Performance Over Time And TRIM
Clean Performance
Examining how a drive might perform over time isn't difficult. First, we fill up the user-accessible space using a sequential write, making the drive "dirty." Then, we overwrite that data using a 4 KB random write pattern. Because the drive is full of data, though, garbage collection can't consolidate the drive's scattered pages into free blocks. When we start writing over with sequential data again, the effects of active garbage collection kick in.
20 Minutes Of Random Writes
According to Iometer, sequential read/write performance should be 340/260 MB/s fresh out of box. Yet, after filling the drive with sequential data, hammering it for 20 minutes with random data, and the writing sequentially to it again, the performance story is quite a bit different. While sequential writes start at 250 MB/s, they quickly drop to 6 MB/s.
The point of this exercise, as accelerated as it is, is to replicate the worst-case situation in which you'd find yourself after writing to every available block, leaving no free blocks for the controller to use a scratch space.
If you introduce some idle time, Octane slightly recovers performance, but we're still dealing with a sequential write speed around 40-50 MB/s.
30 Minutes Of Random Writes
We already know that, even fresh out of its box, Octane's random performance is lower than most of today's modern 6 Gb/s drives. As a result, it's impossible to make all of the blocks "dirty" after 20 minutes of random writes using a queue depth of four. The remaining "clean" blocks end up obscuring what would become a more notable issue.
If we rerun the test with a queue depth of 32 and hammer the drive for 30 minutes, amplifying our previous experiment, performance drops to 7 MB/s and stays there. No matter how long you let the drive sit idle, hoping for background garbage collection to kick in and clean things up, it doesn't recover its performance. This is an issue with the Octane that we asked OCZ to comment on, but received no response. Dirty the drive completely and it appears to be toast.
Increasing Over-Provisioning
Remember a few pages back where we told you that OCZ doesn't set aside any of the Octane's NAND for over-provisioning? Crucial does the same thing with its m4, only it's able to recover more gracefully from a completely dirty state. What, then, could be the issue with Octane?
It might seem to make sense that creating a smaller-than-full partition on OCZ's drive could be a solution, manually over-provisioning it. The solution to Octane's problem isn't as simple as setting aside a little bit of unpartitioned capacity to emulate over-provisioning. Check out the diagrams below:
When we manually specify over-provisioning, the controller is able to actively free up space after the drive gets hammered with random writes. There is one catch. The amount of space the drive is able to make available for "like-new" write performance is proportional to what you over-provision. In the first diagram, we manually set 2.5%, and performance is solid for about that long. In the second diagram, half of the drive is set aside for over-provisioning, and in that case, its performance persists.
The thing is, nobody is willing to set aside half of their 512 GB SSD so that, when the drive is filled with data, performance doesn't fall off. Thus, the only real way to avoid the pitfalls of Everest's garbage collection shortcomings is to make sure you're using a system able to utilize the TRIM command. That means avoiding these things in RAID or on an older operating system.
| Transferring H.264 Movie Average Write Speed | Clean Performance | Dirty Performance |
|---|---|---|
| No Over-provisioning | ~ 250 MB/s | ~ 7 MB/s |
| 2.5% Over-provisioning | ~ 250 MB/s | ~ 10 MB/s |
| 50% Over-provisioning | ~ 250 MB/s | ~ 250 MB/s |
Here's the same sort of experiment, related in real-world terms rather than less-tangible HD Tach stuff. The difference between our "clean" and "dirty" drive with over-provisioning in place is crystal clear when we write a 32 GiB Blu-ray rip to the drive. Without an abundance of over-provisioning, the transfer rate becomes extraordinarily slow.
TRIM Performance
We already know that, if you let the Octane sit idle, it's only able to partially recover its performance. However, if you empty your Recycle Bin in Windows, triggering the TRIM command, speeds recover fully, as illustrated in the above diagram.
