Need for 'Network Coding' simplified code

[Dav]

Hi

I'm looking for a simplified code that treat the approach of Network coding in order to consolidate the theory behind.

Thnx

[nilamo]

https://scapy.net/

[Dav]

https://scapy.net/

Thank you so much nilamo for this tool, it shows that it can do a lot of things.
Actually, I search for something already done, I mean a working code from which I can follow and learn how the 'network coding' can be performed.

Again, thank you

[Skaperen]

do you mean "network" as in communication (TCP/IP or X.25) or "network" as in mathematical graphing and interconnection?

[Dav]

do you mean "network" as in communication (TCP/IP or X.25) or "network" as in mathematical graphing and interconnection?

In my case I'm interested in algorithmic implementation simply. There are: server (the source of data), clients (request data) and intermediate machines (process the data packets sent from server towards client in special way: encoding & decoding). All of this must be within a well-defined topology called "butterfly". here is another detail about it (https://www.youtube.com/watch?v=B0ZcAWEvjCA)

Thank you

[Skaperen]

i would put this under information theory. even though the application for it is network as in communication, it is really not that much of a network thing. this looks a lot like some of the new, more advanced, higher RAID storage levels that are trying to have the error correction and detection of redundancy, while also having as near as possible to 1:1 storage capacity, all with no reduction in performance and perhaps even an improvement by having the next block located where its operation can be performed parallel to its previous block. variations of this are now, also being made to reduce write wear of flash storage.

i think the good places to find code for any of these will be in the published papers. but i would also google for blogs and other postings by those not publishing formal papers. i have seen some of this, and it has me excited for the future (some of which i will not see, given my age and health). but, i have not recorded any of this.

[Dav]

i would put this under information theory. even though the application for it is network as in communication, it is really not that much of a network thing. this looks a lot like some of the new, more advanced, higher RAID storage levels that are trying to have the error correction and detection of redundancy, while also having as near as possible to 1:1 storage capacity, all with no reduction in performance and perhaps even an improvement by having the next block located where its operation can be performed parallel to its previous block. variations of this are now, also being made to reduce write wear of flash storage.

i think the good places to find code for any of these will be in the published papers. but i would also google for blogs and other postings by those not publishing formal papers.

Good idea, thank you !!. The concept of network coding has been tackled by the scientific community about one decade ago and since that there are much scientific papers and codes but not in Python (not found yet )

... (some of which i will not see, given my age and health).

Don't say that. I wish you all the best

[Skaperen]

Python has been around for a long time (almost 3 decades) but its big surge has only been in the past decade or so. i would not expect much scientific code to be in Python until these days. science and math have focused on coding more in Fortran (for numerical science work) and languages like Java and C++ when they code for practical use. a lot of coding is for pure theory in pseudo-code and i think you'll see a lot of that. i suspect you will need to recode a lot of what you do find, and may need to create original code in a lot of cases. given the ultimate use of such logic will be high performance routers, you could very likely see final code in CPU and even GPU-like machine code. Python would be a good choice for simulations as well as management logic running real routers in a management processor. some very high-end routers run on custom hardware (a friend of mine used to design custom chips for one such manufacturer). but i can still see this kind of thing being done even on low-end routers in end-user networks.