CPS Core Performance

Test Environment

...

This should narrow the performance gap in this case.

TODO

Release	Date	CmHandles	100	500	1000	2000	5000	10000	20000	40000	Comments
Kohn	Oct 2022	Time (seconds)	8 sec	16 sec	17 sec	33 sec	581 min	1523 min	ERROR	not testedERROR	Error due to DB running out shared memory
London	May 2023	Time (seconds)	6 sec	7 sec	12 sec	22 sec	621 min	1222 min	61310 min	195532 min
current	Aug 2023	Time (seconds)	6 sec	7 sec	10 sec	16 sec	31 sec	57 sec	71 sec	108 sec

CM-handle registration is multi-threaded, so performance may appear to scale better than linear, until the CPU cores are maxed out.

As can be seen below, CPU usage never reached 100% during the tests , indicating performance is still bottle-neckedof the current version.

Release	Date	CmHandles	100	500	1000	2000	5000	10000	20000	40000	Comments
Kohn	Oct 2022	Time (seconds)	7 sec	1162 min	3957 min	1465	9523	not tested	not tested	not tested	25 min	2.5 hour	est: 10 hour	est: 2 days	est: 7 days	Not tested due to time neededSome values estimated due to time constraints
London	May 2023	Time (seconds)	0.6	1.7	3.2	5.0	16.8	37.2		< 1 sec	2 sec	3 sec	5 sec	17 sec	37 sec	85 sec85.3	ERROR	Error due to 32K limit
current	Aug 2023	Time (seconds)	< 1 sec0.6	1 .4	2.8	3.9	14.2	23.0	65.2	sec	3 sec	4 sec	14 sec	23 sec	65 sec	2 min142.3

Current release has exactly linear performance for CM-handle de-registration (on a single thread):

...