Ocz vertex 2 iops: OCZ Technology Vertex 2 3.5-inch 120GB Solid State Drive

OCZ Technology Vertex 2 3.5-inch 120GB Solid State Drive

Storage

SSDs

OCZ takes it to the next level with a new 3.5″ form factor SandForce SSD designed to be the ultimate desktop SSD.

Published Dec 27, 2010 10:07 AM CST   |   Updated Tue, Nov 3 2020 7:02 PM CST

Manufacturer: OCZ Technology

10 minute read time

Introduction

OCZ Technology is the Swiss Army of solid state drives. The company has a product in their line-up for every occasion. Until now desktop users have been forced to pay a premium for smaller 2.5″ notebook drives that need a desktop adapter bracket. Notebook drives require larger flash modules and need an adapter to fit in desktops. These premium parts raise the cost of the SSD whether you need them or not.

Keeping with the overall strategy of a product for every use (like a Swiss Army Knife), OCZ now has a desktop specific SSD that uses less dense flash and a native 3.5″ form factor enclosure. The new 3. 5″ design is available in both Agility 2 and Vertex 2 programming with the latter having a higher IOPS. At the time of writing the 3.5″ models cost the same as their notebook cousins, but this should change over time since the 3.5″ drives should cost less to produce.

Today we will take a look at the OCZ Vertex 2 3.5″ 120GB desktop solid state drive. On the surface the new drive appears to be just a larger Vertex 2, but internally the design is fresh and we should see a different performance profile. Let’s move on and see what is different.

Specifications, Pricing and Availability

On paper the 3.5″ Vertex 2 has the exact same specs as the 2.5″ model. At this time OCZ is offering the drive in 90, 120, 180, 240, 360 and a massive 480GB. Performance for all but the largest are very respectable; 285MB/s read with 275MB/s write. The 480GB drive takes a performance hit due to its size.

Newegg lists the same 3.5″ 120GB Vertex 2 that we are looking at today for 229. 99 USD. The 2.5″ Vertex 2 is also listed at 229.99, so there isn’t a cost benefit at this time for choosing the desktop model, but this should change over time.

The Packaging

As you can imagine, the 3.5″ packaging is larger than the 2.5″ models we’ve looked at in the past. OCZ kept the same color scheme for the larger Vertex 2, the familiar black and silver bad boy look.

The back of the package lists specifications and features. These go hand in hand with that bad boy color scheme since the Vertex 2 maxes the SATA II specification with 285MB/s read and 275MB/s write speeds.

The inner package encases the drive in closed cell foam so the drive is secure for shipping. A paper manual and sticker is also included.

The OCZ Vertex 2 3.5″ SSD

Just like any other solid state drive, the 3.5″ Vertex 2 is pretty plain on the outside. The top cover that we see here is made from plastic with a dark smoke finish. It would have been really cool to put a clear cover on the SSD so users can see the inside, but smoke is what we got.

The rest of the drive casing is made from aluminum. On the bottom we found standard mounting points for the 3.5″ form factor standard.

Here we see that the new 3.5″ Vertex 2’s height. The drive is thinner than standard desktop drives, but that won’t be an issue for 99% of users. I am leaving a small amount of play in that statement since I don’t have every imaginable configuration possible, but I tried the drive in several without issue.

The SATA power and data connectors are where they should be as per the SATA specification.

The side of the drive has all of the standard mounting points as well. At one point Maxtor and later Seagate released traditional platter drives in this same size.

This is where things get interesting. Here we see the controller side of the Vertex 2. There are 16 flash memory chips on this side, the same amount found on the 2.5″ drive in total.

Here we see the other side; I call it the Supercap side. There are another 16 flash chips on this side. What this tells us is that OCZ is using flash modules that are less dense for the 3.5″ drives. This should allow OCZ to manufacture the drives at a lower cost depending on the prices of flash at the time.

There is also a possibility of a performance increase since more flash is being read and written to at the same time. That is really what we are here to see, so let’s get right to it!

Test System Setup and ATTO Baseline Performance

We would like to thank the following companies for supplying and supporting us with our test system hardware and equipment: AVADirect, GIGABYTE, Cooler Master, LSI, Corsair. and Noctua.

You can read more about TweakTown’s Storage Product Testing Workstation and the procedures followed to test productsin this article.

Even though we are winding down 2010, we plan to use our awesome 2010 test system well into 2011. Look for a few updates in the coming weeks, but this test we are using the same system as we did for the Vertex 2 2.5″ reviews.

ATTO Baseline Performance

Version and / or Patch Used:2.34

ATTO is used by many disk manufacturers to determine the read and write speeds that will be presented to customers.

In ATTO we found a maximum read speed of just over 280MB/s and a maximum write speed of 272MB/s.

Benchmarks — HD Tune Pro

HD Tune Pro

Version and / or Patch Used:4.00
Developer Homepage:http://www.efdsoftware.com
Product Homepage: http://www.hdtune.com

HD Tune is a Hard Disk utility which has the following functions:

Benchmark: measures the performance
Info: shows detailed information
Health: checks the health status by using SMART
Error Scan: scans the surface for errors
Temperature display

HD Tune Pro gives us accurate read, write and access time results and for the last couple of years has been gaining popularity amongst reviewers. It is now considered a must have application for storage device testing.

It’s no surprise that the sequential speeds are identical to the 2.5″ Vertex 2 since both form factors are limited by the SATA bus. Here we see the same speeds that made the Vertex 2 our pick for the ‘what to buy in the less than 256GB’ category.

The write speeds are also just as high. Here we see the Vertex drives dominating the write speed charts.

Benchmarks — Everest Random Access Time

Everest Random Access Time

Version and / or Patch Used:4.60
Developer Homepage:http://www.lavalys.com
Product Homepage:http://www.lavalys.com

Everest Ultimate and Corporate Edition offer several different benchmarks for testing and optimizing your system or network. The Random Access test is one of very few if not only that will measure hard drives random access times in hundredths of milliseconds as oppose to tens of milliseconds.

Drives with only one or two tests displayed in write the write test mean that they have failed the test and their Maximum and possibly their Average Scores were very high after the cached fills. This usually happens only with controllers manufactured by JMicron.

Access times are what give solid state drives their perceived speed advantage over traditional platter drives. Here we see the 120GB Vertex 2 3.5″ running the exact same numbers as the 2.5″ model.

It is difficult to imagine that both 120GB Vertex 2 drives are also running the exact same write latency speeds, even though their configurations are so much different.

Benchmarks — Crystal Disk Mark

CrystalDiskMark

Version and / or Patch Used:3.0 Technical Preview
Developer Homepage:http://crystalmark.info
Product Homepage:http://crystalmark.info/software/CrystalDiskMark/index-e.html
Download here:http://crystaldew. info/category/software/crystaldiskmark

CrystalDiskMark is a disk benchmark software.

Key Features:-

* Sequential reads/writes
* Random 4KB/512KB reads/writes
* Text copy
* Change dialog design
* internationalization (i18n)

Note: Crystal Disk Mark 3.0 is not available to the public yet, but the Technical Preview does allow us to test 4K performance at queue depths of 4 and 32 in addition to 1. The current release Crystal Disk Mark only shows us QD 1.

In CDM we get to look at the 4K and queue depth performance. This is one area that I thought the new 3.5″ design would start to pull away from the 2.5″ model, but it looks like I was wrong. For the most part both drives run nearly identical when measured at the same 120GB capacity.

Keeping with the theme so far, the OCZ Vertex 2 3.5″ drive also matches the 2.5″ model in 4K write speeds at depth. Let’s get to the real world tests and see what happens there.

Benchmarks — PCMark Vantage Hard Disk Tests

PCMark Vantage — Hard Disk Tests

Version and / or Patch Used:1. 0.0
Developer Homepage:http://www.futuremark.com
Product Homepage:http://www.futuremark.com/benchmarks/pcmark-vantage/
Buy It Here

PCMark Vantage is the first objective hardware performance benchmark for PCs running 32 and 64 bit versions of Microsoft Windows Vista. PCMark Vantage is perfectly suited for benchmarking any type of Microsoft Windows Vista PC from multimedia home entertainment systems and laptops to dedicated workstations and high-end gaming rigs. Regardless of whether the benchmarker is an artist or an IT Professional, PCMark Vantage shows the user where their system soars or falls flat, and how to get the most performance possible out of their hardware. PCMark Vantage is easy enough for even the most casual enthusiast to use yet supports in-depth, professional industry grade testing.

FutureMark has developed a good set of hard disk tests for their PCMark Vantage Suite. Windows users can count on Vantage to show them how a drive will perform in normal day to day usage scenarios. For most users these are the tests that matter since many of the old hat ways to measure performance have become ineffective to measure true Windows performance.

HDD1 -Windows Defender
HDD2 — Gaming
HDD3 — Windows Photo Gallery
HDD4 -Vista Startup
HDD5 -Windows Movie Maker
HDD6 -Windows Media Center
HDD7 -Windows Media Player
HDD8 — Application Loading

I dislike it when I’m wrong, but can admit it when things don’t go as I thought they would. In PCMark’s Vantage we see that the real world performance of the 2.5 and 3.5″ Vertex 2 is nearly identical. I really thought we would see the 3.5″ version outperforming the 2.5″ model since it uses more flash, but that wasn’t the case. The good news is that desktop users won’t take a performance hit and get native form factor support, making for an easier installation.

Benchmarks — AS SSD

AS SSD Benchmark

Version and / or Patch Used: 1. 2.3577.40358
Developer Homepage:Alex Intelligent Software
Product Homepage:Alex Intelligent Software
Download here:http://www.alex-is.de/PHP/fusion/downloads.php?cat_id=4&download_id=9

AS determines the performance of Solid State Drives (SSD). The tool contains four synthetic as well as three practice tests. The synthetic tests are to determine the sequential and random read and write performance of the SSD. These tests are carried out without the use of the operating system caches.

In all synthetic tests the test file size is 1GB. AS can also determine the access time of the SSD, the access of which the drive is determined to read through the entire capacity of the SSD (Full Stroke). The write access test is only to be met with a 1 GB big test file. At the end of the tests three values for the read and write as well as the overall performance will be issued. In addition to the calculated values which are shown in MB/s, they are also represented in IO per seconds (IOPS).

Note: AS SSD is a great benchmark for many tests, but since Crystal Disk Mark covers a broader range of 4K tests and HD Tune Pro covering sequential speeds, we will only use the Copy Benchmark from AS SSD.

