The issue relates to the size and the locking mechanism applied to NSS session pools, by default. This leads to poor performance when server-type workloads are applied.
Unfortunately I don't have any performance numbers to share. On 14 Jun 2013, at 19:13, Matthew Hall wrote: > On Fri, Jun 14, 2013 at 06:38:16PM +0100, Vincent Ryan wrote: >> NSS may be initialized to favour performance or to favour memory footprint. >> This fix introduces a new configuration flag to allow Java applications to >> choose. By default, NSS will be initialized for performance. > > The original Sun bug just says, "Non optimized initialization of NSS crypto > library leads to scalability issues". > > But it doesn't say what the scalability issues were, and how bad they were, > and how well it ran without and with the fix. I'm just curious about the > specifics. > > Matthew.
