Don’t you love a great report card? The other day, our marketing folks rolled up all the benchmarking we did comparing Emulex OneConnect OCe10102 with the other second generation CNA in a report card. We got straight A’s. If I had gotten this report card in high school my Dad would have bought me a bicycle.
Each grade was based on a comparisons based on a quotient of the competitors performance divided by Emulex’s performance. It is kind of like setting Emulex OneConnect as the performance standard. For example, IOPS at 512 Block Size for QLogic would be 250131/919268 which equals 27%. The same benchmark for Emulex would be 919268/919268 or 100%.
The report card reflects all protocols – FCoE, iSCSI and TCP/IP and all performance measures Input Output Operations per Second IOPS, to transfer rates, to CPU efficiency. Please refer to the table below for a handy-dandy guide to benchmarking.
These benchmarks leverage the results of the IT BrandPulse: Unified Data Center Networking — Emulex Unveils First in New Class of UCNAs paper plus benchmarks performed at Emulex Labs using a combination of IXIA Chariot Management Console Version 7.0 and IOMETER. For more information, on the configurations and detailed charts, download the report card.
|
Definitions |
|
|---|---|
|
IOPS |
I/O Operations Per Second (also known as IOPS) are often referred to as small-block I/Os. They generally range in size from 512 byte to 8k and are a staple of database, e-mail and supercomputing applications. IOPS have a known performance profile of raising CPU utilization from a combination of CPU interrupt and wait times. |
|
Transfer Rates |
A transfer rate is the amount of data that can be transferred on a specific technology in a specific time period. In storage testing, the transfer rates are usually described in megabytes or gigabytes per second; MB/s and GB/s respectively. Transfer rate is critical to many applications, but some primary examples are: backup and restore, continuous data protection, RAID, video streaming, file copy and data duplication. |
|
CPU Efficiency (based on IOPS) |
This metric examines the ratio of IOPS divided by average CPU utilization. This ratio illustrates the efficiency of a given technology in terms of CPU utilization. Higher numbers of CPU efficiency show that the given technology is friendlier to the host system’s processors. Higher bandwidth or IOPS with lower CPU utilization is the desired result. This is important, as users are trying to maximize their investments, and CPU utilization is much better spent on CPU-hungry applications. |
|
CPU Efficiency (based on transfers) |
This metric examines the number of transfers that can be completed based on one percentage of the server CPU capacity. This is important to users who are trying to maximize their investment, and CPU utilization is much better spent on hungry applications. |
The purpose of our blog are to facilitate an ongoing conversation on what's going on in our industry, with our partners and customers. We encourage your comments. Your ideas and feedback are what makes our blogs interesting, timely and useful for our readers.
We want to publish your comments, however, all comments are moderated. Offensive, off-topic or fraudulent comments won't be approved. We also expect a basic level of civility; disagreements are expected, but mutual respect is a must. We will not post comments that contain vulgar or abusive language; personal attacks of any kind; or offensive terms that target specific ethnic or racial groups. Comments that make accusations will also not be posted.
By submitting a comment, you agree to these terms; having your name displayed with your comment and that you are 18 years old or older. Your name and personal information will not be used for any other purpose, and your e-mail address will not be published.