— Copy Benchmark

In AS SSD we are able to gauge the real world file transfer performance in three different scenarios. For the first time we found the 3.5″ version taking a slight performance hit to the 2.5″ version, or so it may seem.

The 3.5″ drive we used shipped with firmware 1.5. This was the first time we’d seen this firmware, but as the firmware revisions have gone up, the scores in these tests have gone down. We’ll spend more time with this new firmware and add the data to our upcoming SandForce Firmware Revisions article coming next month.

Benchmarks — Passmark

Passmark Advanced Multi-User Tests

Version and / or Patch Used:6.1
Developer Homepage:http://www. passmark.com
Test Homepage:http://www.passmark.com

Many users complain that I/O Meter is too complicated of a benchmark to replicate results so my quest to find an alternative was started. Passmark has added several multi-user tests that measure a hard drives ability to operate in a multi-user environment.

The tests use different settings to mimic basic multi-user operations as they would play out on your server. Variances is read / write percentage as well as random / sequential reads are common in certain applications, Web Servers read nearly 100% of the time while Database Servers write a small amount of data.

The Workstation test is the only single user environment and will be similar to how you use your system at home.

The server tests show more of the same when comparing the Vertex 2 120GB drives. Here we see that the two 120GB Vertex 2’s score nearly identical in all tests.

Final Thoughts

The OCZ Vertex 2 3.5″ form factor SSD is a story of equality for the desktop. For years now SSDs have been designed for notebooks and the desktop remained silent. Those days are over and the desktop has finally spoken up and OCZ heard the cries for lower access times and higher data transfer speeds.

In all seriousness, the OCZ Vertex 2 3.5″ performs exactly like the 2.5″ model, even with the different flash configuration. Going through the list of benchmarks, it’s clear that the two drives are nearly interchangeable when it comes to both synthetic and real world performance. Yes, I am a bit surprised by this and was expecting something different, but it is what it is. The OCZ Vertex 2 3.5″ drive has the same award winning performance as its 2.5″ cousin does.

At the same time, the prices for both drives are exactly the same; 229.99 for the 120GB models at Newegg. The performance isn’t going to change, but I think over time the price will. This is a strange time of year to look at e-tail pricing. The Christmas holiday has pricing bouncing all over the place with sale prices one day and inflated prices the next due to high demand. SSDs are on the top of many «computer guys» lists and OCZ’s top notch marketing and high performance has led to the Vertex 2 being asked for by name more than any other SSD on the market. Once we get out of the holiday season we may see the 3.5″ Vertex 2 and Agility 2 price decline slightly.

The real question for desktop users, though, is if the Vertex 2 3.5″ drive is a proper fit for their systems. To answer that question you should really ask yourself what you plan to do with your drive after your next purchase. If the Vertex 2 is going to have a long term home in your desktop, then having the convenience of a native 3.5″ install has a very positive outlook. If you are someone who will upgrade to the latest and greatest, possibly the Vertex 3 early next year, you might want to look at resale value or moving the Vertex 2 to your notebook.

I’ve said on many occasions that once you go with an SSD for your boot drive, you will never want to use a platter drive for anything. This goes for your notebook too, so even if you think one drive will do for now, you will want a SSD for your notebook sooner than you think. Because of that, the 2.5″ drive might actually be a better buy at this time since the prices are identical and the 2.5″ Vertex 2 ships with a desktop adapter. You lose the ease of installation of a 3.5″ native install, but in the long term you set yourself up for SSD bliss all around.

Shopping Information

PRICING: You can find products similar to this one for sale below.

United States: Find other tech and computer products like this over at Amazon.com

United Kingdom: Find other tech and computer products like this over at Amazon.co.uk

Australia: Find other tech and computer products like this over at Amazon.com.au

Canada: Find other tech and computer products like this over at Amazon. ca

Deutschland: Finde andere Technik- und Computerprodukte wie dieses auf Amazon.de

Newsletter Subscription

Right of Reply

We openly invite the companies who provide us with review samples / who are mentioned or discussed to express their opinion. If any company representative wishes to respond, we will publish the response here. Please contact us if you wish to respond.

Related Tags
  • Storage
  • ocz
  • ssd
  • solid state drive
  • 3.5-inch
  • vertex 2
  • 120GB
  • sandforce

OCZ Vertex 2 100GB SSD

The OCZ Vertex 2 uses the new SandForce SF-1200 controller and proprietary OCZ firmware. That firmware allows for up to 50K IOPS making it the fastest MLC SSD we’ve tested.

Introduction

The OCZ Vertex 2 utilizes the new SandForce SF-1200 controller. We’ve been waiting on a drive with the SandForce controller for a while and when OCZ approached us about taking a look at it we were eager to get it into the lab to take a look. There are other SandForce controller drives out there but this one uses a proprietary Firmware that allows it up to 50k IOPS making it in a class by itself. OCZ has been testing the SandForce SF-1200 and SF-1500 controllers for months internally and during CES they made the formal announcement that OCZ would be bringing out the Vertex 2 series of SSDs using these two new controllers. The SF-1500 is used in the OCZ Vertex Limited edition and the SF-1200 is used in the Vertex 2 lineup. The SF-1200 has performance close to the SF-1500 at a fraction of the cost so we can’t wait to get it on the bench and drive it until it screams.

The Vertex 2 will be available in 50GB (OCZSSD2-2VTX50G), 100GB (OCZSSD2-2VTX100G) and 200GB (OCZSSD2-2VTX200G) drive capacities (50GB  $199.00, 100GB at $409.99, 200GB  for $749.99. They aren’t cheap but the performance is very high on this series.

The drive we have to review is the OCZ Vertex 2 100GB model with part number OCZSSD2-2VTX100G.It has a nice matte black finish and should fit well with any system motif. There really isn’t much you can do to pretty these things up and really since it will be locked away in a laptop bay or tucked discretely into your chassis there is no need.

This is a standard Sata 2 interface and with the revision of BIOS we received on the drive it should run at 285MB/s read and 275MB/s write with up to an impressive 50,000 IOPS. This drive runs better with a heavy Que Depth (Depth 32) so only those applications using heavy Que Depth will see top IOPS performance.

Vertex 2 Maximum Performance 

  • Max Read: up to 285MB/s
  • Max Write: up to 275MB/s
  • Sustained Write: up to 250MB/s
  • Max Aligned 4k IOPS: up to 50,000 IOPS

With Native Trim Support and weighing in at 77g and 99.8 x 69.63 x 9.3mm in size the Vertex 2 has a shock tolerance of 1500G. The drive uses 0.5W at idle and during operation only 2W so it barely sips on your power supply.

Inside the box we found the Vertex 2, a set of screws, a drive tray (Nice Addition!) and a cute sticker. The Vertex 2 has a nice 3 year warranty should it fail inside that period. With a rated MTBF of 2 million hours OCZ won’t be very worried about many RMA’s.

Here’s some more information on the drive for those that like endless specifications and features.

 

  • Available in 50GB to 400GB capacities
  • Available in 60GB to 480GB extended capacities
  • Native TRIM support
  • Max IOPS Firmware
  • Seek Time: .1ms
  • Slim 2.5″ Design
  • 99.8 x 69.63 x 9.3mm
  • Lightweight: 77g
  • Operating Temp: 0°C ~ 70°C
  • Storage Temp: -45°C ~ +85°C
  • Low Power Consumption: 2W in operation,
    .5W in standby
  • Shock Resistant up to 1500G
  • RAID Support
  • Included 3.5″ Desktop adapter bracket
  • Compatible with Windows XP, Vista, 7, and Linux
  • MTBF: 2 million hours
  • 3-Year Warranty

50-240GB Max Performance

  • Max Read: up to 285MB/s
  • Max Write: up to 275MB/s
  • Sustained Write: up to 250MB/s
  • Random Write 4KB (Aligned): 50,000 IOPS

400-480GB Max Performance

  • Max Read: up to 250MB/s
  • Max Write: up to 240MB/s
  • Sustained Write: up to 200MB/s
  • Random Write 4KB (Aligned): 22,500 IOPS

Part Numbers

 

OCZ Toolbox Utility v.

0.60

We got a version of OCZ toolbox utility with the Vertex 2 and the Toolbox isn’t ready for mass production yet so we aren’t going very far into it.

On the first page (not pictured) you can select the drive to work on. This page (tools) you can download Firmware, Set Factory Defaults (didn’t work for us because it’s not ready), change the drive’s name and most importantly format the drive with optimized sector alignment for better performance.

The Security tab allows you to set a drive password, unlock the drive and securely erase the drive. When it’s completely finished the OCZ Toolbox will be a handy little tool for working with your drive. Even partially functional it’s still nice. OCZ informs us the utility should be ready for full functionality soon.

Testing Method

To test the OCZ Vertex 2 we cloned our test rig drive to the SSD. It is the same test drive we’ve been using on all of our drive testing and is nothing more than a clean Windows load with all the drive testing software installed, as well as all the current drivers and patches for the OS. It’s the equivalent of doing a fresh load of Vista from the disc but takes a lot less time and ensures that every drive tested uses exactly the same OS load and drivers. Nothing that may effect the outcome of the testing procedure can creep in. We ran all of the tests a total of 3 times and averaged those results. The Average of the three results are presented here. In the case of a pictorial benchmark we ran the bench 3 times and picked the median result. As with most SSD testing differences from run to run are minimal and the median result is a good indication of what you can expect from the drive.

We ran our usual battery of tests on the drive, and used it as the primary boot drive during testing. All of the drives tested were used as the primary boot drive during testing. That’s a more realistic test than strapping the drive in and testing it with a bare format or as a non-boot drive and it represents real life transfer rates, much like you can expect when you install and operate the drive in your own system. Each test was performed 3 times and the average of the 3 test run is reported here.

