Ivan,
I guess it boils down to whether you are trying to reduce response time or
increase hit rate (at all costs). In other words, I am not sure that
running a complex selection algorithm on a per-query basis is worth the
effort (and would be difficult with today's technology). However, I
understand your requirement for this in a case where (e.g.) there are huge
objects loaded over a satellite and smaller objects loaded from other peers.
At 09:42 AM 04/08/99 +0200, LOVRIC Ivan CNET/DSE/CAE wrote:
> To solve this problem, a peer selection algorithm should know at
>least the size of the requested URL which should be sent back in the
>ICP_OP_HIT message. However, RFC 2186 gives no indication of the object size
>in the ICP_OP_HIT message. So any peer selection algorithm won't work well
>without this information that ICP should return.
I think that returning the size is not unreasonable but I can see probems.
E.g. what if two caches return different sizes for the same object?
> Agreed ! The issue may concern other object discovery protocols. I
>guess it is a good practice to separate object discovery protocols from peer
>selection; however, object discovery protocols should at least give enough
>information (such as object size) to the peer selection algorithm in order
>to make the right choice.
Indeed. There are a lot of different criteria one could apply to peer
selection or object discovery. I don't think that ICP is the right engine
for either, though - it is too lightweight.
> One of the main goals of the caches is to reduce latency to improve
>the QoS from the user point of view; IMO, some work has to be done (within
>the WG) to determine what should be the right level of information to
>transmit between the layer of object discovery protocols and the layer of
>peer selection in order to fully reach this goal.
We will be working on the future of caching and I know a lot of people
share your interest in determining the ``best'' proxy from which to
retrieve a given object - for both end users / clients and proxies
themselves. I expect that WREC will start working on these issues once we
finish our initial phase and it is re-chartered.
Rgds,
John
---------------------------------------------------------------
Network Appliance Direct / Voicemail: +31 654 308 275
Kruisweg 799 Fax: +31 23 567 9699
NL-2132 NG Hoofddorp Main Office: +31 23 567 9600
---------------------------------------------------------------
This archive was generated by hypermail 2b29 : Thu Nov 18 2004 - 11:21:26 MST