[Nfd-dev] File retrieval performance

[Klaus Schneider]

On 05/16/2017 12:06 PM, Lixia Zhang wrote:
>
>> On May 16, 2017, at 10:59 AM, John Scott <scott1091 at yahoo.com
>> <mailto:scott1091 at yahoo.com>> wrote:
>>
>> Hi all,
>>
>> I'm testing a file transfer in NDN between two linux containers. For
>> this scenario I'm using ndnchunks, but I noticed the performance is
>> well below the link capacity.
>
> yes.
> the code has not been tuned for performance.

The default of ndnchunks uses a fixed pipeline. You can try one of the 
AIMD algorithms (ndncatchunks -h) to improve performance.

Best regards,
Klaus


>
>> Are there ways to optimize the ndnchunks application using built-in
>> functions?
>
> wonder what exactly is the goal of your experiments?
>
> TCP/IP code, after 30+ years of tuning, runs way faster than NDN for
> *pumping bits*.
>
>> Thanks,
>> JS
>> _______________________________________________
>> Nfd-dev mailing list
>> Nfd-dev at lists.cs.ucla.edu <mailto:Nfd-dev at lists.cs.ucla.edu>
>> http://www.lists.cs.ucla.edu/mailman/listinfo/nfd-dev
>
>
>
> _______________________________________________
> Nfd-dev mailing list
> Nfd-dev at lists.cs.ucla.edu
> http://www.lists.cs.ucla.edu/mailman/listinfo/nfd-dev
>