Local mode is quite different from how distributed mode would work. I would suggest that you create a distributed setup(with even a few nodes is enough) - Running a Multi-Node Storm Cluster - Michael G. Noll - and then profile the individual workers to find what is consuming the most CPU (which I'm assuming is your intention).
| | | | | | | | | | | Running a Multi-Node Storm Cluster - Michael G. Noll Michael G. Noll Setting up a distributed Storm cluster on RHEL 6 with process supervision | | | On Thursday, August 10, 2017, 3:24:25 AM CDT, shawn.du <shawn...@neulion.com.cn> wrote: #yiv6219386171 font{line-height:1.5;} Hi, We are replaying our data on Local mode storm cluster. Using jvisualvm, we find most CPU are used by following methods (about 60% CPU time): org.apache.storm.shade.org.apache.zookeeper.server.NIOServerCnxnFactory.run()org.apache.storm.shade.org.apache.zookeeper.ClientCnxnSocketNIO.doTransport() Is there any approaches to optimize the performance? #yiv6219386171 a#yiv6219386171ntes-pcmail-signature-default:hover {text-decoration:underline;color:#3593db;cursor:pointer;} #yiv6219386171 font{line-height:1.5;}ThanksShawn