Test Rig

Test Rig
“Quadzilla”
Case Type Top Deck Testing Station
CPU Intel Core I7 980 Extreme
Motherboard Asus Rampage 3
Ram Kingston HyperX 12GB 9-9-9-24
CPU Cooler Thermalright Ultra 120 RT (Dual 120mm Fans)
Hard Drives

OCZ Vertex 2 100GB
Kingston SSDNow V+ 128GB MLC
OCZ Agility EX 60GB (SLC)
Kingston SSDNow V 40GB (MLC)
Kingston SSDNow V+ 128GB (MLC)
G. Skill Titan 256 GB SSD (FM-25S2S-256GBT1) (MLC)
Patriot 128GB Warp SSD (MLC)
Intel 80 GB SSD X25-M
G.Skill 64GB SSD (FM-25S25-64GB) (MLC)
2 WD VelociRaptor’s 300GB (In single and Raid0)
WD 160 GB SATA 2
Maxtor 160 GB SATA 2
WD & Maxtor in Raid 0

Optical Asus BD-Combo
GPU Asus GTX-470
Case Fans 120mm Fan cooling the mosfet CPU area
Docking Stations None
Testing PSU Thermaltake Toughpower 1200 Watt
Legacy None
Mouse Razer Lachesis
Keyboard Razer Lycosa
Gaming Ear Buds
Razer Moray
Speakers None
Any Attempt Copy This System Configuration May Lead to Bankruptcy

Test Suite

Benchmarks

ATTO

HDTach

Crystal DiskMark

HD Tune Pro

AS SSD Benchmark

IOMeter

IOMeter

In IOMeter we are using a preset configuration file for each test. The configuration files were obtained from a trustworthy vendor (Not OCZ) who had no idea what we were going to use them for. Normally they are not optimized or altered in any way after receiving them. That way the vendor doesn’t influence testing results, we don’t optimize for any particular drive, and testing is the same for every drive across the board with no chance of prejudice working its way in. In this case the OCZ Vertex 100GB drive is said to have exceptional 4k write speeds at Que Depth 32, we didn’t use that setting but did modify the Random Write test to Random Write 4k, the updated configuration file we requested from OCZ should be random Read and Write 4k with Que Depth 32. Until then we are running with what we have in IOMeter and acquired an updated version of CrystalDiskMark which has a Que Depth 32 portion.

The Four Test files are:

  • Random Read IOPS Test
  • Random Write IOPS Test
  • Seq Read Test
  • Seq Write Test

We have requested a set of optimized test configuration files from OCZ and when we receive them we will run some optimized tests and those we will clearly label as OCZ Optimized runs.

Random 4k Write IOPS shows that the heavy Que depth of 32 is pretty much a requirement for good IOPS. Using the configuration file we always use with default settings and the parameters changed to 4k 100% random 4k Write we got nowhere close to 50K IOPS. We got a transfer speed of 12.54 MB/s which is close to what we got in other benchmarks with low Que Depth.

Random Read IOPS we got a great result for IOPS coming in at 32605.14 IOPS and the transfer rate hit 127.36.

Sequential Read we got 1973.98 IOPS with a transfer speed of 246.75 MB/s which is a pretty amazing transfer rate.

Sequential Write we got 1050.19 IOPS with a transfer speed of 131.27 MB/s and the question we find ourselves asking is since the Vertex 2 responds best to a Que Depth of 32 is: How many programs and applications use a que depth of 32 to take advantage of the Vertex’s high IOPS capabilities.

CrystalDiskMark

In CrystalDiskMark we find ourselves in an unusual position. We have been using version 2. 2 but 3.0 is the newest version and it has a Que Depth 32 portion. What we have done in this review is used both and made note of which version we are using. It’s a stopgap until we get time to re-bench a stack of drives that would make your head spin. We also wanted to study Aligned and Un-Aligned as offered by the OCZ Toolbox. Unfortunately we Aligned the drive not knowing the OCZ Toolbox would give us this error when we tried to reset factory default settings.

SInce the drive is supposed to have exceptional 4K performance with a Que Depth of 32 we limited our exploration of CDM 3.0 to that until we have time to re-bench a slew of drives.

We were lucky though, most of the tests we needed were already in the bag when we got the Function Disabled message.

Sequential write we are using the same 2.2 version we’ve been using for a while and we got 154.1 MB/s

Sequential Read we get a score of 209.6 MB/s and you can tell that CDM 2.2 doesn’t use a particularly heavy Que Depth which is probably going to force us to look for more flexible Hard Drive Benchmarks like IOMeter.

Still using version 2.2 we got a 70.57 MB/s on the Vertex 2 in the 512K Write. Not to worry we included a screen shot of both 2.2 and 3.0 a little later.

The 512 Read test shows us 190.5 MB/s and the Vertex 2 beats all but two of our drives. It’s a little refreshing to see a drive that wasn’t completely optimized for the benchmark suites and focuses on real life performance.

Here’s where we are using CDM 2.2 and 3.0 and keep in mind that since they are different versions this is somewhat Apples to Oranges in comparison. Without aligning the drive in version 2.2 we got a lackluster 1.993 MB/s in the 4k Write test. When we aligned the drive using the same version we got 63.49 MB/s and it’s looking like the OCZ Toolbox is indeed a very important tool for this drive.

Shifting to version 3.0 aligned we got an almost identical score of 63.54 MB/s but when we read the Que Depth score with a Que Depth of 32 we almost choked. We got a whopping 132.4 MB/s and the proprietary Firmware OCZ is using on the SandForce SF-1200 controller is looking like gold to us.

Here again we use both version 2.2 and version 3.0 and Version 2.2 with the drive unaligned we got another lackluster score of 16.77 in the read test. Align the drive using the OCZ toolbox and we got 28.14 MB/s and the second fastest 4k Read we have ever gotten and the only drive beating it was a SLC drive which is a totally different critter.

Move to CDM Version 3.0 and again we got an almost identical score to version 2.2 after the drive was aligned. Move to Que Depth 32 and again we almost choked and saw a score of 195 MB/s and again the question of the day is: How many applications use that heavy a Que Depth? We would bet that a Que Depth of 32 is the exception and not the rule.

Putting Version 2.2 and Version 3.0 on top one another in screenshots you can see that the only real difference between the two in scores is the new 4k read/write with a Que Depth of 32.

We’ve tried contacting the CrystalDiskMark team to find out what prompted them to add a Que Depth of 32 to their benchmark but as yet we have received no response.

AS SSD

AS SSD is an infant in the benchmark world and has been going through enough revisions that we just use screen shots of it. Trying to chart performance on a benchmark that is being changed weekly is a little pointless but since it’s written specifically for SSD’s we are tossing in the screenshots.

The COpy portion of AS SSD give us a little more realistic look at speed than raw benchmarks of a synthetic nature. In the copy portion we got 141.20 MB/s for ISO’s, the program sections shows us 118.05 MB/s and the gaming section shows 137.57 MB/s. We’ve actually ran platter drives on this benchmark and they take hours to run, SSD’s finish in mere minutes (3 – 5 Minutes). That should tell you the inherent difference between SSD and Platter Drives.

The important section of the IOP’s test in this case is the 4k-84k Threaded. Most drives we’ve seen hit 5k in the 4k or 4k-64k Threaded threaded if they are lucky. The Vertex 2 blows that out of the water and hits 29 – 32k(ish) in the 4k – 64k Threaded test.

Straight speed benchmarks show us that the OCZ Vertex 2 does indeed excel at 4k and 4k – 64k Threaded testing. A lot of other drives we’ve seen top out at 32 MB/s and here we see a top end of 117.5 MB/s Write.

ATTO

ATTO in the 8MB Write test shows us the Vertex 2 100Gb SSD running away with the performance crown and probably laughing at the competition.

The Read test shows up pretty much the same thing and the drive approaches the theoretical limit of Sata 2 transfer speeds. We would love to get a couple of these drives on a hardware RAID controller and see just how they make us smile.

Kicking to the 512k test we see the Vertex 2 65MB/s ahead of the closest drive and once again probably can’t even see the other drives in it’s rear view mirror.

The Vertex 2 makes a clean sweep of things in ATTO and topped the charts in every test.

 

We wanted to see the difference between short 8 MB tests and 32 MB tests on this drive so we included the screen shots of the tests. If you will notice the tests are more choppy in the 8 mB portion.

Move to the 32 MB test and the drive hits stride at about 16kb and remains consistently above 275MB/s read and write beyond that.

Hard Drive Tune Pro

Minimum speed on the Vertex 2 comes in at 204.5 MB/s and from its position on the chart you can see this drive is a contender for overall fastest drive. (Considering the Agility EX is an SLC drive anyway)

The Vertex 2 hit 217 MB/s in the Maximum speed test and the only MLC drive to top it was the Intel X25-M.

The burst speed on the drive hit 166.9 and isn’t representative of the drives true performance. Burst tests tend to be cache only hits and not very indicative of real performance.

The Average speed of the Vertex 2 is 214 MB/s and again the only MLC (multi Level Cell) drive that managed to best it is the Intel X25-M and HD Tune Pro seems to favor the Intel drive somewhat because this is about the only test where it bests the Vertex 2.

HDTach

Here again, let us iterate that we don’t weigh burst testing heavily in the final scoring of a drive. The Vertex 2 hit more than respectable burst speed of 262.8 MB/s but burst tests usually consist of cache only hits and as for real life performance don’t do a lot for you unless your stuck in a cubical doing the same redundant things over and over for an 8 hour shift. Normal usage we tend to wander from program to program and the cache flushes more often than the toilet at a raw oyster bar.

The Average Speed of the Vertex 2 where the smart money is at hits 238.9 MB/s and at this point we would have to say the Vertex 2 is the fastest overall drive we’ve tested.

Conclusion

Performance wise the OCZ Vertex 2 does an excellent job of satisfying our need for speed. Price wise we aren’t ready to throw our endorsement completely behind any SSD at this time. With a $400(ish) price tag and formatted capacity of around 93GB that exceeds $4/Gigabyte and that cost sets us back a little.

If you can spring the price the Vertex 2 does provides blazing speed and a 3 year warranty but you are trading storage capacity for speed and that has to be considered. We’ve seen (admittedly slower) 1TB platter drives on sale for $100 and for the same price as the Vertex 2 you can have 4 TB of storage running in RAID 0 or 100 GB of SSD storage. That’s a hard choice to make.

All that being said if you have need for speed SSD’s simplify things and provide great performance but they do so at a premium price. The OCZ Vertex 2 has the premium speed and premium price but like most SSD’s storage is limited. We love the speed, grumble about the price and plain don’t like the limited size.

Still with Native Trim support, the SandForce controller, OCZ proprietary Firmware, and exceptional speed it’s hard not to like the Vertex 2 100GB.

When considering purchasing an SSD the Vertex 2 is hard not to recommend as far as SSD’s go. It’s the overall fastest SSD we’ve seen.

OUR VERDICT: OCZ Vertex 2
Performance 9. 5
Value 8.5
Quality 9
Features 9
Innovation 10

We are using a new addition to our scoring system to provide additional feedback beyond a flat score. Please note that the final score isn’t an aggregate average of the new rating system.

Total 9

TRIM support

Good sequential read and write

Great Depth 32 Que 4k Reads and Writes

New SandleForce Controller looks good!

