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

Re: [RE] Updated paper



David V James wrote:
> Varuni,
> 
> Thanks for the review and most insightful comments.
> A quick response follows.
> 
> 
>>>Is there anything I'm missing that makes this different from regular 
>>>rate monotonic scheduling?
> 
> 
> Yes. In this case, the full capacity can be allocated but the
> guaranteed latency increases. While I believe the guaranteed
> latency increase is on the order of 20%, I'm not sure of the
> theory.
> 
James,

I failed to notice this on my first reading of your whitepaper..

I guess

"Assuming the cumulative traffic is limited to less than the link 
capacity, the latency of each traffic class is guaranteed (the latency 
guarantee is approximately an MTU more than an inter-arrival period)."

is the admissibility criteria

Do you happen to have a reference that validates this statement?

On a similar vein I suspect this scheduling algorithm has a well known 
name within the network community and it would be helpful if you used it 
as well.

Regards
Varuni