firmware, but I need more time with the drive to really be certain.
Finally if you're deploying a server with lots of random writes, the Octane isn't for you. OCZ will eventually release an Everest based drive for the enterprise, but the Octane is not that drive.
Facebook
Twitter
RSS
75 Comments
View All Comments
jwilliams4200 - Friday, November 25, 2011 - link
You need to carefully check the dates on that data. It takes a while for this kind of data to trickle in, so you are mostly looking at returns that were sold about a year ago.PubFiction - Saturday, November 26, 2011 - link
I don't think anandtechs claims about a turn around took place in the last year.OCZ has a low rank in most areas it plays in. This reminds me of the fiasco with the different nand in the same product for OCZ and it seemed that alot of people were unhappy with how anandtech was dealing with that. IMO OCZ still has a bad name and I do not touch anything they make unless it is significantly less expensive and reviews fine. I guess the turn around is that OCZ was worse than terrible back then. The amazing thing is that being a bad company seems to pay off because OCZ is aquiring major players in multiple fields meaning they must be making money I guess.
jwilliams4200 - Saturday, November 26, 2011 - link
Sigh. I did not say that OCZ stats will look better for parts sold after April 1, 2011. They may very well look worse. I was just pointing out that the data he was looking at may not reflect what he thought it did. For one thing, it will include very little of the SF22XX BSOD issues.faster - Friday, November 25, 2011 - link
What a great article. Not only a good technology review, but a great story about the transformation of OCZ. The perspective of this article could only be told by Anand because it was his own story and experience. This was a fantastic piece of writing that gripped me, interested me, and left me feeling good about OCZ. Great job!ericore - Friday, November 25, 2011 - link
This drive is very slow and relies poorly if you compare it with the Plextor px-m3s. In fact, that drive is probably the best SSD on the market yet Anandtech has never reviewed a single Plextor drive. It comes with a 5 year warranty, and uses a marvel controller. The specs for the 128GB alone are Random Read (IOPS 4K) 70,000, Random Write (IOPS 4K), 50,000 which hands down owns this drive and probably the Crucial M4 variant as well. Yet there has not been a single Plextor review, sadface. Note, Plextor also does a 20 hour burn-in on all drives.Also I remember reading the octane uses exclusively designed nand scaled for capacity at the cost of reliability therefore having much shorter lifespan than competition.
This is a positive step for OCZ though.
Beenthere - Friday, November 25, 2011 - link
It would be useful to see a proper review of the Plextor PX-M3 SSD. Perhaps Plextor hasn't shipped any review drives yet as I don't see any online reviews for them.Fiah - Saturday, November 26, 2011 - link
This reputable Dutch website reviewed one (I take their reviews as seriously as I do anandtechs)http://translate.google.com/translate?sl=nl&tl...
They're calling it the fastest SSD yet
Beenthere - Saturday, November 26, 2011 - link
The Dutch site appears to have tested an MP2 model not the M3S - unless Plextor is using odd model numbering in Europe?The test data showed the MP2 to be faster in some benches and slower in others. I think we'll need to see Anandtech's or StorageReview.com results to get a clear picture on SSD performance for the M3S.
I'm not so interested in speed as I am in reliability and compatibility. I have no interest in being an Un-Paid beta tester for PC companies.
Beenthere - Saturday, November 26, 2011 - link
It turns out that the MP2 is not the M3S. SR reviewed the MP2 and it's a middle-of-the pack SSD based on performance. The M3S may actually be slower based on the specs Plextor is quoting. The M3S may be geared more toward enterprise?http://www.storagereview.com/plextor_pxm2p_ssd_rev...
http://www.plextoramericas.com/index.php/ssd/px-m2...
http://www.plextoramericas.com/index.php/ssd/px-m3...
Questor - Sunday, November 27, 2011 - link
I echo the echo about seeing more smaller capacity SSD reviews. I personally have one 120 GB Intel , one 128 GB Samsung 470 and two 128 GB Plextor SSDs. I am very happy with them but would love to see all the technical stuff you folks here and Anandtech extract from this gear.I am also interested in percentage of drive sizes shipping to consumers (IE: what capacitie(s) are consumers buying most and ratio of brand to units shipped for sale, if at all possible).
I have stayed away from OCZ like it was the next coming of the plague. As a matter of fact, 1) I posted recently on another article I wouldn't touch OCZ until they could make a reliable drive and prove it and...
2) When someone around me started talking Sandforce and OCZ I would run.
Your 12 month idea of watching and waiting for reliability results is right along the lines of my thinking.
Other than those requests, thanks for a great article. It was so interesting, the wife is annoyed I stayed up to finish reading! =O I had better order those roses, come up with a real good verse for the card and get the dinner reseverations ready at some place we can't afford! If I didn't leep doing this, I could afford some 512 GB drives!