3 year warranty

Price Vs Storage Capacity

 

 

Summary: The OCZ Vertex 2 proves to be a signifigant force in the SSD world. Overall we would say it’s the fastest drive we’ve tested. The future looks bright for the OCZ lineup of drives and their OCZ Toolbox just makes it even sweeter.

OCZ Vertex 2 100GB Performance Preview – Techgage

Today we have a quick preview of the OCZ Vertex 2 SSD that is currently breezing its way through our SSD testing benchmarks. Sure, hexa-core Gulftown and Thuban CPUs are all the current rage, and graphics cards like the Radeon HD 5870 and GeForce GTX 480 are literally sizzling right now. But any enthusiast will probably say that solid-state drives are still where things are at, and we have to agree given the remarkable boost in performance and system responsiveness which is seen over traditional hard disks.

For most, prices have been the deterring factor, but as anyone familiar with Intel’s benchmark-setting X25-M G2 series will recall, demand was so strong that it took over six months before drives settled around Intel’s recommended MSRP!

The SSD we have just received is not a value drive (even amongst other SSDs), but rather it is targeted squarely at the cutting-edge enthusiast that wants top-notch performance and is willing to pay a bit more to get it. And it promises to blow Intel’s X25-M G2 out of the water. SandForce is a name quickly circulating amongst tech enthusiasts and with good reason.

OCZ’s Vertex 2 is powered by the SandForce SF-1200 controller and uses the “older” SATA 2 interface, but that doesn’t seem to pose a problem for this drive from what we have seen so far! Officially, the SF-1200 “only” offers 10,000 4K random write IOPS performance, but the SF-1200 powering the Vertex 2 uses a special sauce firmware that is rated for 30,000 4K random write IOPS.

To put those numbers in perspective, Intel officially rates the X25-M G2 80GB for 6,600 4K random write IOPS, and the 160GB for 8,600 IOPS. Other drives such as the Indilinx Barefoot-powered Vertex (still held in high regard for its excellent performance) actually offers even lower IOPS results.

IOPS are not everything though, with OCZ quoting max reads of up to 285MB/s and nearly identical max writes of 275MB/s. Offered in 50, 100, and 200GB capacities, the Vertex 2 will be OCZ’s new flagship SSD and of course includes TRIM, RAID, and a metal bracket to allow easy mounting in traditional 3. 5″ desktop drive bays. OCZ wraps it all up nicely with a three year warranty and extensive user support through its forums.

No preview would be complete without a sneak peek at the results, so here’s a quick look. PCMark Vantage is an excellent program comprised of a series of real-world usage tests that target specific parts of a system, and then combines them to form a single overarching system-wide performance suite. The overall PCMark score is a reflection of our test platform’s configuration and shows the drive performing well, but it isn’t until the overall HDD suite score that we see results take off!

The original Vertex Turbo manages an already impressive standalone score of 24,174 PCMarks, but the Vertex 2 eclipses this with a result of 34,609! That is a 69.8% gain over the previous best SSD result for the test! If we are even comparing it to the hard disk drive we are looking at a whopping 10x improvement in the HDD benchmark score alone. This isn’t bad for an older Q6600 test platform running at a stock 2. 4GHz frequency, but a modern Core i7 system has the potential to break the 40,000 mark in the HDD suite with this drive!

While our batch tests are fairly unique, they have served us well in determining what SSDs come to the table prepared and which drives have that special something extra to handle even the most demanding multitasking workloads. That said we are going to need to devise even more demanding tests as the system itself is becoming the bottleneck point , as evidenced with the Vertex 2 being the third drive to post impressive results a little over four and a half minutes. With the heavy batch test the drive has a better chance to show off its capabilities and does so by posting the quickest time we have seen, even though this test is also becoming bottlenecked by our test platform hardware.

Whether for breaking world records or simply turning an already fast PC into the pinnacle of performance, the Vertex 2 does not disappoint. We haven’t even completed half of our tests but one thing is already clear from our experience with the batch testing, which is that the Vertex 2 delivers some of the most consistent results we have seen to date. Despite the unusual methods the SF-1200 uses to achieve its remarkable performance, in five runs the medium batch test time always fell within a seven second range from previous runs. We will explore more about how the SF-1200 operates along with the full results in the full review, but suffice to say we already know the Vertex 2 is the fastest SSD we have tested to date.

Discuss this article in our forums!

Have a comment you wish to make on this article? Recommendations? Criticism? Feel free to head over to our related thread and put your words to our virtual paper! There is no requirement to register in order to respond to these threads, but it sure doesn’t hurt!

Support our efforts! With ad revenue at an all-time low for written websites, we’re relying more than ever on reader support to help us continue putting so much effort into this type of content. You can support us by becoming a Patron, or by using our Amazon shopping affiliate links listed through our articles. Thanks for your support!

OCZ Vertex 2 — LanOC Reviews

Details
Category: Storage

Even with the price of SSD’s still out of range of some users we see an amazing variety of models giving the most hardcore and wealthy enthusiasts a great selection to pick from. OCZ itself manufactures a model for any price or performance range. We already took a look at their budget/mainstream focused Agility; today we have a chance to put their new Vertex 2 to the test. Will the added cost equal better performance?

Product Name: OCZ Vertex 2

Review Sample Provided by: OCZ

Review by: Wes

Pictures by: Wes

Specifications




  • Available in 50GB to 400GB capacities
  • Available in 60GB to 480GB extended capacities
  • Native TRIM support
  • Max IOPS Firmware
  • Seek Time: . 1ms
  • Slim 2.5″ Design
  • 99.8 x 69.63 x 9.3mm
  • Lightweight: 77g
  • Operating Temp: 0°C ~ 70°C
  • Storage Temp: -45°C ~ +85°C
  • Low Power Consumption: 2W in operation,
    .5W in standby
  • Shock Resistant up to 1500G
  • RAID Support
  • Included 3.5″ Desktop adapter bracket
  • Compatible with Windows XP, Vista, 7, Mac OS X and Linux
  • MTBF: 2 million hours
  • 3-Year Warranty

50-240GB Max Performance

  • Max Read: up to 285MB/s
  • Max Write: up to 275MB/s
  • Sustained Write: up to 250MB/s
  • Random Write 4KB (Aligned): 50,000 IOPS

400-480GB Max Performance

  • Max Read: up to 250MB/s
  • Max Write: up to 240MB/s
  • Sustained Write: up to 200MB/s
  • Random Write 4KB (Aligned): 22,500 IOPS

Packaging

The packaging for the Vertex 2 has a picture of the SSD on the front with the Vertex 2 logo in a nice metallic. Around back they have a small specifications listing including the max performance you should expect and anything else you may need to know. Inside the drive came in a paper and foam tray just like the agility before it. Of course this time around the back of the tray had a 3.5 inch to 2.5 inch adapter tucked away.

Installation

With the included adapter OCZ really has made installing the Vertex 2 easy in any rig. The mount included also has a nice OCZ logo next to the mounted drive, so you wouldn’t want to install a competitor’s drive in this mount. They even included the screws needed to mount everything together in case you don’t have a collection of extra screws sitting around someplace like most enthusiasts.

Performance

To put the SandForce SF-1200 controller in the Vertex 2 to the test, I put it through our normal barrage of tests CrystalDiskMark, ATTO, HDTune, and 3dMark Vantage. We did have some issues with inconstant numbers on our test rig but were able to get repeatable results using both H55 and x58 test rigs. Because of that I am sure that the problem is on our end and I will be sorting out the issues on the test rig. Enough about our issues and the delay in publishing this review, let’s take a look at the results..

As you can see the Vertex topped the charts with everything we have tested to date. Even though the performance numbers we used to compare are from HDtune and are a little lower than you would expect. The Vertex 2 performed to what we expected with CrystalDiskMark and ATTO. The 249128 read speed in ATTO is only bested by its write speeds at 260616, very impressive! The fast write speeds are where this drive really shines. You can see that same performance with CystalDiskMark coming in with a very respectable 258.3 MB/s write speed to go with its 239.1MB/s read speed. What the fast write speeds mean is that you won’t have any bottlenecks unlike some other SSD’s. Read performance is important, but fast writes will speed up file transfers, software installations, and even game saves.

The reason OCZ is able to get that edge in performance with the Vertex 2 comes from a few of its features, namely the SandForce SF-1200 SATA-2 controller. The SF-1200 is a one chip design that removes the DRAM buffer to speed things up. The SF-1200 also includes native Trim support along with support for S.M.A.R.T. and Native Command Queuing (NCQ) with 32 command slots. Together they keep a close eye on your performance to keep it in tip top shape and to let you know if there is an upcoming problem. The SF-1200 controller also supports a 5-year customer life cycle with OCZ warranting the first three years. This means you can expect over 2 million operating hours of use out of the Vertex 2. By the time you have to worry about it the Vertex will be long out of date in this fast moving world.

Overall

Having put the Vertex 2 to the test, I was amazed to see its outstanding overall performance. The write speeds were especially impressive when compared to drives we have tested in the past. Although their Agility series is known for its mid-range pricing, the Vertex is not far off when compared by their price per Gb. Combine the great performance with a reasonable price and you have the makes of a great buy. I would recommend the Vertex 2 for anyone looking for top notch speed with a reasonable price.

  • Plextor PX-B120U

  • RunCore Pro-V 200GB SSD

OCZ Vertex 2

OCZ Vertex 2 60GB SSD review 

OCZ Technology manufactures a model for any price or performance range. We already took a look at their budget/mainstream focused Agility; today we have a chance to put their new Vertex 2 to the test.

We have in our lab’s the 60GB OCZ Vertex 2 SATA II 2.5″ Solid State Drive (SSD). The drive promises a performance in 4k file writes up to 50,000 IOPS and it is rated to last an extra 500,000 hours beyond the 1.5 million hour MTBF of the original Vertex. Building on the features of its predecessor, the Vertex 2 incorporates the Sandforce 1222 controller design and features transfer rates up to 285MB/s read and 275MB/s write.

Built with SATA 3Gb/s, the Vertex 2 Series easily integrates into today’s mobile and desktop platforms and features TRIM support to optimize performance over the drive’s lifespan as a Windows 7 notebook or desktop upgrade. The drive also comes backed by a 3-year warranty.

Features:— Native TRIM support
— Max IOPS Firmware
— Seek Time: 0.1ms
— Slim 2.5″ Design
— 99.8 x 69.63 x 9.3mm
— Lightweight: 77g
— Operating Temp: 0°C ~ 70°C
— Storage Temp: -45°C ~ +85°C
— Low Power Consumption: 2W in operation, 0.5W in standby
— Shock Resistant up to 1500G
— RAID Support
— Compatible with Windows XP, Vista, 7, Mac OS X and Linux
— MTBF: 2 million hours


The OCZ Vertex 2 60GB SSD (OCZSSD2-2VTXE60G) retails online for about $90 and it is backed by a generous 3-year warranty.

