[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Two AES encryption modes?



At 08:15 AM 7/30/02 , Housley, Russ wrote:
>Scott:
>
>Thanks for updating your analysis.
>
>>Speed of Software Implementation:
>>
>>Both transforms can be executed in approximately the same speed.
>>However, both CBC and COUNTER also require authentication transforms to
>>be executed.
>
>In this area, I think that it is worth pointing out that the COUNTER key 
>stream can be precomputed (at least on the encryptor side).  While 
>authentication must be performed when the packet is available, the key 
>stream can be precomputed to allow for reduced latency.

Yes, this is an advantage, and I'll add it to the list the next time I
publish it.  In addition to the possibility for reduced latency, it also
allows you to use CPU idle time to do some of the encryption work.

-- 
scott