Related Stories :
- OCZ Octane 512 GB SSD Review: Meet Indilinx's Everest Controller
- Google Introduces Octane Browser Benchmark
- Surface Pro Benchmark Showcases Impressive Performance
- League Of Legends Performance, Benchmarked
- A Free-To-Play MMO? Neverwinter Performance, Benchmarked
- Dota 2 Performance, Benchmarked
- Crysis 3 Performance, Benchmarked On 16 Graphics Cards
- Far Cry 3 Performance, Benchmarked
- Call Of Duty: Black Ops II Graphics Performance, Benchmarked
- Medal Of Honor Warfighter Performance, Benchmarked
I think in 2 to 3 years we can get a affodable and fast 1tb ssd in market
Wow CRYSIS 2 reads and writes above 1 gb data per 20minutes
I think in 2 to 3 years we can get a affodable and fast 1tb ssd in market
Fast yes, affordable no. My guess is atleast 5 years for a 1 TB ssd to be under $100
Still comparing Crysis 2 to everything that moves? I had WD Blue in RAID 0 for quite a while and was relatively happy. Before Christmas however, I have replaced them with just simple, SATA 300 Intel 320 SSD 80Gb. Boy what a difference! No more HDD scratchy sounds, no heat from them, no vibrations, no annoying ticks when idle, silent.. Speed wise PC boots up within 30 sec, and I am only running Intel i3 2100 with no OC. To those who are holding onto HDD I would say unless capacity is the key - sell it off for an SSD. Especially now that HDD prices are skyroketting it is proving easier and easier to do the swap.
Fast yes, affordable no. My guess is at least 5 years for a 1 TB ssd to be under $100
it's so much fun to dream....don't expect prices to drop that much....that's what people people said about CPUs a few years back, yet nothing has changed.... another example is the mid and top end video cards....since manufacturing techniques have improved and have become more efficient one would think that the products would be cheaper....that's not the case....it's called demmand....people demand faster components and will pay a premium price for it, why would manufacturers drop the prices?...they still have to make a profit
I think in 2 to 3 years we can get a affodable and fast 1tb ssd in market
yeah.
and in 2 to 3 years we can get a 20 core intel 9999 X edition for $50.
and gtx990X2 for just $100.
Still comparing Crysis 2 to everything that moves? I had WD Blue in RAID 0 for quite a while and was relatively happy. Before Christmas however, I have replaced them with just simple, SATA 300 Intel 320 SSD 80Gb. Boy what a difference! No more HDD scratchy sounds, no heat from them, no vibrations, no annoying ticks when idle, silent.. Speed wise PC boots up within 30 sec, and I am only running Intel i3 2100 with no OC. To those who are holding onto HDD I would say unless capacity is the key - sell it off for an SSD. Especially now that HDD prices are skyroketting it is proving easier and easier to do the swap.
And I recommend folks hold onto their current hard drives and get a boot SSD. 80GB may be enough for you, but a lot of us have bigger storage needs. Its gonna take about a year for the hard drive market to recover, so hang on to those mechanical drives.
I think in 2 to 3 years we can get a affodable and fast 1tb ssd in market
yeah.and in 2 to 3 years we can get a 20 core intel 9999 X edition for $50.and gtx990X2 for just $100.
I do believe that 3-5 years from now we will see a huge increase on performance accompanied by a huge drop in price (compare with today's prices and performance)
Then we will probably have SATA 4 on the market and the "right price/GB/TB" will be on SATA 3 SSD's.
With that in mind, I have always build my systems a generation "behind" which is always more than "a few" generations of whatever I had built last, I have always double or triple my previous built performance for around the same money invested on it.
(plus/minus a few new "tricks" that probably were not on the previous built that could raise my budget
200 dollars or so)
Is is possible to get an 1TB SSD for around $100-$200 dollars in 3-5 years?
I believe it will be.
just don't expect to also be the faster SATA 4, you will have to "compromise" by been a little "behind"
in speed.