The Vertex 2 uses a sleek dark design. On the front of the package is a picture of the drive as well as the capacity and some of the specifications. On the back of the package there is a more detailed list of the specifications and the features of the Vertex 2.

Included with the Vertex 2 drive is the 2.5″ to 3.5″ converting bracket, which allows you to mount the drive into your desktop chassis, the mounting screws and an information guide: 


The OCZ Vertex 2 uses a SATA II interface for the data transfer. The maximum transfer speed on SATA II interfaces is 3Gbps. Next to the SATA data interface is the SATA power interface to provide the low power needed to run the Vertex 2 drive. 
(The drive uses 34nm NAND MLC flash, the current standard for this class of SSD. Thesolution is based on the Sandforce SF-1222 controller, which has been proved to offer high reading/writing speeds.) 

The SF-1200 is a true single-chip SSD Processor designed for volume manufacturing and high performance. The DuraClass technology delivers reliability, endurance, performance and power consumption:

As each generation of flash memory evolves and the silicon geometries shrink — performance, reliability, endurance, and data retention are negatively impacted. DuraWrite technology extends the life of the SSD over conventional controllers, by optimizing writes to the flash memory and delivering a write amplification below 1, without complex DRAM caching requirements.

SandForce DuraClass technology enables the SSD to maximize both the endurance and performance for the life of the drive.

The SF-1200 has also built in AES-128 bit encryption controllable by a configurable user password. This feature prevents would-be thieves from extracting data directly from the flash memory should they ever have access to the drive.

Summary

The OCZ Vertex 2 series is completely taking advantage of the capabilities of the Sandforce 1222 controller, to offer a great write and read performance. Our benchmarks showed that the drive is able to reach transfer rates up very close to the promised 285MB/s for read and 275MB/s for write, while it also delivers more than 40K IOPS for 4KB random reads and writes. The drive comes in a 2.5″ format and a SATA II interface. With such an increased performance, running two of these drives in RAID 0 would almost bottle neck an ICh20R 3.0GBps controller. In addition, with the supported TRIM function under Windows7, you will get the best performance and keep your SSD running good as new with little user interaction.

Compared to the OCZ Agility 2 60GB offering, which is available at the same price tag, the Vertex 2 60GB is offering a little bit higher small file random write performance, so generally choosing any of these drives will most probably not make significant performance difference, as least for desktop users. Still, if you have a workload that demands it, the Vertex 2 doesn’t disappoint.

Spending $90 for 60GB of storage is generally to much, but that’s the situation in the SSD market. So under the currently available SSD options and if your PC is not equipped with a 6GBps SATA interface, the Vertex 2 is a good buy in order to boost your PC’s performance.

Enthusiasts with more patience could also wait for the first SSDs that will be equipped with the upcoming Sandforce SF-2000 controllers, who are promising to deliver 500MB/s sequential reads and writes and up to 60K IOPS for 4KB random reads and writes. OCZ is expected to release the Vertex 3 series based on SandForce’s SF-2582 controller later this year.
Positive

  • Very fast reading/writing
  • 3 years warranty
  • Very active forums and constant support with firmware upgrade
  • Good price
  • Native TRIM Support
  • Up to 50,000 IOPS in random 4K aligned writes
  • Includes the 3.5″ desktop adapter
  • Upgradable firmware via OCZ Toolbox

PC Hardware & Custom PC Rating : 10 out of 10

If you like to see an unboxing, click on the video link below

OCZ Vertex 3 240GB SSD review — Storage

OCZ’s Vertex 2 was arguably our pick of the bunch when it came to high-performance SATA 3Gbps SSDs, so we’ll be first to admit that we’ve been expecting great things from its SATA 6Gbps successor — the Vertex 3.

First unveiled at the Consumer Electronics Show back in January, the Vertex 3 SSD makes use of a second-generation SandForce controller that hopes to fully utilise — if not saturate — the bandwidth available to the SATA 6Gbps interface.

OCZ’s officially quoting mind-boggling sequential read and write speeds of up to 550MB/s and 525MB/s, respectively, while 4K random write performance is rated at 60,000 IOPs. Putting those numbers into perspective, a last-generation Vertex 2 was capable of read speeds of up to 285MB/s, write speeds of up 275MB/s and 4KB random writes at up to 50,000 IOPs.

In terms of pure sequential speed, the jump from Vertex 2 to Vertex 3 sees a performance increase of over 90 per cent. Such is the speed of the new-generation drive that the 600MB/s peak throughput of a SATA 6Gbps interface starts to feel a whole lot less future proof than it did just a few months ago.

So how has OCZ managed to eke out an almost twofold increase in performance? Outwardly, very little has changed. Our £430, 240GB pre-production review sample looks a lot like any other 2.5in SSD — it’s small, lightweight, costly and armed with two connectors; a SATA power connector and a SATA 6Gbps interface that’s backward compatible with SATA 3Gbps.

The SATA 6Gbps interface is one piece of the puzzle, but most of the magic is found under the hood…

With the aluminium enclosure removed, we get a look at the chip that enthusiasts the world over have been lusting over — the SandForce SF-2281.

Located at the heart of OCZ’s Vertex 3 PCB, the controller touts a bevy of enhancements (more on those shortly) and comes accompanied by 16 Micron 16GB 25nm MLC NAND flash memory devices (eight on the front, eight on the back) in an eight-channel configuration.

That provides the 240GB model of the Vertex 3 with 256GB of physical memory, so where’d the additional capacity go? Like most other SSDs, a portion of the NAND flash is set aside to preserve long-term performance and longevity. In the case of the 240GB Vertex 3, 16GB (just over six per cent) is reserved for over-provisioning and when the remaining space is translated to a usable formatted capacity in Windows, you’re left with 223GiB.

A closer look at SF-2281

OCZ and SandForce have historically enjoyed a special relationship, but it’s worth noting that the SF-2281 controller isn’t exclusive to OCZ and the Vertex 3 won’t be the only drive to use it. Although OCZ’s interpretation is expected to be first to market, it’ll be followed by a horde of SandForce-driven competitors in the coming weeks, so let’s take a closer look at what the new-generation SandForce controllers have to offer.

  SF-1222 SF-2141 SF-2181 SF-2281 SF-2282 SF-1565 SF-2582 SF-2682
Target Market Client Client Client Client Client Enterprise Enterprise Enterprise
SATA Interface 3Gbps 3Gbps/s 3Gbps 6Gbps 6Gbps 3Gbps 6Gbps 6Gbps
Sequential Read* (MB/s) 250 250 250 500 500 250 500 500
Sequential Write* (MB/s) 250 250 250 500 500 250 500 500
IOPS* (Sustained 4 KB Reads) 30K 60K 60K 60K 60K 30K 60K 60K
IOPS* (Burst/Sust 4 KB Writes) 10K 60K/10K 60K/20K 60K/20K 60K/20K 30K 60K/60K 60K/60K
SMART Attributes Basic Basic Basic Basic Basic Extended Extended Extended
Flash Channels/Byte Lanes 8ch/16bl 4ch/4bl 8ch/8bl 8ch/8bl 8ch/16bl 8ch/16bl 8ch/16bl 8ch/16bl
Flash Interface A A / S A / S A / S A / S A A / S A / S
Package Type BGA-361 BGA-256 BGA-256 BGA-256 BGA-400 BGA-361 BGA-400 BGA-400
Package Size (mm) 13×13 14×14 14×14 14×14 14×14 13×13 14×14 14×14
Max Capacity (GB) 512 64 256/512 256/512 512 512 512 512
ECC (max bits/512B sector) 24 (RS) 55 (BCH) 55 (BCH) 55 (BCH) 55 (BCH) 24 (RS) 55 (BCH) 55 (BCH)
AES Encryption (bits) 128 256 & 128 256 & 128 256 & 128 256 & 128 128 256 & 128 256 & 128
Military Erase No No No No No Yes Yes Yes
iTemp Option Yes No No No No Yes Yes Yes
non-512 Byte Sectors No No No No No No No Yes
Power Fail Circuit Support No No No No No Yes Yes Yes
Minimum Over Provisioning 7% 0% 0% 0% 0% 28% 28% 28%
* Up to. ..
** A = Asynchronous, S = Synchronous

As illustrated by the above table, SandForce has a range of SF-2000 controllers making their way to the market, including enterprise-specific variations tailored toward optimum data protection and superior 4KB random/write performance.

OCZ will employ the SF-2582 in its own enterprise-orientated drive — the Vertex 3 Pro — but we’re more interested in the comparison between the controllers used in OCZ’s high-end consumer drives; namely the SF-1222 used in the Vertex 2 and the SF-2281 in the follow-up Vertex 3.

Starting with reliability, SF-2281 brings a couple of noteworthy upgrades to the table including TCG OPAL security with 256-bit AES encryption (up from 128-bit) and an improved ECC engine capable of correcting up to 55 bits per 512-byte sector.

Furthermore, all SandForce SF-2000-series controllers continue to utilise the company’s patented DuraWrite technology — an on-the-fly compression algorithm that reduces the size of the data written to the drive. With a typical write amplification of 0.5x, 20GB of data can require as little as 10GB of flash writing. Couple that with the controller’s intelligent block management and wear leveling, you’re left with a drive that requires fewer write cycles during regular use.

This is becoming increasingly important as NAND flash memory can withstand a finite number of program-erase cycles before the wear begins to deteriorate the integrity of the storage. That number tends to decrease with each die shrink, and with 25nm NAND flash offering on average less than 5,000 program-erase cycles, technologies such as DuraWrite are essential to endurance and reliability.

And the controller’s data-protection capabilities don’t stop there. SF-2281 continues to offer support for RAISE (Redundant Array of Independent Silicon Elements), a RAID-like protection mechanism that can utilise a single NAND die to protect against flash block failures. OCZ’s Vertex 3 is over-provisioned to support the technology, but what’s interesting is that SandForce is now allowing manufacturers to disable RAISE completely and reclaim the additional capacity — an option OCZ doesn’t yet appear to be exploring.

There’s a lot going on inside this tiny controller, but what’ll matter most to many enthusiasts are the improved performance numbers. Armed with a SATA 6Gbps interface, SF-2281 is capable of churning out sequential speeds of up to 500MB/s and sustained 4KB writes at 20,000 IOPs. The official numbers are quick, but they don’t quite match up to the specification of the Vertex 3, so what’s going on?

