Per-Entity Metadata Working Group

[David Walker <>]

(Thanks, Ian, for correcting my misreading.)

Getting back to Nick's original question, I guess I am concerned about the latency times for 12KB objects (~0.4 secs in the US), unless we think InCommon sites have much better connectivity than Frost and Sullivan's average (and that the predominant latency factor is the last-mile network).

This, of course, brings us back to Chris's admonition that we focus on business needs.  What delay is acceptable when an IdP or SP needs to retrieve the metadata for some other SP or IdP?  0.4 seconds sounds pretty long to me, but it is a rare event, we think.

Which brings me back to my offer to analyze some IdP and SP logs to see what the churn of metadata retrievals looks like.  Anyone want to contribute some logs?

David

On 08/26/2016 08:33 AM, Nick Roy wrote:

Thanks Patrick, good points.

Nick

On 8/26/16 9:20 AM, Patrick Radtke wrote:

I think it is unfortunate that they did not compare with a non-CDN
solution, or show the 'backbone' testing so we could see what portion
of the time was on the last mile vs accessing the CDN vs accessing a
non-CDN.

As for distribute large aggregates: any issues I've encountered with
large aggregates are related to memory consumption and signature
verification time on my end, and client processing time is much
greater than download time. I don't think I'd notice changes in
download time for the large aggregate, in particular because it
happens in the background.

-Patrick

Attachment: signature.asc
Description: OpenPGP digital signature