Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: Does only decimal system work for data rate specification?





For informational purposes only :-)....

802.3ad specifies a mechanism for aggregating multiple similar speed
ethernet links into a single "aggregate port".  Customers use this today to
provide performance scaling when they don't want to upgrade to the next
higher speed ethernet.  In some cases, customers use a single bundle of
copper to transport both physical ports, and it seems that some clever
implementer or could design a way for customers to use WDM to transport the
parallel paths over a single fiber.

JR


At 09:09 AM 12/9/99 -0700, Mike Wincn wrote:
>
>At 09:55 AM 09-12-99 +0100, Fenghao Mu wrote:
>>1) Why we prefer to only specify 10GE rather than suggest other 
>>reasonable data rates related to gigabit Ethernet, say  2.5/5 gigabit 
>>Ethernet? 
>
>suppose i turn the question around -- what would be the point of having many
>more datarate communication standards, when we already have 1, 10, 100, and 
>1000-base-whatever?  it isn't clear to me that finer granularity offers any 
>useful
>advantage.
>
>>2) Why we need Hari coding to tie four individual links together instead 
>>of sending data separately?
>
>Hari doesn't code anything.  Hari is a proposed method of standardizing
>PMD interconnect that uses existing line codes.  otherwise, the four serial
>links are separate, other than for issues related to word- or byte-striping.
>
>>3) Does our data rate specification fit with market requirement and 
>>technology progress well? 
>
>yes, else there would not be much interest in developing a new standard
>for interoperable, higher-datarate systems.  you should understand that
>no one will be compelled to use any standards-based 10GbE system, 
>other than for desire to participate in that market space.  that has been
>true for all 802.3 stuff, since inception.
>
>>1. From a customer point of view, only 1GE and 10GE give less choices 
>>for one to build a good switching structure. He has to chose between two
>>possible data rates, either 1GE or 10GE. It might be very difficult for him
>>to upgrade his data links, in other words, 10GE is too luxurious and 1GE 
>>is insufficient for his requirement. What is his best solution? 
>
>there is already interest in developing a higher datarate standard.  issues
>such as feasibility, interoperability, and so on are to be resolved in the 
>standards development process.  there will always be technological 
>and dollar cost barriers to general market entry, and though the standards
>process helps to reduce that cost it cannot be removed entirely.
>
>>2. Data rate evenly distributed networks are seldom met in reality, and it 
>>exists only in people's imagination. Designing a switcher fabric based on
>>this imagination will cause a big waste of resource. 
>
>i think you will need to define your use of "wasted resource" more carefully
>before anyone can address this.  do you want 100% capacity, 24 x 7?  you
>won't find it here.  define your conditions and we can proceed.
>
>>[...] If we could use 2.5/5/10 GE data 
>>rates, we can individually assign a data rate for a path in a network 
>>according different requirements. Multiple data rates can be very helpful
>>for optimal I/O pad usage.
>
>you can do that now, at 1, 10, 100, or 1000MBaud. (aside:  i offer this for 
>the sake of completion.  does anyone know whether Starlan is still in use?)
>
>>3. There is no reason why we only specify 10GE, which bonds 4 individual 
>>links together. Probably there isn't any information source or sink today 
>>which has to be specified at 10Gb/s.
>
>if that were true, there would not be much interest in HSSG participation.
>
>>5. At today's technology, few devices could support 10GE within one link
>>for a meaningful distance. 
>
>have you ever heard of OC-192/SDH-48?  you may want to try a search 
>engine.
>
>> From technology point of view, there isn't any
>>remarkable breakthrough so that we can utilize a new and cheap device 
>>to implement 10GE links. 
>
>fast VCSELs, DMD-flattened M-M fiber, fast (linear) equalization, FEC, MAS
>and the like all offer potentially cheap solutions for 10GbE.  you may want
>to have a look at presentation notes at the reflector site.
>
>>6. In fact, we could simply extend the Gigabit Ethernet standard to 
>>2.5/5 GE specification by adapting new speed as a smooth move. If we 
>>use 8B10B encoding, the speed will be 3.125/6.25 Gb/s, respectively. 
>>Simple and efficient! Why not?
>
>i gather that people considered that, early on.  seems there wasn't much
>interest.
>--
>jmw
>
>