Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : The New QLogic (ANCR) -- Ignore unavailable to you. Want to Upgrade?

To: George Dawson who wrote (13111)	12/11/1997 9:53:00 AM
From: Craig Stevenson	Read Replies (1) \| Respond to of 29386

George, I learned something new yesterday regarding the Amoco installation. (Where FC replaced ATM.) The ATM gear that was replaced was 155 Meg, not 25 Meg as I had thought. The reasons for replacement had more to do with ATM's complexity than any throughput issues, although I'm sure the extra 100 megabits per second was a nice bonus. I've also been doing a lot of research into the 512 byte vs. 2048 byte payload issue. Suffice it to say that if my calculations are correct, this is absolutely a non-issue. Consider this quote from Brocade's "Achieving Maximum Effective Throughput in a Fibre Channel Fabric" technical note on their web site: "So, 240 bytes of overhead are needed to transmit 8Kbytes of data. This is minimum overhead, because other delays within a system can lengthen the gaps between transmitted frames. For this discussion, minimum gaps are assumed in order to explore minimum overhead, and thus maximum throughput." This is an interesting quote, because it totally ignores latency, intentionally! (That's what those "other delays" are. <g>) If latency is factored in, my calculations show the MKII as being 6.8% faster than SilkWorm when both switches are using the same payload size. If SilkWorm uses a more efficient 2048 byte payload, and the MKII uses a 512 byte payload, the improved speed of the MKII almost compensates for the fact that four times as many frames must be sent. The difference is only 1.6%, in favor of SilkWorm. The whole topic of payload sizes is probably a moot point, since the MKII does support 2048 bytes payloads, but it was an interesting excercise to see just how the two switches compare. In my opinion, Ancor currently has the technological edge. Now, they have to sell it to customers. Craig