OCZ’s special sauce?

Cast your memory back to Vertex 2, and you might remember that OCZ employed a custom firmware that boosted the performance of its SF-1222 drive beyond that of its competitors. Will we see a repeat of that process with the SF-2281? To find out, we asked OCZ if the Vertex 3 would utilise an exclusive firmware and were told that «perfected would be a better description».

Not the most cryptic of answers, and the performance optimisations are clear to see; the standard SandForce firmware should limit the SF-2281 to a sustained 4K write performance of 20,000 IOPs, but OCZ’s drive touts figures of up to 60,000 IOPs. Whether or not these performance gains will be available to other SandForce partners remains to be seen, but for now let’s see what the Vertex 3 is really capable of.

IOPS

In the neutrality of this article is disputed . A related discussion can be found on the talk page. Please do not delete this post until the conditions for doing so are met. (August 2016) (find out how and when to delete this message template)

input / output per second ( IPS , brazen ) is an input-output performance measurement used to characterize computer storage devices like hard disk drives (HDDs), solid state drives (SSDs), and storage area networks (SANs). Like benchmarks, IOPS numbers published by storage device manufacturers are not directly related to actual application performance. [1] [2]

Content

  • 1 Background
  • 2 Performance characteristics
  • 3 Examples
    • 3. 1 Mechanical hard drives
    • 3.2 Solid -unit 2
  • 4 See also
  • 5 Recommendations

von

In order to meaning the performance characteristics of any storage device, it is necessary to simultaneously indicate at least three indicators: IPS , response time and workload (applications). In the absence of both response time and workload indications, IOPS is meaningless. In isolation, IOPS can be thought of as analogous to the «rpm» of a car engine, i.e. an engine capable of spinning at 10,000 rpm with its transmission in neutral does not transmit anything of value, however an engine capable of delivering a given torque and power in horsepower at a given number of revolutions fully describes the capabilities of the engine.

The exact number of IOPS possible in any system configuration will vary greatly depending on the variables the tester puts into the program, including balance of reads and writes, mix of sequential and random access patterns, number of worker threads, and depth queues, as well as the sizes of data blocks. [1] There are other factors that can also affect IOPS results, including system setup, storage drivers, OS background operations, etc. In addition, when testing SSDs in particular, preconditions need to be taken into account. [3]

Performance

Random versus sequential access.

The most commonly measured performance characteristics are sequential and random operations. Sequential access operations to locations on a storage device in a continuous manner and are usually associated with large amounts of data transfer, for example 128kB. Random access cells for storage access operations are not contiguous and are typically associated with small transfer sizes, such as 4kB.

The most common operational characteristics are as follows:

Dimension Description
Total IPS The total number of input-output operations per second (when performing a reading and recording tests) ,000
Average random read IOPS
Random write IOPS Average random write IOPS
Settling reading IPS The average number of input and output operations per second in
Squirrel record IPS Average Supporting Entry Recording Operations in seconds

devices, the random IOPS numbers are primarily dependent on the random number of storage device operations. seek time, while for SSDs and similar solid state storage devices, random IOPS numbers are primarily dependent on the speed of the internal storage controller and the speed of the memory interface. On both types of storage devices, consecutive IOPS numbers (especially when using a large block size) typically indicate the maximum sustained bandwidth that the storage device can handle. [1] Often consecutive IOPS are logged as a simple MB/s number like this:

IOPS×TransferSizeInBytes=BytesPerSec{ displaystyle { text{IOPS}} times { text {TransferSizeInBytes}} = { text{BytesPerSec}}} ( response is usually converted to Megabytes per second)

Some hard drives will improve performance as the number of pending I/Os (i.e. queue depth) increases. This is usually the result of more advanced controller logic on the drive performing command queuing and reordering, commonly referred to as either Tagged Command Queuing (TCQ) or Native Command Queuing (NCQ). Most commercial SATA drives either cannot do this, or their implementation is so poor that there is no visible performance improvement. [ citation needed ] Enterprise SATA drives such as the Western Digital Raptor and the Seagate Barracuda NL will improve by almost 100% with large queues. [4] High-end SCSI drives, which are more commonly found in servers, typically show much greater improvement, Seagate Savvio exceeding 400 IOPS — more than double its performance. [ citation needed ]

Although traditional hard drives have roughly the same IOPS for read and write operations, most NAND flash-based SSDs are much slower to write than to read due to the inability to overwrite directly to a previously written location , calling a procedure called garbage collection. [5] [6] [7] This has caused hardware test sites to start providing independent measurement results when testing IOPS performance.

Flash drives such as the Intel X25-E (released in 2010) have much higher IOPS than traditional hard drives. In a test done by Xssist, using Iometer, Random Transfers 4K, Read/Write Ratio 70/30, Queue Depth 4, IOPS provided by the Intel X25-E 64GB G1 started at around 10,000 IOPS. IOPS and dropped sharply after 8 minutes to 4000 IOPS and continued to gradually decrease over the next 42 minutes. IOPS varies from 3000 to 4000 for about 50 minutes and beyond for the remaining 8+ hours of the test. [8] Even with the drop in random IOPS after the 50th minute, the X25-E still has much higher IOPS than traditional hard drives. Some SSDs, including the OCZ RevoDrive 3 x2 PCIe using the SandForce controller, have shown much higher consistent write performance that more closely matches read speeds. [9]

Examples

Mechanical hard drives

The block size used in testing significantly affects the number of IOPS performed by a given drive. Below are some typical performance indicators: [10]

9000 — 135 — 135 — 135 — 135 — 135 — 135

9000 97 — 123

9000 9000 73.5 — 127.5 — 127.5 ° C.

9000 9000 — 143

9000 — 76

9000 — 63 9000 9000.3 24.3 — 32.19000 9000 43.4 – 97.8

Solid -state devices

Drive (Type / rpm) IPS

(Block 4 KB, Random)

IPS

(Block 64 KB, random)

MB/s (64K block random) IOPS

(512K block random)

MB / s (block 512 KB, random) MB / s (large block, consistent)
SAS / 15k 188 — 203 175 — 192

11.2 — 12.3 58.9 — 68.9 91.5 — 126.3
FC / 15k 163 — 178

151 — 169 9,000 — 10.8 49.7 — 63.1 FC / 10K 142 — 151 130 — 143 8.3 — 9.2 80 — 104 40.9 — 53.1 58.1 — 107.2
SAS / 10K 141 8. 3 —3 —3 9.2 80 — 104 40.9 — 53.1 58.1 — 107.2
SATA / 7200

73 — 79 4.4 — 4.9
SATA / 5400 57 55 3.5 44 22.6

9000 9000 9000 Examples . Please improve the article by adding more descriptive text and removing less relevant examples. See the Wikipedia guide to writing the best articles for further suggestions.

90 000 000 000 000 000 000 000 000 read IOPS, 135,000 write IOPS [23]

[28] [29] [30] [31]

30-

360 000 Write IOPS [35]

9000 9000 9000 Single SSD OCZ SuperScale Z-Drive R4 PCI-Express

9000 + 550,000 + 550. 000 + 550,000 + Settle

9000 9000

9000 3U 3U 3U 3U 3U 3U 3U 3U 3U 3U 3U 3U 3U 3U 3U 3U 3U — 7U 1007 — 1447 W (nominal) 95 lb (43.1 kg) fully loaded + 44 lb per expansion shelf Flash

Intel X25-E (SLC) SSD ~ 5000 IOPS [13] SATA 3Gb/s Intel Passport

[14] 30031 and 35,000 IOPS for writing and reading, respectively. The mixture measures 5,000 IOPS. The Intel X25-E G1 has approximately 3 times the IOPS compared to the Intel X25-M G2. [15]
G. Skill Phoenix Pro SSD ~ 20,000 IOPS [16] SATA 3Gb/s SandForce -1200 SSDs with enhanced firmware show up to 50,000 IOPS, but benchmarks show for this particular drive ~25,000 random read IOPS and ~15,000 random write IOPS. [16]
OCZ Top 3 SSD to 60,000 input and output operations per second [17] SATA 6 Gbts / C Random Record 4 KB (with alignment)
Corsair Fories GT 9000 9000 9000 SSD to 85 to 85 to 85 000 IOPS [18] SATA 6 Gb/s Disk 240 GB Sequential Read 555 MB/s Sequential Write 525 MB/s Random Write Test 4 kB (aligned)
Samsung SSD 850 PRO SSD 100000 IOPS Read

IOPS Write

[19]

SATA 6Gb/s Aligned Random I/O4 KB at QD32
10,000 read IOPS, 36,000 write IOPS at QD1
550 MB/s sequential read, 520 MB/s sequential write on 256 GB models and above
550 MB/s sequential read, 470 MB/s sequential write on 128 GB model [19]
STROLROOGRAL DRIVER MEMBLAZE PBLAZE5 910/916 NVMe [20] 1000K of arbitrary reading (4 KB) IPS 9000 303 KB of arbitrary recording (4 KB) IPS

0 9000 9000 9000 9000 9000 NVMe)

Performance data taken from PBlaze5 C916 NVMe SSD (6. 4TB).
OCZ Pinnacle 4 SSD Up to 120,000 IOPS [21] SATA 6 Gb/s Drive 256 GB, 560 MB/s sequential read and 510 MB/s sequential write, random read 4 KB test 90 KB IOPS, random write 4 KB test 85 KB IOPS
( IBM) Texan-20 120000+ RASTRAILARTS DRIVATIONS / Recording IOPS [22] PCIe includes Kesh RAM
Fusion-IODRive SSD PCIe
Virident Systems TACHION SSD 320,000 sustained read IOPS with 4KB writes per 0 IOPS and 200 I/O sustained IOPS Using blocks 4 KB [24] PCIE
OCZ Revodrive 3 X2 200000 4K IPS [25] 9000

PCIE PCIE
Solidfire SF3010 / SF6010

SSD ,0005 250. 000 4 Kb of Operations 32] ISCSI MASSISS based on flash memory (5RA)
STEL HOLDOGROUM ANTEL 750

SSD 440 000 Operations in a second when reading
2

input operations in the second when recording

[33] [34]

