Re: [DISCUSS] Replace Customized Map with ConcurrentHashMap

2023-09-06 Thread mattison chao
> Moreover, this process should not consume excessive time. Simply initiate > the script, await the deployment, and execute the test. I have personally > performed this procedure numerous times with success. :) That will be cool. Looking forward to it. btw, I am not sure if I remember correctly. D

Re: [DISCUSS] Replace Customized Map with ConcurrentHashMap

2023-09-06 Thread PengHui Li
Hi Qiang, Lastly, it would be beneficial to conduct a test on the distributions as the ConcurrentHashMap has demonstrated improved efficiency. However, a drawback is the potential increase in JVM GC overhead. If we can assess the final outcome of the Pulsar benchmark, the findings would be more pe

Re: [DISCUSS] Replace Customized Map with ConcurrentHashMap

2023-09-06 Thread mattison chao
Hi, Guys. IMO, It's a data structure difference. We can give the benchmark of two structures in different situations (pulsar using situations). > we can built a distribution for the benchmark. After we get the result, > then we can decide if it's fine to go to the new solution. The change should

Re: [DISCUSS] Replace Customized Map with ConcurrentHashMap

2023-09-06 Thread PengHui Li
I also support the idea. For the execution. I think we can have a branch to apply the changes and then we can built a distribution for the benchmark. After we get the result, then we can decide if it's fine to go to the new solution. The change should also be applied to bookkeeper. And I can help

Re: [DISCUSS] Replace Customized Map with ConcurrentHashMap

2023-09-06 Thread Yunze Xu
I support replacing it with JDK's ConcurrentHashMap. Maintaining a customized concurrent hash map whose algorithm is essentially the same with the implementation of a very old version of JDK is painful. The PROs listed like no boxing and linear probing have not proved to be better by any benchmark.

RE: [DISCUSS] Replace Customized Map with ConcurrentHashMap

2023-09-05 Thread the tumbled
On 2023/09/06 04:22:58 QQ wrote: > Hi Pulsar Community, > > I’d like to start a discussion about whether replacing the customize util > class like ConcurrentOpenHashMap with ConcurrentHashMap, as the performance > of ConcurrentHashMap is better than those customize util significantly. > Worse

[DISCUSS] Replace Customized Map with ConcurrentHashMap

2023-09-05 Thread QQ
Hi Pulsar Community, I’d like to start a discussion about whether replacing the customize util class like ConcurrentOpenHashMap with ConcurrentHashMap, as the performance of ConcurrentHashMap is better than those customize util significantly. Worse, these customize util class cannot ensure cons