Web Replication and Caching (wrec)
0================================0
Meeting November 11, 1999 at the 46IETF, reported by Ingrid Melve
Agenda
1. Introductions, agree agenda (5 min)
2. Taxonomy document (10 min) Ingrid / Ian or Gary
draft-melve-wrec-taxonomy-00.txt
3. Known Problems (15 min) John Dilley
draft-ietf-wrec-known-prob-00.txt
4. Research Issues (10 min) Joe Touch
draft-wrec-res-00.txt [NB: If there has been no progress on this
document since July then we need to either drop it, publish it or
move it to the re-chartered WREC].
5. Emerging protocols (15 min) Alberto/Jeremy
draft-cerpa-wrec-necp-00.txt
6. Other related work (10 min) Ivan Lovric
draft-lovric-icp-ext-02.txt
7. Rechartering (30 min)
We need to discuss re-chartering WREC. The mood I have gauged from
the group is that we need to define a charter with some real work
items and development in it. In particular, the areas of:
+ replication methods
+ inter-proxy communication,
+ proxy to network-element communication,
+ proxy / replica discovery
John Martin chaired the group. John introduced the group and pointed
out that this is the last meeting of the working group in its current
form. WREC is proceeding, taxonomy a few months late, rechartering
discussion (NECP draft is out) as the work we set out to do is done.
More than half the participants had read the drafts!
Taxonomy
[2]draft-ietf-wrec-taxonomy-00.txt is documenting current taxonomy for
web replication and caching. Unless there are fundemental problems
with the draft, it should be published closed to its current form.
NECP draft-cerpa-wrec-necp-00.txt is implemented and should be added.
Discussion on the term reverse proxy, consensus that the current
wording covers current use.
Two weeks is the cutoff for comments to the mailing list, otherwise
the document goes to Last Call.
Known Problems draft
No comments recieved the last weeks, and John Dilley considers the
work to be done (John Dilley was unable to attend the meeting and Jay
Kistler presented this part.
Two weeks for comments, then it goes to Last Call.
Research Issues draft
There really is not a document. The input so far has been research
projects, not issues and what to do with it. Want to postphone the
document until the taxonomy is done and discuss what the issues might
be then. Decide to drop this document for now.
Emerging protocols
Network Element Control Protocol (NECP) draft is out as individual
Internet-Draft, presented by Jeremy Elson. L4 switches may use this
protocol for load balancing or intercepting for transparent proxies.
NECP allows the cache and switch to exchange control traffic
What control traffic?
- When server come up, they can tell the switch:
"add me to your group for Service X"
- Servers can send load information; switch does better balancing
- Switches immediately stop sending work to dead servers
using periodic KEEPALIVES
- Transparent Proxy Caches can tell switches to allow direct
connections for certain clients (e.g. on auth failure)
Key features
- minimal
- assumed per-flow state available on switch
- extensible load metrics
- authentication
Non-features
- specific load balancing policies
- IP addresses of friendly servers/caches
- configuration management
The group suggested and the NECP authors agreed to cleaning up the
terminology to be in compliance with the taxonomy draft. This included
referencing taxonomy terms and replacing text with standardized terms.
WPAD is a working group Internet-Draft draft-ietf-wrec-wpad-01.txt
(documenting current protocol) presented by Josh Cohen.
The DNS part of this is not good enough, but this is currently
deployed. Comment on the security consideration section being weak, it
should include PAC non-compliance (PAC files intepreted by client in
inconsistent ways) and DNS searching if not hit in first zone.
Want to move forward with WPAD, issues raised from the AD (Keith)
about IAB/IESG and security considerations.
Rechartering
Suggested starting point:
1. Definition of problem
2. Requirements of a (web) replication architecture
+ replication methods, service vs. content replication
+ loosely coupled vs. tightly coupled
+ content discovery
3. Specific related technology development (e.g.)
+ replica/proxy to network-element communication
+ inter-proxy/replica communication (content distribution)
+ proxy/replica discovery
Agreement was reached to use a top-down architecture. Time expected on
this is 3-4 months. At the same time there will be specific related
technology development that needs to integrated with the solutions.
There was a discussion on rechartering, but no conclusion was reached.
The discussion is to continue on the mailing list.
Joe Touch promised to send outline of architectural requirements to
the list.
Anycast
Anycast may be used for setting up multiple servers at one IP address,
given that the servers reside within the same AS. If the servers are
divided between several different service providers, there scaling
issues will most probably prevent the number of anycast services from
reaching more than ten. Anycast is a local solution for delivering
redundance or for failover, but does not scale in the global Internet.
An interesting proposal for tunnelling Anycast was presented by Dina
Katabi
[3]http://www.ana.lcs.mit.edu/dina/draft-katabi-global-anycast-01.pdf
Towards an integrated web replication infrastructure?
IPv6 took seven years, we can do this in two. The goal is to ask the
network for objects, not to go to specific servers and ask the servers
for objects.
The current model is to configure the client and then afterwards being
configured the client can look for objects. If the configuration
subsitutes a proxy for other lookup mechanisms, there is less
configuration to be done at the client level and more at the proxy.
Object that exist at more than one location, either because they were
cached or because they were replicated should be findable from the
clients. Finding the closest version of a object is difficult today,
unless the client is configured to use a proxy and there is some
intelligence at the proxy.
This archive was generated by hypermail 2b29 : Thu Nov 18 2004 - 11:21:27 MST