In NDN Pub-Sub approach, node-A has been disconnected from network for several days. After resuming the connection, the same node-A wants to receive all the data packets which were shared during its absence. My question is:
How Node-A can request only for those contents which are unavailable in its storage or shared during its absence?
I would say that there are two approaches, one that relies on naming to minimize the number of duplicate data packets and another that builds in extensions to NDN to explicitly exclude certain data.
For the first approach, imagine streaming a video that is broken up into packets with names like video-name/index/01:02:03 where 01:02:03 contains the packets that whole the video data during the 1 second interval starting at 01:02:03 point of the video. If you know the last complete data packets of the movie are that you read in, then you can request the same appropriate interval based name so you wouldn't have to get all of the video, just the ones during that last second.
For the second approach, see the following paper and the exclude functions:
https://named-data.net/wp-content/uploads/TRstreaming.pdf
It also talks some about the first approach.
Jeff Sedayao Thank for your answer.
The question mark still exists. Actually, I want a subscriber to receive all those packets which are latest and unavailable in the subscriber's CS. A publisher should send all latest packets to the subscriber which are unavailable to subscriber.
One way to solve this issue is to use Pysync which compares the subscriber and publisher list of packets and subtracts the subscriber list from publisher's lists using invert bloom filter. However, this is an end-to-end communication that does not follow the core architecture of NDN.
Regards
Arif Hussain Magsi - NDN is a best effort datagram approach like IP - you need a layer above to guarantee end to end transmission of packets. I don't think that breaks the core NDN architecture - it's just that there are no guarantees.
- Badges
- Science topic
- Similar topics
- Biological Science
- Bioecology
- Systems Ecology
- Landscape Ecology
- Connectivity