NVMe over PCIe 3.0 x4, U.2 and HHHL expansion card Aligned 4K random I/O with four worker processes on QD32 (effectively QD128), model 1, 2 TB [34]
Sequential read up to 2.4 GB / s, sequential write 1.2 GB / s [33]
Samsung SSD 960 EVO operations SSD read IOPS
360,000 write IOPS [35]
NVMe over PCIe 3.0 x4, M.2 Aligned 4K random I/O with four workers in QD4 (effectively QD16), [36] Model 1TB
14000 IOPS in Read IOPS, 50,000 Write IOPS at QD1
330,000 Read IOPS, 330,000 Write IOPS on 500 GB model
300,000 Read IOPS, 330,000 write IOPS on 250 GB models
Sequential read up to 3. 2 GB/s, sequential write 1.9 GB/s [35]
Samsung SSD 960 PRO SSD 440,000 read IOPS
NVMe over PCIe 3.0 x4, M.2 Aligned 4K random I/O with four worker processes in QD4 (effectively QD16), [36] 1 TB and 2 TB models
14,000 read IOPS, 50,000 write IOPS at QD1
330,000 read IOPS, 330,000 write IOPS on 512 GB models
Up to 3.5 GB/s sequential read, 2.1 GB/s sequential write [35]
(IBM) Texan-720 memory systems Flash / DRAM 500 000 Reading Optimal Operations, 250,000 optimal recording operations 4 KB IPS [37] FC / Infiniband
SSD to 500,000 input-output operations per second [38] PCIE
Whiptail, Invicta

SSD Fiber Channel, iSCSI, Infiniband/SRP, NFS Flash Storage Array
Violin Systems

Violin XVS 8

Total Latency Flash Array 6 3RU 9005 | Latency 400 µs at 1 M IOPS | 1ms latency at 2M IOPS LUN — 340,000 IOPS at 1ms Fiber Channel, ISCSI

NVME via FC

Violis system

XIO G4

SSD massive

IOP to: 400000 at 400000 Active, ISCSI 000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 , 8 Gb/s FC2

4 ports per controller

(IBM) Texas RamSan-630 memory systems Device Flash/DRAM 1000000+ 4 KB random read/write IOPS [40] FC / INFINIBAND
IBM FlashSystem 840 Flash / DRAM

1100000+ 4 KB of arbitrary reading / 600000 4 KB recordings IOPS [41]

0 9000 9000 8G FC / 16G FC 10G FCoE / InfiniBand

Modular Storage Shelf 2U — 4TB — 48TB
Fusion-io ioDrive Octal (single PCI Express Card) SSD 1,180,000+ Random Read/Write IOPS 903 300006

PCIE
OCZ 2X SUPERSCALE Z-DRIVE R4 PCI-Express SSD

to 1200000 input-output per seconds [38] 9000 9000 9000 9000 PCIE 000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 9000 Texas RamSan 70 Flash/DRAM 1200000 random read/write IOPS [43] PCIe Includes RAM cache
4 Kaminario 20005 SSD Up to 2,000,000 IOPS. [44]
1.200,000 IPS in the SPC-1 test, imitating the business application [45] [46]
MLC Flash
NetAPP cluster FAS6240

9000. 145 SPECsfs2008 nfsv3 IOPS using 1440 x 15K disks on 60 virtual tiered storage shelves. [47] NFS, SMB, FC, FCoE, iSCSI SPECsfs2008 is the latest version of the Standard Performance Evaluation Corporation test suite for measuring file server throughput and response time, providing a standardized method for comparing performance across vendor platforms. http://www.spec.org/sfs2008.
Fusion-io ioDrive2 SSD Up to 9,608,000 IOPS [48] PCIe Only through demo for now.
E8 Storage SSD to 10 million input and output operations per second [49] 10-100 Gbit Ethernet Flash Capital on the racks
EMC D5000 9000 9000 9000 9000 Up to 10 million IOPS [50] PCIe Out of Box, up to 48 high availability clients. PCIe Rack Scale Flash. The product has been discontinued. [51]
Clean storage M50 Flash to 220 000 32K IPS [52] 16 Gbre Channel 10 Gbit / s Ethernet ISCSI 10 Gbit / s Ports of 1 Gbits / S ports Up to 1.4 million IOPS 9 «Pure Storage Datasheet» (PDF).

OCZ Vertex 460 Solid State Drive Continues a Glorious History

One of the most unexpected developments in the hardware market last year for me was the deal in which Toshiba bought OCZ’s SSD business. The latter at that time was experiencing financial difficulties, so the deal was completed quickly enough. Fortunately, this turned out to be just the case when both companies benefit from the merger. Toshiba has a well-established brand with an excellent reputation (discs continue to bear the proud letters “OCZ” on the case), sales channels and experienced employees, and OCZ now has access to all Toshiba’s developments that play a significant role in the flash chip market. One of the first disks released already with the participation of Toshiba was the OCZ Vertex 460, which we will talk about.

But before looking at a specific drive, let’s take a look at how flash memory works in general.

Back in the middle of the 20th century, the American Bosch Arma Corporation under the leadership of Wen Tsing Chow developed a simple technology for long-term data storage — PROM (programmable read-only memory). Each cell of such a memory was a grid of horizontal and vertical conductors connected at each intersection with a small fusible insert. In order to get the value at a given address, it is enough to check whether the current flows between the corresponding horizontal and vertical conductors. If there is a current, then a unit is stored there, otherwise it is zero. Such microcircuits were also programmed quite simply: high voltage was applied to those intersections that needed to be reset, burning out the jumper. Such a memory could be programmed only once, and it did not differ in compactness, but it is very fast and, in fact, is not afraid of electromagnetic impulses.

The next step was taken in the Intel lab by Dov Froman, who was researching defective microcircuits with destroyed transistors. He invented a new type of memory called EPROM. Each memory cell is a field effect transistor with two gates. The first of them is floating, separated from the entire circuit by a layer of insulator, “above” it there is another gate — the control one. This memory is also programmed by applying high voltage to the desired cells. At the same time, due to the Fowler-Nordheim effect, high-energy electrons overcome the insulation layer and accumulate on the floating gate, the transistor goes into the open state and is considered a unit. Due to the isolation of the floating gate, the charge on it remains for decades. To erase such microcircuits, a quartz glass window was made in them, which made it possible to irradiate the circuit with ultraviolet light. Such irradiation leads to an increase in the energy of electrons, and they leave the gate, zeroing the cells.

In 1978, George Perlegos, also working for Intel, developed chips similar to EPROM, but with a thinner gate insulation layer, so the chip could be washed without UV exposure. Such chips are called EEPROM or EEPROM (Electrically Erasable ROM). Their disadvantages include a limited number of rewriting cycles (up to a million for modern microcircuits) and gradual data erasure over time (current chips are guaranteed up to 10 years of information storage).

Since EEPROM chips for erasing required the creation of a strong electric field in a thin dielectric, this imposed restrictions on the growth in the density of memory cells on a single chip. This has led to the development of high-capacity write-once chips and much smaller rewritable chips. The solution to the problem was invented in 1984 by Toshiba engineers, and since the process of erasing chips reminded the creators of a flash, the chips were called flash memory. The first commercial samples were released by Intel in 1988 (NOR memory) and in 1989 Toshiba (NAND).

As you can see, Toshiba stood at the very origins of the technology, so the transition of OCZ to chips from Toshiba was expected with great hope. Of course, a company that makes all the components for an SSD in-house will be in a better position than those forced to source components from others, so the OCZ/Toshiba duo makes for a third «unified» manufacturer (besides Micron’s Samsung and Crucial).

  • NAND Chips: 19nm Toshiba Multi-Level Cell (MLC) Flash
  • NAND Controller: OCZ Barefoot 3 M10
  • Interface: SATA 3 6Gb/s4 9004 II Form2 (backward compatible with SATA2 Form2
  • ) -Factor: 2.5 ″, Ultraton 7 mm
  • Dimensions: 99.7 x 69.75 x 7 mm
  • Weight: 113G
  • Consumption: V, 0.6 W, 2.7 WT
  • Operating temperatures: 0°C — 70°C
  • Peak Sequential Read: 540 MB/s
  • Peak Sequential Write: 525 MB/s
  • 4K random write I/O operations:

    IOPS

For a long time I could not remember the difference between the two lines of OCZ disks: Vector and Vertex — all the more confusion was added by the fact that Vertex is translated from English as “top” , but in fact the top of the ruler is just Vector. In the end, I just remembered this fact as a given.

The disc comes in a stylish slim black cardboard box. Inside you will find the drive itself, a funny sticker, an adapter for installing the drive into a desktop case, and a few pieces of paper. By the way, do not rush to throw them away, one of them contains the serial number for Acronis True Image HD, a wonderful program for backing up and restoring data. A nice bonus that can come in handy more than once.

The Vertex 460’s logic is based on the well-established Barefoot 3 controller (more precisely, its slightly stripped-down version of the M10), developed by Indilinx, which OCZ acquired back in 2011. Despite the fact that the controller has been developed for a long time by the standards of the fast-growing hi-tech world, it is still very competitive. The controller is based on the ARM Cortex core and the OCZ Aragon RISC coprocessor. The controller supports the SATA 6 Gb/s standard.

Inside, the SSD looks almost the same as its older brother Vector 150: the controller is in the center, the case acts as a heatsink for the heat sink. Micron chips are used as cache memory, in the 240 GB model, the cache size is 512 MB, in the older 480 GB model, 1 GB is installed. The flash chips are, of course, new from Toshiba, made using the 19 nm process. There are 16 chips in total, 8 on each side of the board, the capacity of each chip is 16 GB. In total, this gives us 256 GB, but only 240 GB is available to the user, the rest is reserved for service needs (in the next article I will tell you where the “extra” memory goes).

The disc is very thin, only 7 mm, making it an ideal choice for a variety of mobile applications.

The manufacturer provides the drive with a three-year warranty, with write volumes up to 20 GB per day. The figures are very reasonable for a budget model, and there is no need to worry about the disk, it will serve its full life.

In general, the transition to Toshiba technology had a positive effect on disk performance. The Barefoot controller works well with Toggle Mode NAND, which is a speed improvement over the ONFI interface used in older OCZ models.

I deliberately did not do a bunch of various synthetic tests, the Internet is full of measurements showing that the OCZ Vertex 460 is not far behind its older brother Vector 150, and the latter, in turn, competes almost on an equal footing with all competitors. The performance of most modern SSD models with a solid margin is enough for any needs of home users, so tests that determine the advantage of certain models in the area, fluctuating around 5-7%, are actually the lot of techno-maniacs and interested people. But since graphics still adorn any text, I will give the results of measuring the speed of copying 5 GB of files from disk to disk and comparing with several other SSDs. As you can see, the results are close to the declared ones.

The case when the first pancake never came out lumpy. I don’t even know if it can be considered the first, since both companies have a lot of experience behind them, but I think the current models are just the beginning of a long journey of companies that have joined forces.

