SSD is not for everyone, or how I talked with Fusion-io ioDrive Duo
We are already used to the fact that everyone is talking about the advantages of SSDs (also remembering to intimidate them with faults in terms of fault tolerance), and we are used to the fact that SSDs are “fast but expensive.” So, one of these days just one of such expensive models of SSD fell into my hands, in particular Fusion-io ioDrive Duo. Of course, this is already a fairly old model, which appeared on the market back in 2009, but it can still be found on sale. So why is this thing so good, and why were there so many enthusiastic screams in the IT press about the fastest SSD (which it certainly was 3 years ago)? Let's take a look at her today.
The first minus "opened" in the first minutes after installation - this drive is not detected by the motherboard at all, that is, it cannot be bootable, and it is impossible to put the system on it, but you can use it only as an additional storage for data. For example, the same PCI-E SSD from OCZ are quietly detected by the motherboard and can be used as bootable ones.
Well, we will look further ... Fortunately, the second minus did not take long: in addition to the driver, which was found on the manufacturer’s website only after a long registration form, we needed firmware compatible with the version of this particular driver, as well as a separate utility Dell IO Manager (why Dell, for me personally, remains a mystery, because IO Data supplies its models for both Dell and HP, and our copy was just HP), to manage and maintain this drive (by the way, with rather meager functionality). So, it took 1.5 hours, but so far on our own to get this thing to work and see the disk failed. (and some Western colleagues on their own could not cope with this task and contacted the manufacturer’s support).
And here is our hero, in all its glory, so to speak.
But all these shortcomings could be forgiven him if he showed himself in work, because the manufacturer promises us amazing speed characteristics:
As can be seen from the table, our model 1.28 TB uses MLC memory (manufactured by Samsung), but which one the controller is installed in it, for some reason the manufacturer kept silent. Although the autopsy showed that it contains two FPGAs (programmable, logical integrated circuit) Virtex-5, manufactured by Xilinx .
Bearing in mind that this drive contains 2 controllers and accordingly is defined as 2 different disks, I really wanted to find in it the possibility of creating a RAID array of levels 1 and 0, but ... And then we were cruelly disappointed - software RAID only .
Let's get down to the most interesting part - the tests, and check what he is capable of, what peaks of reading and writing he will submit to? Due to the fact that it is impossible to create a “hard” raid array from it, and the soft raid is not perceived by testing programs, each of the “sections” will be tested separately.
Testing was conducted on a Dell PowerEdge R710 server .
Dell PE R710 Server Configuration:
Let it not surprise you that the volume of each of the partitions is 320GB, I set the maximum write performance in the disk settings.
Fusion-io supplies its disks for both Dell and HP , and each company has its own disk manager. In the screenshot above, this is Dell IO Manager , and HP's manager is the web interface - HP IO Accelerator Management Server .
On the manufacturer’s website, along with the latest drivers, lies Dell IO Manager , and HP IO Accelerator Management Servercan be found on the HP website respectively. It is worth considering that at the time of testing the driver on the HP website was version 3.1.1, which was still listed in the Private Version on the official website (as I understand it, it was something like the Beta version), and Current Stable was 2.3.10
Well, as a starting testing points I will give the performance of our RAID10 on 6xSAS 146GB 15k:
Well, now the performance of two partitions of our experimental ioDrive Duo
Wewere extremely surprised at such a low disk performance. Rollback and reinstallation of drivers did not give any result! Moreover, the disk showed the same performance in all operating modes Maximum Write Performance and Maximum Capacity. We even had time to think that there might be a problem with the PCI-E port on our Dell server, so it was decided to replace the test platform. Important factors for the replacement were: the HP sticker on the disk itself, as well as more recent drivers on the HP website, which are not available to mere mortals on the developer's website. Our new test platform was already familiar from comparative tests - HP DL380 G7 , in the following configuration:
To begin with, the same HD Tune test was launched to check if there is any difference ....
And we saw it!
The difference is almost 2 times! Here's what drivers can do. This is how it looks on the diagram:
Please note that the total disk capacity is 512GB, that is, in the Maximum Write Performance configuration, it "eats away" not only 50%, but only 20%, which is also very good. Of course, failures on one of the sections do not look very good, but there is nothing to be done - the lack of TRIM support makes itself felt. Now we have seen that everything is in order with the disk, and you can proceed directly to testing, omitting unnecessary conversations.
AIDA64 Disk Benchmark
CrystalDiskMark
AS SSD
Of course, to the speeds declared by the manufacturer - almost like to the moon, but we have long been accustomed to the fact that marketing numbers still quite differ from the realities.
In the near future, we plan to conduct comparative testing of several SSD drives, including both conventional SATA and RAID, as well as PCI-E from other market players, and we will compare all this with arrays from ordinary 10 / 15k SAS disks. Moreover, many horror stories about SSD drives being unreliable have long passed into the myths section, and they work no less than regular HDDs. Given the warranty periods, you can safely rely on these products, but you still should not forget about backups!
