[jira] [Created] (HBASE-18214) Replace the folly::AtomicHashMap usage in the RPC layer

Previous Topic Next Topic
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

[jira] [Created] (HBASE-18214) Replace the folly::AtomicHashMap usage in the RPC layer

JIRA jira@apache.org
Devaraj Das created HBASE-18214:

             Summary: Replace the folly::AtomicHashMap usage in the RPC layer
                 Key: HBASE-18214
                 URL: https://issues.apache.org/jira/browse/HBASE-18214
             Project: HBase
          Issue Type: Sub-task
            Reporter: Devaraj Das
            Assignee: Devaraj Das

In my tests, I saw that folly::AtomicHashMap usage is not appropriate for one, rather common use case. It'd become sort of unusable (inserts would hang) after a bunch of inserts and erases. This hashmap is used to keep track of call-Id after a connection is set up in the RPC layer (insert a call-id/msg pair when an RPC is sent, and erase the pair when the corresponding response is received). Here is a simple program that will demonstrate the issue:
folly::AtomicHashMap<int, int> f(100);
int i = 0;
while (i < 10000) {
    try {
      LOG(INFO) << "Inserted " << i << "  " << f.size();
      LOG(INFO) << "Deleted " << i << "  " << f.size();
    } catch (const std::exception &e) {
      LOG(INFO) << "Exception " << e.what();
After poking around a little bit, it is indeed called out as a limitation here https://github.com/facebook/folly/blob/master/folly/docs/AtomicHashMap.md (grep for 'erase'). Proposal is to replace this with something that will fit in in the above usecase (thinking of using std::unordered_map).

This message was sent by Atlassian JIRA