[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: OID compression/suppression support




>>>>> Kristine Adamson writes:

Kristine> We have had a lot of complaints from network management apps
Kristine> about their ability to get all the data they want from SNMP
Kristine> efficiently.  And with IPv6 IP addresses in instance values
Kristine> in the new version-neutral MIBs, this problem will only get
Kristine> worse.  Object oriented protocol operations may be the best
Kristine> future direction but, we really need an improvement now, and
Kristine> it will probably be a long time before object oriented
Kristine> operations become a reality.  So, I would like to see the WG
Kristine> invest time on the OID compression/suppression proposals.

My experience with IETF stuff lately is that both options will end up
taking long. By pursuing only one option, the amount of work needed to
be done within the IETF and implementation wise gets reduced. Why
should we not be able to have new PDUs at the same speed we can get
OID compression?

Also note that some of the latency is also due to the fact that the
agent can not optimize access to instrumentation data, access control
etc. because with current PDUs, the manager does not really tell the
agent what he really is interested in (such as "give me this slice of
the following table"). With new PDUs, this can be fixed as well.

/js

-- 
Juergen Schoenwaelder    <http://www.informatik.uni-osnabrueck.de/schoenw/>