This Confluence has been LDAP enabled, if you are an ASF Committer, please use your LDAP Credentials to login. Any problems file an INFRA jira ticket please.

Page tree

Versions Compared


  • This line was added.
  • This line was removed.
  • Formatting was changed.


The root cause of the performance issue mentioned above is the receive API implementation in the network-based TMB, which uses the sleep system call for blocking. We need a better implementation, e.g., event-driven based.

One alternative is to avoid the client side polling, by refactoring the Receive RPC API the server-side blocking implementation, instead of client-side streaming.

10-node-cluster Setup in CloudLab