Articles: Storage

Bookmark and Share

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 ]

Degradation and Steady-State Performance

Unfortunately, SSDs are not always as fast as in their “fresh” state. In most cases their performance goes down after some time and in real life we deal with completely different write speeds than what we see on the diagrams in the previous chapter of our review. The reason for this phenomenon is the following: as the SSD runs out of free pages in the flash memory, its controller has to clear memory page blocks before saving data into them, which causes substantial delays. Although, modern SSD controllers can alleviate the performance drop by erasing unused flash memory pages ahead of time, when idle. They use two techniques for that: idle-time garbage collection and TRIM.

Of course, users are more interested in the consistent performance of their SSDs over a long period of time rather than the peak speed they are going to see only during the initial short-term usage period, while the drive is still “fresh”. The SSD makers, however, declare the speed characteristics of “fresh” SSDs for marketing reasons. That’s why we decided to test the performance hit that occurs when a “fresh” SSD becomes a “steady” one.

To get a complete picture of SSD performance degradation we ran special tests based on the SNIA SSSI TWG PTS (Solid State Storage Performance Test Specification) methodology. The main idea of this approach is to measure write speed consecutively in four different cases. First we measure the “fresh” SSD speed. Then we measure the speed after the SSD has been fully filled with data twice. The third test occurs after a 30-minute break during which the controller can partially restore performance by running the idle-time garbage collection. And finally, we measure the speed after issuing a TRIM command.

We ran the tests in synthetic IOMeter 1.1.0 RC1 benchmark, where we measured random write speed when working with 4 KB data blocks aligned to flash memory pages at 32 requests queue depth. The test data were pseudo-random. The following diagram shows the history of the relative speed changes, where 100% refers to the SSD performance in “fresh-out-of-box” state.

SandForce-based SSDs suffer a performance hit when they run out of free memory pages. They slow down with use due to poor garbage collection algorithms. This is actually the controller’s problem because it can be observed with any firmware. The performance hit has become less severe with faster flash memory as is indicated by the Intel SSD 335 as well as the Corsair Force GS (which is equipped with 24nm Toggle Mode NAND flash). The Intel SSD 330 would slow down by 15% when writing with a long request queue whereas the Intel SSD 335 only slows down by 10% when writing 4KB data blocks with a long request queue.

Anyway, we wouldn’t recommend using SandForce-based SSDs, including Intel ones, in operating systems without TRIM support. Otherwise, you may find yourself disappointed with the eventual and noticeable deterioration in the performance of your SSD.

Since the characteristics of most SSDs do change once they transition from fresh out-of-the-box state into steady state, we measure their performance once again using CrystalDiskMark 3.0.1 benchmark. The diagrams below show the obtained results. We use random data writing and measure only performance during writes, because read speed remains constant.

Intel’s SSDs are closer to each other in terms of steady-state performance, but the SSD 520 proves its higher positioning by being slightly faster than the SSD 335 and SSD 330.The two affordable drives are similar in this test and do not impress much compared to their opponents. Well, second-generation SandForce controllers have been around for over 2 years already, so it’s no wonder that they fail to compete with newer solutions. Intel’s exclusive firmware cannot save the day, so the only good news for the SF-2281 controller is that it is good in typical desktop usage scenarios that involve a lot of reading.

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 ]


Comments currently: 19
Discussion started: 02/06/13 04:21:14 AM
Latest comment: 01/07/17 08:44:48 AM

View comments

Add your Comment