TEST BENCH AND PROTOCOL
In testing the 200GB Kingston E100, we’ll be using Red Hat Enterprise Linux and/or CentOS, as we do for much of our enterprise testing to date.
Linux has less overhead and is generally more flexible when it comes to evaluating performance. That said, our enterprise test bench is OS agnostic.
We’ll apply a few new standardized testing techniques in addition to some of our older test protocols.
We want to isolate and explore the individual performance of the review drives as accurately as we can.
As the test bench evolves, we hope the result is a more tangible, relevant performance evaluation.
A special thanks to Asus, Crucial, OCZ, and Fractal Design for sponsoring our Enterprise Test Bench.
S.M.A.R.T. DATA
SMART, or Self-Monitoring, Analysis and Reporting Technology data, yields insights into drive health and behavior. Originally, SMART was used as a method of predicting drive failure for mechanical HDDs. In the age of solid state storage, SMART data can tell you how much a drive has read, written, how many blocks have been removed from service, temperature, etc. Some SSDs have virtually no useful attributes, while others are quite comprehensive. Since there aren’t really industry standards governing SMART attributes, each manufacturer’s information is different.
The SandForce 2582 Flash Storage Processor, LSI’s fancy name for their ASIC, isn’t much different from its 2281/82 cousins found in every consumer ‘LSI SandForce Driven’ SSD. Moving up to the enterprise versions nets you much improved S.M.A.R.T. data, support for power-loss-protection schemes, temperature protection, and mil-spec secure erasure. Both consumer 228x and enterprise 258x silicon support 55 bits/sector error correction and AES encryption. The 2282 and 2582 variants possess wider 16 byte lanes on their controller’s 8 channels, while the 2281 has to make due with 8 byte lanes/channel. With wider lanes, more flash can be adequately addressed.
Our E100 came flashed with SandForce Reference 510ABBF0 FW. As it turns out, Kingston has identified a TRIM stability issue with LSI SF’s 510-series firmware, one which could lead to the drive dropping out on an issued TRIM command. As a result, Kingston would prefer that no one use the E100 in TRIM environments. That explains the sticker which covers the SATA signal and power connectors and most of the rear chassis:
For enterprise-oriented drives, this isn’t a very big deal. Basically, if a drive is on a HBA, RAID card, in a RAID array, in an enclosure, or otherwise not being used on a single motherboard SATA port (or in RAID 0 with Intel RST drivers 11.5 or higher). Previous issues with SandForce firmware and TRIM-related instability often took some time to occur (if at all), and since all of our SATA and SAS testing is performed with a LSI 9207-8i HBA, we wouldn’t have experienced the issue anyway.