IOPS — what it is and how to calculate it / Sudo Null IT News (HDD), Solid State Drives (SSD) and Network Attached Storage (SAN).

In fact, IOPS is the number of blocks that can be read or written to the media. The larger the block size, the smaller the chunks that make up the file, and the lower the IOPS will be, since reading a larger chunk will take more time.

So, to determine IOPS, you need to know the speed and block size during a read / write operation. The IOPS parameter is equal to the speed divided by the block size of the operation.

Performance data

The main measured quantities are linear (sequential) and random (random) access operations.

Linear read/write operations, in which parts of files are read sequentially, one after the other, refers to the transfer of large files (more than 128 K). Random operations read data randomly from different areas of the media, usually associated with a block size of 4 KB.

The main specifications are as follows:

Description
Total IOPS Total IOPS (both read and write)
Random Read IOPS Average random reads per second
Random Write IOPS Average random writes per second
Sequential Read IOPS Average linear reads per second
Sequential Write IOPS Average linear writes per second

Approximate IOPS

Approximate IOPS values ​​for hard drives.

Device Type IOPS Interface
7,200 rpm SATA drives HDD ~75-100 IOPS SATA 3Gb/s
10,000 rpm SATA drives HDD ~125-150 IOPS SATA 3Gb/s
10,000 rpm SAS drives HDD ~140 IOPS SAS
15,000 rpm SAS drives HDD ~175-210 IOPS SAS

Approximate IOPS for SSDs.

Device Type IOPS Interface
Intel X25-M G2 MLC SSD ~8600 IOPS SATA 3Gb/s
OCZ Vertex 3 SSD ~60,000 IOPS (4K Random Write) SATA 6Gb/s
OCZ RevoDrive 3 X2 SSD ~200,000 IOPS (4K Random Write) PCIe
OCZ Z-Drive R4 CloudServ SSD ~1,400,000 IOPS PCIe

RAID penalty

Any read operations that are performed on disks are not subject to any penalty since all disks can be used for read operations. But everything is reversed with write operations. The number of write penalties depends on the type of RAID selected, for example.

In RAID 1, in order for data to be written to disk, there are two write operations (one write per disk), and therefore RAID 1 has two penalties.

In RAID 5, there are 4 operations to write data (Read existing data, RAID parity, Write new data, Write new parity), so the penalty in RAID 5 is 4.

This table shows the penalty value for more commonly used RAID configurations .

RAID I/O Penalty
RAID 0 1 (Edited by Reader)
RAID 1 2
RAID 5 4
RAID 6 6
RAID 10 2

Working load rating

The workload characteristic is generally considered as the percentage of reads and writes that the application generates or requires. For example, in a VDI environment, the IOPS percentage is considered to be 80-90% for writing and 10-20% for reading. Understanding the characteristics of the workload is the most critical factor, since it determines the choice of the optimal RAID for the environment. Write-intensive applications are good candidates for RAID 10, while read-intensive applications can be placed on RAID 5.

Calculate IOPS

There are two scenarios for calculating IOPS.
One of the scenarios is when there is a certain number of disks and we want to know how many IOPS these disks will give out?
The second scenario, when we know how many IOPS we need, and want to calculate the required number of disks?

Scenario 1: Calculating IOPS based on a certain number of disks

Let’s imagine that we have 20 450GB 15k RPM disks. Consider two workload scenarios 80%Write-20%Read and another scenario with 20%Write-80%Read. We will also calculate the number of IOPS for both RAID5 and RAID 10.

Formula for calculating IOPS:

Total Raw IOPS = Disk Speed ​​IOPS * Number of disks
Functional IOPS =(((Total Raw IOPS×Write %))/(RAID Penalty))+(Total Raw IOPS×Read %) and there are those IOPS that include RAID penalties, and these are the “real” IOPS.

And now let's substitute the numbers and see what happens.

Total Raw IOPS = 170*20 = 3400 IOPS1316
Option 1 (80%Write 20%Read) Functional IOPS = (((3400*0.8))/(4))+(3400*0.2) = 1360 IOPS
Option 2 (20%Write 80%Read) Functional IOPS = (((3400*0.2))/(4))+(3400*0.8) = 2890 IOPS

For RAID-1

Option 1 (80%Write 20%Read) Functional IOPS = (((3400*0.8))/(2))+(3400*0.2) = 2040 IOPS
Option 2 (20%Write 80%Read) Functional IOPS = (((3400*0.2))/(2))+(3400*0.8) = 3100 IOPS

Scenario 2: Counting the number of drives to reach a certain number of VA IOPS

Consider a situation where we need to determine the type of RAID and the number of disks to achieve a certain number of 5000 IOPS and with certain workloads, for example 80%Write20%Read and 20%Write80% Read.

Again, to begin with, the formula by which we will calculate:

Total number of Disks required = ((Total Read IOPS + (Total Write IOPS*RAID Penalty))/Disk Speed ​​IOPS)

Total IOPS = 5000

Now let's plug in the numbers.

Note: 80% of 5000 IOPS = 4000 IOPS and 20% of 5000 IOPS = 1000 IOPS with these figures we will operate.

For RAID-5

Option 1 (80%Write20%Read) - Total Number of disks required = ((1000+(4000*4))/170) = 100 disks.
Option 2 (20%Write80%Read) - Total Number of disks required = ((4000+(1000*4))/170) = 47 disks approx.

For RAID-1

Option 1 (80%Write20%Read) - Total Number of disks required = ((1000+(4000*2))/170) = 53 disks approx.
Option 2 (20%Write80%Read) - Total Number of disks required = ((4000+(1000*2))/170) = 35 disks approx.

Understanding and calculating IOPS, RAID penalties, and workload characteristics is a very critical aspect of planning. When the write load is more intense, choose RAID 10 and vice versa for read loads RAID 5.

IOPS measurement software

IOmeter - IOPS test
IOzone - IOPS test
FIO - IOPS test
CrystalDiskMark - IOPS test
SQLIO - a set of tests for calculating performance (IOPS, MB, Latency) for a database server
wmarow - a RAID group calculator for IOPS performance

Testing a PCI-E SSD drive OCZ Z-Drive R4

See also:
SSD reliability drives and monitoring tools
Testing the Dell PowerEdge PCI-E SSD
SSD solution is not for everyone, or as I talked with Fusion-io ioDrive Duo

IT technologies are developing by leaps and bounds, and fairly young SSD technology is a direct confirmation of this. Not so long ago, the first models of SSD drives only appeared, not so long ago, their prices became acceptable for use in workstations and servers, and now, SSD drives are slowly approaching the limitation of the SATA interface, resting on the available 6 gigabits per second. With a story about OCZ Z-Drive R4, I would like to continue our short review cycle PCI-E SSD drives (at the end of which, we will summarize, compare different models of different storage systems and try to determine what is best to use and for what purposes at the moment). We started with the already rather old Fusion-io ioDrive Duo model, and we will continue with last year's model from OCZ - Z-Drive R4, designed for the corporate segment.

Just the other day, she arrived at our test laboratory and we began to study her with interest.

In our hands was the CM88 model, with a volume of 1.6Tb, and this is what the manufacturer promises us.

MLC memory, SandForce 2282 controller (4 on each board), with declared read/write speed up to 2.8Gb/s.

Specifications:

  • 1.6TB - ZD4CM88-FH-1.6T (usable 1490GB)
  • Max read speed up to 2800MB/s
  • Max write speed up to 2800MB/s
  • Random writes ) 410000 IOPS
  • Random writes (8 kB) 275000 IOPS
  • PCI Express x8
  • NAND controller: 8 x SandForce SF-2200 SSD
  • Dimensions (L x W x H): 242 x 98. 4 x 17.14 mm
  • Weight: 283g
  • Power consumption: 23W Idle, 26W activity

The board is "two-layered", but still takes only one slot, which is convenient and economical.

Compared to Fusion-io ioDrive Duo.

SandForce 2282 controller.

MLC NAND modules manufactured by Intel, made according to the 25nm process technology.

Unfortunately, OCZ does not provide any means to configure the device at all. This is not a Fusion-io ioDrive Duo, in which we can choose the operating mode of the device ourselves. As they say - we use what the manufacturer gives. The only thing we can "manage" is the firmware and BIOS version of the device. To do this, there is an OCZ Toolbox on the manufacturer's website (unfortunately, the version is only for Windows), with which we can update the firmware and BIOS of the card in a matter of seconds.

Well, let's check if this is true. First of all, we started testing with HD Tune as usual, but were extremely surprised by the results shown.

The speed turned out to be lower than after the first run of the test, which indicates that TRIM does not work in the best way for it, i.e. the garbage collector simply "freebies". Of course, we did not try to leave him alone for a few days, perhaps the picture would be somewhat different. As for the dips in speed up to 200Mb/s, this can only be explained by transitions between sets of memory modules (we observed exactly the same picture when testing ioDrive, apparently this is normal for such large drives).

Well, let's leave the test results as they are, because from the previous review we have already been able to conclude that not all tests can unequivocally show the performance of the disk, so let's proceed to our tests.

AIDA64 Disk Benchmark


Linear reading speed, extremely pleasing to the eye.

But the speed of random reading leaves much to be desired. Here Z-Drive even loses to the RAID array from the younger brothers OCZ Vertex 4.

Buffered reading is also on top, leaving competitors far behind. Good result, not great, but extremely good.

But the access time is also not happy at all, in comparison with io-Fusion - everything is extremely sad, apparently the rather slow MLC memory says (although it is thanks to it that the device has the lowest price on the market).

CrystalDiskMark


ATTO Disk Benchmark

AS SSD


More than pleasant final results. The speed, of course, is far from the declared 2.8Gb / s, but we have long been accustomed to dividing marketing figures by 2. The result is extremely good, and as tests in real conditions will show, these are not just numbers.

Iometer


File copy/ripping


Conclusions

Pros:

  • The performance of the device is at a very high level, not yet available for other types of disk subsystems.
  • Relatively low cost of 1Gb of disk space compared to competitors.
  • This drive can be bootable, i.e. you can install the system on it and use it as the only disk in the PC / server. Not all PCI-E SSD drives can boast of this.

Cons:

  • MLC NAND
  • SandForce 2282

Well, we can conclude that the drive fully justifies the expectations placed on it. It did not disappoint us with either synthetic or real tests. If we return to our article about Fusion-io ioDrive Duo, we can conclude that technologies do not stand still and for 2 years of work, manufacturers have done a good job on their products, especially since competition in the market is becoming more aggressive every day and more and more manufacturers trying to bring their products to market.

2024 © All rights reserved