[Ndn-interest] Maximum Throughput in NDN

[Chengyu Fan]

Hi Junxiao and César,

On Mon, Feb 4, 2019 at 7:05 AM Junxiao Shi <shijunxiao at email.arizona.edu>
wrote:

> Hi César
>
> I was wandering if somebody has already analyzed the maximum Throughput
>> that can be achieved with NDN.
>>
>
> NDN is a protocol. It can theoretically achieve unlimited throughput.
> When IP was first invented, its throughput was a few kilo bits per second.
> In 2019 we are going to see 400Gbps IP routers.
>
> has someone already set a transport layer over Ethernet, UDP, TCP and then
>> transmitted some large files?
>>
>
There are several factors that may impact the NFD throughput, including the
number of name components in Interests, Data packet size, content store,
congestion control algorithm, the nfd log level, and others.


>
> ndncatchunks file transfer tools achieved
> <https://imsure.github.io/ndn-vegas-report.pdf> a 15~25Mbps goodput with
> ndncatchunks file transfer tool, using his TCP Vegas congestion control
> scheme.
>

@Junxiao Shi <shijunxiao at email.arizona.edu> , could you tell us more about
the test, such as the topology, the machine specification, nfd version, and
CS is used or not?
My recent test uses ndncatchunks running on nfd 0.6.4 (CS size was set to
0) on my local laptop is around 400Mbps.


> NFD's forwarding performance was benchmarked
> <https://redmine.named-data.net/issues/3564#note-24> in 2016 to be 6.8K
> Interest-Data exchanges (tested with 10-byte segments, but if it were
> 1300-byte segments, the goodput is 70Mbps); no newer benchmark reports are
> available.
>

The NFD benchmark was done for version 0.4.2.
I remember there were some improvements after the benchmark, and the
results should be better than 6.8K Interest-Data exchanges.


> The fastest known NDN forwarding performance (not file transfer) has
> exceeded 1333K Interest-Data exchanges (tested with 1300-byte segments,
> goodput is 13.8Gbps), as reported in NDN Community Meeting 2018
> <https://www.nist.gov/news-events/events/2018/09/named-data-networking-community-meeting-2018>
> (day 2 part 3, first 22 minutes).
>

@Junxiao Shi <shijunxiao at email.arizona.edu> , I think you mean the NDN-DPDK
Forwarder that is at the beginning of the last video. It is an
implementation for the dedicated forwarder that contains some new data
structures to help multi-threading.
Is the code for NDN-DPDK open to the public? Is there a timeline for the
release?


>
>
>> In case the answer is negative, is there any suggested configuration for
>> setting up 'state of the art' configuration?
>>
>
> See the NDN Community Meeting 2018 video for hints.
>
> Yours, Junxiao
> _______________________________________________
> Ndn-interest mailing list
> Ndn-interest at lists.cs.ucla.edu
> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest
>


-- 
Thanks,

Chengyu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/ndn-interest/attachments/20190204/79c1eeaa/attachment.html>