AkG
Well-known member
- Joined
- Oct 24, 2007
- Messages
- 5,270
IOMETER Results / Stutter Test
IOMeter is heavily weighted towards the server end of things, and since we here at HWC are more End User centric we will be setting and judging the results of IOMeter a little bit differently than most. To test each drive we ran 5 test runs per HDD (1,4,16,64,128 que depth) each test having 8 parts, each part lasting 10 min w/ an additional 20 second ramp up. The 8 subparts were set to run 100% random, 80% read 20% write; testing 512b, 1k, 2k,4k,8k,16k,32k,64k size chunks of data. When each test is finished IOMeter spits out a report, in that reports each of the 8 subtests are given a score in I/Os per second. We then take these 8 numbers add them together and divide by 8. This gives us an average score for that particular que depth that is heavily weighted for single user environments.
Eeep! The underpowered RAID controller chickens have certainly come home to roost. These numbers are what you would expect from a middle of the road spindle drive….and certainly NOT a high tech SSD. I am not sure why Samsung continues the love affair with JMicron but it really is handicapping their drives. IF you are going to do a RAID 0 SSD….for the love of all that is holy please drop JMicron like the bad apple they are. This is not a cheap drive, yet when the commands start coming fast, furious and totally random the JMB390 runs crying home to momma. This is JMicron and Samsung’s third chance here at HWC and they are obviously missing the boat.
In our usual IOMeter test we are trying to replicate real world use where reads severly outnumber writes. However, to get a good handle on how well a Solid State Drive will handle a worse case scenario (and thus how likely the dreaded stutter issue will happen) we have also run an additional test. This test is made of 1 section at que depth of 1. In this test we ran 100% random. 100% writes of 4k size chunks of information. In the .csv file we then found the Maximum Write Response Time. This in ms is worst example of how long a given operation took to complete. We consider anything higher than 333ms (one third of a second) to be a good indicator that stuttering may happen, with the higher the number the worse the duration of the stutter will most likely be.
While the Titan may totally and completely tank at the regular IOMeter test, at least it passes the Stutter test. 0.4 seconds may not be wonderful but they are certainly a vast improvement over the previous generation…and are probably low enough that any stutters which do occur will be short enough to be imperceptible. Of course, this reduced stutter latency does come at a price, namely the fact that the average latency of the Titan is not all that great. Its not that bad either as less than a tenth of a second is barely perceptible but there is still room for improvement.
IOMETER
IOMeter is heavily weighted towards the server end of things, and since we here at HWC are more End User centric we will be setting and judging the results of IOMeter a little bit differently than most. To test each drive we ran 5 test runs per HDD (1,4,16,64,128 que depth) each test having 8 parts, each part lasting 10 min w/ an additional 20 second ramp up. The 8 subparts were set to run 100% random, 80% read 20% write; testing 512b, 1k, 2k,4k,8k,16k,32k,64k size chunks of data. When each test is finished IOMeter spits out a report, in that reports each of the 8 subtests are given a score in I/Os per second. We then take these 8 numbers add them together and divide by 8. This gives us an average score for that particular que depth that is heavily weighted for single user environments.
Eeep! The underpowered RAID controller chickens have certainly come home to roost. These numbers are what you would expect from a middle of the road spindle drive….and certainly NOT a high tech SSD. I am not sure why Samsung continues the love affair with JMicron but it really is handicapping their drives. IF you are going to do a RAID 0 SSD….for the love of all that is holy please drop JMicron like the bad apple they are. This is not a cheap drive, yet when the commands start coming fast, furious and totally random the JMB390 runs crying home to momma. This is JMicron and Samsung’s third chance here at HWC and they are obviously missing the boat.
IOMETER Stutter Test
In our usual IOMeter test we are trying to replicate real world use where reads severly outnumber writes. However, to get a good handle on how well a Solid State Drive will handle a worse case scenario (and thus how likely the dreaded stutter issue will happen) we have also run an additional test. This test is made of 1 section at que depth of 1. In this test we ran 100% random. 100% writes of 4k size chunks of information. In the .csv file we then found the Maximum Write Response Time. This in ms is worst example of how long a given operation took to complete. We consider anything higher than 333ms (one third of a second) to be a good indicator that stuttering may happen, with the higher the number the worse the duration of the stutter will most likely be.
While the Titan may totally and completely tank at the regular IOMeter test, at least it passes the Stutter test. 0.4 seconds may not be wonderful but they are certainly a vast improvement over the previous generation…and are probably low enough that any stutters which do occur will be short enough to be imperceptible. Of course, this reduced stutter latency does come at a price, namely the fact that the average latency of the Titan is not all that great. Its not that bad either as less than a tenth of a second is barely perceptible but there is still room for improvement.
Last edited by a moderator: