- Email |
- Print |
- Comments (30) |
- Share
If you took the time to flip through all of the benchmark pages, then you probably won’t need to read this conclusion to know what we're going to say. Intel’s first SSD, the X25-M, which aims at the premium desktop and mobile market, was already impressive. It still dominates many benchmarks, pairing high performance with great efficiency. But the X25-E is something different altogether.
Performance Madness
The new device is based on the same controller and cache memory architecture. It does not provide more maximum throughput than the X25-M (200 MB/s), and it is limited to 32 GB and 64 GB capacities for now. But it offers serious write performance (160 MB/s) thanks to single level cell flash memory, which the mainstream drive doesn’t possess. More importantly, it introduces I/O performance that is 10x to 25x higher than what you can get from the latest 15,000 RPM server hard drives. In almost every I/O benchmark, except the Web server test, the X25-E is three to five times faster than its direct flash SSD competitors.
Revealing the Inefficient
Describing the X25-E as the most efficient server drive would be correct, but I prefer to endorse it as the flash SSD storage product that finally redefines server storage performance, and resets the standards for high I/O devices. It isn’t so much more efficient than hard drives, but hard drives are simply extremely inefficient when it comes to random workloads.
Sophisticated flash memory technology has reached a level at which a single storage product is capable of delivering performance levels formerly reached only on complex RAID arrays with 6-12 hard drives. Not only does it outperform those good old hard drives, but this single X25-E storage product does it while consuming only a bit more than 1 W, on average, compared to at least 100 W for a RAID array.
This doesn’t mean that the hard drive is going to disappear, of course. High capacity applications and fast throughput remain an undisputed domain of magnetic storage products. But the days of hard drives being used in I/O intensive server applications are numbered. Hitachi and Seagate had better do their homework before releasing their flash SSD products in late 2009 or 2010, as Intel has set the bar higher than it has ever been before in the server storage market.
- Need help with intel X25-E SSD in sony sz 120 laptop [Storage]
- X25-e / adaptec 5805 /raid0 throughput problem [Storage]
- Another i7-920 gaming build, please critique [Homebuilt Systems]
- Best i7 mobo [Motherboards & Memory]
- WHO HAS THE BEST SCORE ON WIN7!!! YOU?? [Windows 7]
Questions? Ask Tom's community!
- 1 / 2
- Next
-
Sponsored links
Related forums topics
Best offers
|
X25-V 40GB 2.5" Solid State Drive (SATA) | $129.99 Newegg.com More info |
|
Portable External 640GB 2.5" Hard... | $99.99 Newegg.com More info |
|
Caviar Black 1TB Hard Drive (Serial... | $115.00 STAPLES More info |
|
eGo Desktop External 1TB Hard Drive... | $124.00 CDW.com More info |
|
Barracuda 7200.11 1.5TB Hard Drive... | $99.99 Dell Small Business More info |
All performance charts
- Insane Storage: Store 10 TB in a Quarter
- Fixing Modern SSDs for the Enterprise
- Silicon Power's 256 GB SSD Released
- server hard drives
- intel x25e
- which hard drive is the best for server
- hard drive benchmark 2009
- most efficient hard drive
- benchmark times faster
- storage server old hard drives
- fast server drives
- latest hard drive 2009
- hard drive benchmarks 2009
- ssd 2009
- intel x25m 160
- intel ssd x25m
- intel x25e ssd
- ssd storage intel
Recently, I've seen an article about performance degradation in the X25-M due to internal fragmentation, have you looked into if this also applies to the X25-E?
hard drive makers will be scratching their heads on this one. i've read some social networking and dating sites that uses around 200 to 600GB only of data.
Your previous review was of another flash-based product, the ioDrive from Fusion-io. If only you had included it's benchmarks in this article's figures, it would have been a most interesting shootoff. Any chance of updating the charts? Or do I just have to flip back and forth between the two articles?
Your previous review was of another flash-based product, the ioDrive from Fusion-io. If only you had included it's benchmarks in this article's figures, it would have been a most interesting shootoff. Any chance of updating the charts? Or do I just have to flip back and forth between the two articles?
I agree right after I clicked on this article the first thing I thought is I wonder how this compairs to the IODrive form IOFusion.
.... and a costs/performance chart with maybe total yearly costs with drive price, watt etc.. maybe even say something about the raid hardware you'll be able to save.
nice article though, good to see that for (rich) consumers the X25-M is the right choice.
So when can we expect Tom's to upgrade their servers with Intel X-25E's so that we can truly experience this amazing drive in real time???
I have one of these going into my new server I'm building actually. We'll eventually raid 4 of them. Should be a nice addition to the 4x Quad-core Shanghai's.
Your previous review was of another flash-based product, the ioDrive from Fusion-io. If only you had included it's benchmarks in this article's figures, it would have been a most interesting shootoff. Any chance of updating the charts? Or do I just have to flip back and forth between the two articles?
That's coming, probably next week, in a head-to-head between the two!
omg walks over the competition?
fusion-IO has over 100000 IO operations in the same benchmark as where this intel reaches ~6000...
don't believe me? read the IOdrive review. it utterly crushes the intel, the Mtron and all other SSD's, not to mention normal hard drives
I think you may have another bottleneck in your configuration. I have been able to get 250MB sustain read from both my MLC and SLC drives.
Could it be that the Nocona platform's chipset is limited on bandwith with its SATA connections? By the way I was refering to the Intel MLC and SLC.
That's coming, probably next week, in a head-to-head between the two!
Can you please include cost compassion? I am trying to figure out that for the same $$ which gives me a better value. So if I had $10,000 which would be a better buy.
Besides theoretical numbers, I haven't read anything about the real-world life-expectancy of these drives in workstations and/or high-volumn servers.
Can you please include cost compassion? I am trying to figure out that for the same $$ which gives me a better value. So if I had $10,000 which would be a better buy.
Area, the piece is already written, but I'll check to see if there's a discussion on cost.
Usually I'd say who cares about power consumption, but it actually makes sense in this article! An extra couple watts for a desktop becomes much much more in a server full of drives.
One thing I'd like to see is some sort of mainstream desktop hard drive in there purely for sake of comparison. Throw a Caviar blue in the mix or something - with just these benchmarks, I find it difficult to get a sense of the performance gap between these drives and the one in my desktop.
But good article guys, lots of data.
TEST THE PERFORMANCE OF THE DRIVES AT CERTAIN % POINTS - EMPTY, 20% 50% 70% AND 99% - the performance changes from what i hear - would love to know the results
Intel SSD's - making your porn load quicker!
When looking at power efficiency one should compare equal setups and I see power usage of storage of 32GB compared against up to 450GB here. One will need 14 of the smaller SSDs to hold the same data as one HD. Where does this comparo account for this?
When looking at power efficiency one should compare equal setups and I see power usage of storage of 32GB compared against up to 450GB here. One will need 14 of the smaller SSDs to hold the same data as one HD. Where does this comparo account for this?
When looking at power efficiency one should compare equal setups and I see power usage of storage of 32GB compared against up to 450GB here. One will need 14 of the smaller SSDs to hold the same data as one HD. Where does this comparo account for this?
The article looses a lot when the OCZ Vertex is not taken into account!
or, is it only available in EU?