Altera Forum






Threads: 18,987
Posts: 77,595
Members: 29,300
Welcome to our newest member, Max10
User
Reputation
9135
7620
5891
4200
3030
2197
2056
1706
1388
1300




 
Register
Quick Search
 
  Altera Forums > Device and Tools Related > FPGA, Hardcopy, and CPLD Discussion

100G Virtual Lanes Number

 
 
Thread Tools Display Modes
Prev Previous Post   Next Post Next
  #1  
Old November 19th, 2009, 06:52 AM
ethstud ethstud is offline
Altera Beginner
 
Join Date: Nov 2009
Posts: 1
Rep Power: 298
ethstud is on a distinguished road
Default 100G Virtual Lanes Number

Dear all, after reading the AN570 document on the implementation of the 40G/100G ETH protocol in Stratix IV devices, I have some doubts related on the number of internal Virtual Lanes (VLs). In the document Altera asserts that for the 100G solution, the internal data flow should be given to 20 VLs, while for the 40G solution only to 4 VLs.

The question is simply, why (when for 40G are only 4 VLs) and for 100G 10VLs are not enough? Is here not possible to connect only 10 VLs to 10 FIFOS and 10 Gearboxes, omitting the MUX/Demux?

In the white paper WP-01080-1.2 "Using 10-Gbps Transceivers in 40G/100G Applications", they explain this number as the least common multiple of the electrical (n=10) and PMD(m=?) lanes. But does not a single PMD lane correspond to one 10G transceiver? So in total only 10 10G Transceiver are required to implement a 100G link, leading therefore to a least common multiple of 10, i.e. 10 VLs.

Thank you for the support
Reply With Quote
 



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Quartus in a virtual machine alt1000 Quartus II and EDA Tools Discussion 9 July 30th, 2008 10:56 AM
virtual FPGA mik General Altera Discussion 0 April 9th, 2008 10:24 AM


All times are GMT -8. The time now is 06:50 AM.