Per-Entity Metadata Working Group

[Nicholas Roy <>]

On 9/6/16 5:16 PM, Tom Scavo wrote:
> On Tue, Sep 6, 2016 at 6:51 PM, Scott Koranda 
> <>
>  wrote:
> > [...]
> > If anyone disagrees with that simple analysis please let me
> > know.
> I don't disagree but...
>
> > The continued discussion on latency and gathering more numbers
> > and more data is fine, but I don't see that they change
> > anything about the approach we will recommend.
> >
> > Again, if anyone disagrees please say so, but my conclusion
> > after seeing the numbers and understanding how clients will
> > cache (per our earlier discussions) is that an InCommon MDQ
> > service built on a commercial CDN is going to work well.
> I'm not convinced the CDN layer is necessary. My initial timings
> suggest the CDN does not provide huge benefit beyond what mdq-beta
> alone can provide. More testing is needed, however.

The CDN also has the virtue of offloading the traffic from the Ann Arbor 
and redundant site networks.

Nick

> Tom