White paper performance numbers

Question

If this is true: 2060 x 700 Kbps ~= 1.4Gbps
=> Then the claim that the results published in the FMS3
White Paper are limited to 1Gbps throughput (
http://www.adobe.com/products/flashmediaserver/pdfs/FlashMediaServer3_WhitePaper_ue.pdf
, page 17) can not be true!

Can someone comment this please, maybe i am not understanding
the issue correctly BUT the numbers published are very confusing
(pages 16, 17), the complete benchmark for the FMS 3 would be a
benefit to us all...

techeye · Answer

There is reference to 2 bonded nics (why? I have no idea....), but the result of ~1.4 Gbps is consistent with my bonded nic tests on rhel4 (2.6.9-22ELsmp). Fyi, the results with the newer rhel5 kernel (2.6.18) was able to actually get up to ~1.9 Gbps. The newer Linux kernel (& tcp optimizations) make a big difference. I don't know about the results on windows.

But yeah, I agree, apparently whomever was editing this was unable to portray exactly what they meant. They probably wanted to target the average customer, who does not usually use bonded nics (yet). Anyway, the take away, I think, is that performance numbers improved greatly, and that in a typical 1 Gbps configuration one is able to fully take advantage of the entire nic (minus the OS's tcp stack), and in fact that FMS can go even beyond 1 Gbps by utilizing a second nic.

Sign up

To post, reply, or follow discussions, please sign in with your Adobe ID.

Sign in to Adobe Community

To post, reply, or follow discussions, please sign in with your Adobe ID.

Scanning file for viruses.

This file cannot be downloaded