Hi Malaka, lets report these findings on architecture group as a investigation we did. regards Ramtih
On Tue, Jan 26, 2016 at 8:16 AM, Malaka Gangananda <[email protected]> wrote: > Hi, > > While profiling the MB we found out that there is a performance bottle > neck in using util.ConcurrentSkipListMap. Because every time size call is > made it has to loop through the data structure so it takes O(n) time. But > looking at GC ConcurrentHashMap we found out that it only takes constant > time (O(1) time) to make the size call.Other operations on GC > ConcurrentHashMap are also faster than uti.ConcurrentSkipListMap. > > Here I have attached JMH Benchmark results which acquired by testing > util.ConcurrentSkipListMap vs GC ConcurrentHashMap > > > ConcurrentSkipListMap Vs Goldmansachs Concurre... > <https://docs.google.com/a/wso2.com/spreadsheets/d/1y5w1aIDBN89C0ctB83iDOULtYPnB_-3OuGYArd0lbt0/edit?usp=drive_web> > > > Malaka Gangananda > Software Engineer Intern > mobile:0713564340 > [email protected] > <[email protected]> > -- Ramith Jayasinghe Technical Lead WSO2 Inc., http://wso2.com lean.enterprise.middleware E: [email protected] P: +94 777542851
_______________________________________________ Dev mailing list [email protected] http://wso2.org/cgi-bin/mailman/listinfo/dev
