You are viewing a plain text version of this content. The canonical link for it is here.

Posted to notifications@asterixdb.apache.org by "Taewoo Kim (JIRA)" <ji...@apache.org> on 2017/09/28 01:03:00 UTC

[jira] [Created] (ASTERIXDB-2109) B-tree search can use a lot of buffers as the number of components increases.

Taewoo Kim created ASTERIXDB-2109:
-------------------------------------

             Summary: B-tree search can use a lot of buffers as the number of components increases.
                 Key: ASTERIXDB-2109
                 URL: https://issues.apache.org/jira/browse/ASTERIXDB-2109
             Project: Apache AsterixDB
          Issue Type: Bug
            Reporter: Taewoo Kim
            Priority: Minor


LSM B-tree search uses a priority queue to hold a buffer frame from each in-memory component or component on disk. Based on a merge policy that we choose, there can be many components on disk and if they are not merged frequently, the number of required buffer frames will also increase. And, this would become an issue since if we don't have any budget constraint on LSM B-Tree search. 



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)