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

Re: proposed row ops requirements



Nit alert...

(a) Make SNMP efficient so that retrieving complete tables or portions
    of potentially large tables is aproximately as fast as retrieving the
    same information via other channels (e.g. FTP or CLI).

I think the goal for performance should be of the same order, with a negligable or small constant.
Setting the bar for SNMP to perform as fast or faster seems unreasonable as a goal.

\\Mark

Juergen Schoenwaelder wrote:

>>>>> Sharon Chisholm writes:

Sharon> My problem with the way to (a) is written, is that I don't
Sharon> usually want the entire table - I only want some of the
Sharon> columns.  If we focus on efficiency for getting the entire
Sharon> table, then we may not actually gain anything.  If I get
Sharon> something 4 times more efficient, but I have to retrieve 4
Sharon> times the information to use the feature, then I'm exactly
Sharon> where I started.

So what about this:

(a) Make SNMP efficient so that retrieving complete tables or portions
    of potentially large tables is at least as fast as retrieving the
    same information via other channels (e.g. FTP or CLI).

/js

--
Juergen Schoenwaelder      Technical University Braunschweig
<schoenw@ibr.cs.tu-bs.de>  Dept. Operating Systems & Computer Networks
Phone: +49 531 391 3289    Bueltenweg 74/75, 38106 Braunschweig, Germany
Fax:   +49 531 391 5936    <URL:http://www.ibr.cs.tu-bs.de/~schoenw/>