Intel, ATTO, CliBench Tests
This content was originally featured on Amdmb.com and has been converted to PC Perspective’s website. Some color changes and flaws may appear.
Click to Enlarge
IOMETER is an I/O subsystem measurement and characterization tool for single and clustered systems. Iometer is pronounced “eye-OM-i-ter,” it measures performance under a controlled load. Iometer is both a workload generator (that is, it performs I/O operations in order to stress the system) and a measurement tool (that is, it examines and records the performance of its I/O operations and their impact on the system). It can be configured to emulate the disk or network I/O load of any program or benchmark, or can be used to generate entirely synthetic I/O loads. It can generate and measure loads on single or multiple (networked) systems. Questionably, Iometer is most reliable and accurate disk measurement performance utility currently available.
Five Seagate X15 II hard drives were used during this benchmark. One X15 II hard drive was used as the boot drive via Adaptec’s on-board controller. The additional four X15 II hard drives were attached and balanced between both channels A and B of the AMI Elite 1600 RAID controller. The illustration above, demonstrates the significant difference in performance between the Elite 1600 and Atto’s UL3D.
Click to Enlarge
Atto’s disk benchmark tests disk speed by reading and writing packets of various sizes to the disk. There is a substantial increase in performance from transfer sizes 32KB to 64KB. From 128KB transfer size and beyond, channel A of the Elite 1600 is completely saturated the U160 specification. However, under a 64-bit/33MHz PCI bus, allows for a maximum transfer rate of 264MB/sec, across both A and B channels. The highlighted area of the illustration demonstrates the peak differences in write and read performance at 1MB between the AMI Elite 1600 and Atto’s UL3D. The Elite 1600 quite remarkably walks away with the lead with Atto’s own disk benchmark utility.
Click to Enlarge
CliBench is a suite of synthetic benchmarks that includes the following benchmarks: Dhrystones, Whetstones, Eight queens’ problem, Matrix operations, Number Crunching, Floating point, memory throughput, hard disk’s throughput and the CPU usage. Again, the Elite 1600 easily outshines the UL3D. Also, there is a tremendous delta between the CPU usage of the Elite 1600 and UL3D. The UL3D’s high CPU usage is a result of it firmware/software RAID support. Whereas, the Elite 1600 is a dedicated hardware RAID implementation.
