Re: [bmwg] New WG Deliverable Proposal - Core Router Software Accelerated Life Testing

In section 3.4, the I-D proposes "evaluation benchmarks". (Since the
proponents produced an I-D, I'm going to assume the I-D's a more complete
statement of intent and scope. My question is, do these tests and
their collection scenario present good BMWG benchmarks?

Consider the ol' manual, cable pull test or its higher-end relative, the
module hotswap test. Experience tells us that these two tests are some
of the finest system-level tests known to routerkind.  You can even measure
some aspect of "recovery time".  But do they lend themselves to good 
"benchmarks"?

-Kevin

SENGUPTA,ANANDA (A-NewEngland,ex1) wrote:
> Well, this is the reason I was readily agreeing to the work on the nomenclature draft, and then debating what tests fit in this WG in the actual test plan draft. It is clear that there is value to this work, and though all the test cases may not fit in this WG, some will.
> Ananda

_______________________________________________
bmwg mailing list
bmwg@ietf.org
https://www1.ietf.org/mailman/listinfo/bmwg