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* Frank Coluccio Technology Forum -- Ignore unavailable to you. Want to Upgrade?

To: ftth who wrote (474)	7/19/2000 9:47:24 PM
From: Raymond Duray	Respond to of 46821

Re: IEEE 802.3ae - One more view Hi ftth, Here's EETimes view of the La Jolla event: eetimes.com 10-Gbit Ethernet proposals advance By Craig Matsumoto EE Times (07/19/00, 4:40 p.m. EST) SAN MATEO, Calif. — Participants in the IEEE's 10-Gigabit Ethernet task force reported progress on all fronts at last week's meeting, save for one unresolved choice of physical-layer medium. [This would be the gist of the Lightreading article, eh] The meeting, held July 11 and 12 in La Jolla, Calif., was part of the larger IEEE 802 plenary meeting. The 10-Gigabit Ethernet task force is working under the label IEEE P802.3ae. Participants managed to map out the logical layer of 10-Gigabit Ethernet, but a minor dispute arose around the physical medium-dependent (PMD) sublayer, where the task force voted on which wavelengths of light to be included in the standard and whether those signals should be serial or split into a four-way wave-division-multiplexed scheme. In the end, the task force settled on two transmission options for the physical layer: 1,310-nanometer serial and 1,550-nm serial. A 1,310-nm wave-division-multiplexed (WDM) option, as well as two 850-nm choices, fell short of the three-quarters vote required to pass a technical proposal. Each of the three garnered a majority of votes, however, and the task force, by voting multiple times on different combinations of the three, made an effort to be fair, said Frederick Weniger, product marketing manager of Vitesse Semiconductor Corp.(Camarillo, Calif.). Other PMDs were proposed but failed; the task force appears to have settled on the two options that were approved and the three that received majority votes as the five preferred candidates. Members of the 802.3ae task force said the failure of the three options to win the necessary amount of votes shouldn't be a major roadblock, as it is partly the result of companies jockeying to get favored positions cemented into the specification. Weniger said that some systems providers, for example, would prefer to work with a smaller number of PMDs. "The issue is not whether any of these are feasible," said Jonathan Thatcher, chairman of the 802.3ae task force and an engineer with World Wide Packets Inc. (Spokane, Wash.). "There are different sets of objectives for different groups of the membership, and that doesn't allow them to all align on the same combination." It so happens that the 1,310-nm and 1,550-nm wavelengths are better suited for long-distance transmissions, covering three of the five distance grades that the task force is chartered to include in the 10-Gigabit specification. The contention lies in how to cover the two shorter distances: 300-meter and sub-100-meter "very short reach" connections. The 1,310-nm WDM option would cover both; an 850-nm serial option would handle only 300-meter connections but would use cheaper 850-nm lasers. The 850-nm WDM choice can handle both of the short-reach distances but only with the assistance of an equalizer to combat signal dispersion. A case for this technology was presented at the meeting by Bill Wiedemann of Blaze Network Products Inc. (Dublin, Calif.). Because the shorter-distance options are the ones that failed to pass in the voting, certain members have discussed splitting up the task force, creating a separate group to focus on short-haul and LAN applications. But those views are on the extreme side, Weniger said. "I do not believe for one moment there is a majority interested in breaking things up," he said. "New members become eligible [to vote] as they attend more meetings. So the dynamics could easily change next time, just on the numbers." The remaining PMDs will come up for vote again at the next 802.3ae meeting, which will be held in Boston, Sept. 12 to 14. Papers presented at the July 802.3ae meeting are now available online.