Posted by KorP
The first minus "opened" in the first minutes after installation - this drive is not detected by the motherboard at all, that is, it cannot be bootable, and it is impossible to put the system on it, but you can use it only as an additional storage for data. For example, the same PCI-E SSD from OCZ are quietly detected by the motherboard and can be used as bootable ones.
Well, we will look further ... Fortunately, the second minus did not take long: in addition to the driver, which was found on the manufacturer’s website only after a long registration form, we needed firmware compatible with the version of this particular driver, as well as a separate utility Dell IO Manager (why Dell, for me personally, remains a mystery, because IO Data supplies its models for both Dell and HP, and our copy was just HP), to manage and maintain this drive (by the way, with rather meager functionality). So, it took 1.5 hours, but so far on our own to get this thing to work and see the disk failed. (and some Western colleagues on their own could not cope with this task and contacted the manufacturer’s support).
And here is our hero, in all its glory, so to speak.
But all these shortcomings could be forgiven him if he showed himself in work, because the manufacturer promises us amazing speed characteristics:
As can be seen from the table, our model 1.28 TB uses MLC memory (manufactured by Samsung), but which one the controller is installed in it, for some reason the manufacturer kept silent. Although the autopsy showed that it contains two FPGAs (programmable, logical integrated circuit) Virtex-5, manufactured by Xilinx .
Bearing in mind that this drive contains 2 controllers and accordingly is defined as 2 different disks, I really wanted to find in it the possibility of creating a RAID array of levels 1 and 0, but ... And then we were cruelly disappointed - software RAID only .
Let's get down to the most interesting part - the tests, and check what he is capable of, what peaks of reading and writing he will submit to? Due to the fact that it is impossible to create a “hard” raid array from it, and the soft raid is not perceived by testing programs, each of the “sections” will be tested separately.
Testing was conducted on a Dell PowerEdge R710 server .
Dell PE R710 Server Configuration:
- Processor - Xeon E5607
- Memory - 16GB
- 6xSAS 146GB 15k RAID10 Hard Drives
Let it not surprise you that the volume of each of the partitions is 320GB, I set the maximum write performance in the disk settings.
Fusion-io supplies its disks for both Dell and HP , and each company has its own disk manager. In the screenshot above, this is Dell IO Manager , and HP's manager is the web interface - HP IO Accelerator Management Server .
On the manufacturer’s website, along with the latest drivers, lies Dell IO Manager , and HP IO Accelerator Management Servercan be found on the HP website respectively. It is worth considering that at the time of testing the driver on the HP website was version 3.1.1, which was still listed in the Private Version on the official website (as I understand it, it was something like the Beta version), and Current Stable was 2.3.10
Well, as a starting testing points I will give the performance of our RAID10 on 6xSAS 146GB 15k:
Well, now the performance of two partitions of our experimental ioDrive Duo
Wewere extremely surprised at such a low disk performance. Rollback and reinstallation of drivers did not give any result! Moreover, the disk showed the same performance in all operating modes Maximum Write Performance and Maximum Capacity. We even had time to think that there might be a problem with the PCI-E port on our Dell server, so it was decided to replace the test platform. Important factors for the replacement were: the HP sticker on the disk itself, as well as more recent drivers on the HP website, which are not available to mere mortals on the developer's website. Our new test platform was already familiar from comparative tests - HP DL380 G7 , in the following configuration:
- Processor - Xeon E5606
- Memory - 4 GB
- Hard Drives - 2xSAS 146GB 10k
To begin with, the same HD Tune test was launched to check if there is any difference ....
And we saw it!
The difference is almost 2 times! Here's what drivers can do. This is how it looks on the diagram:
Please note that the total disk capacity is 512GB, that is, in the Maximum Write Performance configuration, it "eats away" not only 50%, but only 20%, which is also very good. Of course, failures on one of the sections do not look very good, but there is nothing to be done - the lack of TRIM support makes itself felt. Now we have seen that everything is in order with the disk, and you can proceed directly to testing, omitting unnecessary conversations.
Synthetic tests
AIDA64 Disk Benchmark
CrystalDiskMark
AS SSD
Of course, to the speeds declared by the manufacturer - almost like to the moon, but we have long been accustomed to the fact that marketing numbers still quite differ from the realities.
In the near future, we plan to conduct comparative testing of several SSD drives, including both conventional SATA and RAID, as well as PCI-E from other market players, and we will compare all this with arrays from ordinary 10 / 15k SAS disks. Moreover, many horror stories about SSD drives being unreliable have long passed into the myths section, and they work no less than regular HDDs. Given the warranty periods, you can safely rely on these products, but you still should not forget about backups!
Posted by KorP