spark-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From lovasoa <...@git.apache.org>
Subject [GitHub] spark issue #18263: [SPARK-21039] [Spark Core] Use treeAggregate instead of ...
Date Sat, 10 Jun 2017 23:31:44 GMT
Github user lovasoa commented on the issue:

    https://github.com/apache/spark/pull/18263
  
    I ran some tests on a cluster with 5 nodes, creating a bloomfilter (with an fpp of 10%)
of the column `o_orderkey` of the table `ORDERS` of TPC-H with a scale factor of 100.
    
    
    ## The cluster
    ![image](https://user-images.githubusercontent.com/552629/27006958-82c9ec76-4e4b-11e7-934a-44e3adf2b0a0.png)
    
    ## Results
    
    ### Using `RDD.aggregate`
    ![image](https://user-images.githubusercontent.com/552629/27006977-134147ea-4e4c-11e7-95a4-44bb976ff22f.png)
    
    Notice the huge scheduler delay.
    
    ** total time: 52 seconds**
    
    ### Using `RDD.treeAggregate`
    
    ![image](https://user-images.githubusercontent.com/552629/27007049-e081c968-4e4d-11e7-8b74-56d12e675b20.png)
    
    #### first stage
    ![image](https://user-images.githubusercontent.com/552629/27007052-fc562cba-4e4d-11e7-9080-5b46f62c161e.png)
    
    #### second stage
    
    ![image](https://user-images.githubusercontent.com/552629/27007058-15887300-4e4e-11e7-8646-daecdde0056f.png)
    
    ** total time: 17 seconds**


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org


Mime
View raw message