Performance of Shibboleth IDP Ver 2.3.5

[Cantor, Scott]

On 12/3/13, 12:29 PM, "kotesh201" <koteshwarv at gmail.com> wrote:

>Most of the CPU is consumed by following two methods.

I seriously doubt that. Signing takes up 90% of the CPU used by an IdP
doing real work. If your results are accurate, which I doubt, you aren't
doing real world testing.

>Is there any way that we can bring down the CPU usage by these two
>methods.

We have no plans to further optimize performance for a code base in
maintenance mode while we work on the next version.

-- Scott