Add "SSD" to the title.
that's what people people said about CPUs a few years back, yet nothing has changed
AMD Athlon 64 4000+ San Diego 2.4GHz circa 2005 - $475.99 inflation adjusted to 2011 ~$548.22
Intel Core 2 Duo E6850 Conroe 3.0GHz circa 2007 - $279.99 inflation adjusted to 2011 ~$304.10
Intel Core i5-2500K Sandy Bridge 3.3GHz circa 2011 - $219.99
I'm sorry, you were saying?
I'm not sure why this product is being released. If its performance degrades so much, and the Vertex 3 is the same price and doesn't suffer from such an issue, who exactly is going to buy this? Seems like they are released a beta product on the public.
I am not seeing anything special here either. Crucial M4 series still seems like the best option out there for speed, reliability and price. I use them at home and several at work and they've all been great.
AMD Athlon 64 4000+ San Diego 2.4GHz circa 2005 - $475.99 inflation adjusted to 2011 ~$548.22Intel Core 2 Duo E6850 Conroe 3.0GHz circa 2007 - $279.99 inflation adjusted to 2011 ~$304.10Intel Core i5-2500K Sandy Bridge 3.3GHz circa 2011 - $219.99I'm sorry, you were saying?
they are still expensive....they can by sold for a lot less....here is another example....when stores have theses so called specials around 60% off retail prices they are still making a profit....so a cpu that's around 300 dollars could be sold at 180 dollars and the company would still make a profit....like said before, we empower the companies you set such high prices
they are still expensive....they can by sold for a lot less....here is another example....when stores have theses so called specials around 60% off retail prices they are still making a profit....so a cpu that's around 300 dollars could be sold at 180 dollars and the company would still make a profit....like said before, we empower the companies you set such high prices
Its called R&D... It allows progress but is also adds the "high prices" If you say getting i5 2500k is too pricey then think again.
"AMD Athlon 64 4000+ San Diego 2.4GHz circa 2005 - $475.99 inflation adjusted to 2011 ~$548.22"
Man where were you buying your athlon 64s, i cant remember what i paid. But it certainly was under 200, i dont buy $200+ processors. I bought 3 A64s and 2 A64x2s and none of them were 200 let alone 476.
I can find a athlon 64x2 3800+ in my newegg history from 2006 for 154, they were not expensive. (dont remember where i bought the others, so cant check exactly what i paid)
I'm not sure why this product is being released. If its performance degrades so much, and the Vertex 3 is the same price and doesn't suffer from such an issue, who exactly is going to buy this? Seems like they are released a beta product on the public.
um, core 2 duo's are around 150-180$ online in many places still, thats pretty high for somewhat old components... and a core 2 quad was like 300$ when i looked.
Wow this is an amazing ssd!!!!!
I think in 2 to 3 years we can get a affodable and fast 1tb ssd in market
The prices dropped about 50 cents per GB in the last year. That's roughly 25% less.
Next year if they drop another 25%, they will be roughly 1.20$ per GB
I'm being optimist by saying you will be able to get a decent SSD at around 90 cents per GB.
A year later, we can maybe look at 85Cent per GB on average.
And then another year after that, 65 Cents.
1 TB at 65 cents is 650$
Unless there's a huge surge in flash memory production, I don't see a 1TB SSD for less than 500$ in the next 3 years.
By the way, I don't see the sandisk SSD's on here. They seem to be the best deal around price wise.
they are still expensive....they can by sold for a lot less....here is another example....when stores have theses so called specials around 60% off retail prices they are still making a profit....so a cpu that's around 300 dollars could be sold at 180 dollars and the company would still make a profit....like said before, we empower the companies you set such high prices
You're conflating the retail space with the manufacturing space here to make a point that would not otherwise stand. When a retail store sells at 60% off retail prices it has absolutely nothing to do with manufactuers and what profits they may or may not be making and everything to do with retail strategy. They're either loss-leading or trying to move excess inventory. Either way they're not making a profit on the item being sold they're recovering the loss some other way or just writing off the loss. Just because some retail store offers a sale of 60% off doesn't mean that every item on the market sold at 100% MSRP was a 40% profit for the retail outlet. Which, again, is not the manufacture anyway so even if you point stood it wouldn't demonstrate what you're attempting to argue.