Hello Lixia and all:
I'm a student in Tsinghua University, BeiJing and make bold to submit a
proposal for the current scalability problems of the Internet. This is about
a layered mapping system (perhaps we can call it "LMS") to store and lookup
mapping information of edge addresses and core addresses in the separation
base. It is similar to LISP+ALT while differs from it in a number of ways. I
made a elementary summary and introduced the design in another document.
Summary is here: <http://docs.google.com/View?id=dc7cz575_6gwxds4hb>
http://docs.google.com/Doc?docid=0AQsJc7A4NTgeZGM3Y3o1NzVfNmd3eGRzNGhi&hl=en
Details:
<http://docs.google.com/fileview?id=0BwsJc7A4NTgeMGViYmYzZWMtYjZhYS00ZDc2LWE1NWMtMDNkMWI4YTM3ZDI5&hl=en>
http://docs.google.com/fileview?id=0BwsJc7A4NTgeOTYzMjFlOGEtYzA4OC00NTM0LTg5ZjktNmFkYzBhNWJhMWEy&hl=en
My key idea is similar to the LISP+ALT in adopting a hierarchical
mapping system to guarantee scalability. Initially i thought my
proposal is just a specification of ALT, but later i realize there're
some differences. I also made a comment on
today's proposals and my devised two-stage cache on the ITR can work
for many proposals to enhance the performance.
My experiements haven't added to it but prelimenary results are satisfied.
Could you do me a favor to add this proposal to the rrg-wiki Active Proposal
list? You may have a look at it and discard it if it's not novel at all. I
apologize if it wastes rrg's time.
Best wishes,
Sun Letong
_______________________________________________
rrg mailing list
rrg@irtf.org
http://www.irtf.org/mailman/listinfo/rrg
