[jira] [Commented] (VXQUERY-92) Benchmark and Framework for Parallel XQuery

["Preston Carman (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/VXQUERY-92?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14295967#comment-14295967 ] 

Preston Carman commented on VXQUERY-92:
---------------------------------------

The benchmark over weather data is now apart of the code base. The benchmark includes selection, aggregation and join queries. What tasks do you think are still open related to this issue.

Do we need documentation or something added to our website about the benchmark?

> Benchmark and Framework for Parallel XQuery
> -------------------------------------------
>
>                 Key: VXQUERY-92
>                 URL: https://issues.apache.org/jira/browse/VXQUERY-92
>             Project: VXQuery
>          Issue Type: New Feature
>            Reporter: Preston Carman
>            Assignee: Preston Carman
>              Labels: gsoc2013
>
> Create a set of queries and a framework to test the parallel execution of a XQuery query processing engine. The set of queries shall outline the strengths and weaknesses of using VXQuery implementation. The framework should return measurement details for determining performance.
> Utilizing these queries a person may create both speed up and scale out tests scenarios.
>  
> The benchmark queries should have the following properties.
>  * Include at least one query for each scenario being testing.
>  * The queries are based on a real data set.
>  * The data set is publicly available.
>  * The benchmark details will be available as an open source project.
>  * Scale out and speed up will have a default test case defined.
>  
> Initial suggested scenarios.
>  * Filtering results.
>  * Computing an aggregate result.
>  * Sorting the result.
> Side Note: While this benchmark and framework is new, VXQuery has a test suite that runs thousands of queries against the system. This test suite could form the basis of the new benchmark and framework.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Re: [jira] [Commented] (VXQUERY-92) Benchmark and Framework for Parallel XQuery

[Michael Carey <mj...@ics.uci.edu>]

Ideally you'd want to add instructions to others on how to:
  - Get/install VXQuery
  - Get/install the benchmark files and queries
  - Run the benchmark queries
I.e., info on how to reproduce the results.
Then others could get/run the "same" thing and compare against it, etc.

On 1/28/15 2:31 PM, Preston Carman (JIRA) wrote:
>      [ https://issues.apache.org/jira/browse/VXQUERY-92?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14295967#comment-14295967 ]
>
> Preston Carman commented on VXQUERY-92:
> ---------------------------------------
>
> The benchmark over weather data is now apart of the code base. The benchmark includes selection, aggregation and join queries. What tasks do you think are still open related to this issue.
>
> Do we need documentation or something added to our website about the benchmark?
>
>> Benchmark and Framework for Parallel XQuery
>> -------------------------------------------
>>
>>                  Key: VXQUERY-92
>>                  URL: https://issues.apache.org/jira/browse/VXQUERY-92
>>              Project: VXQuery
>>           Issue Type: New Feature
>>             Reporter: Preston Carman
>>             Assignee: Preston Carman
>>               Labels: gsoc2013
>>
>> Create a set of queries and a framework to test the parallel execution of a XQuery query processing engine. The set of queries shall outline the strengths and weaknesses of using VXQuery implementation. The framework should return measurement details for determining performance.
>> Utilizing these queries a person may create both speed up and scale out tests scenarios.
>>   
>> The benchmark queries should have the following properties.
>>   * Include at least one query for each scenario being testing.
>>   * The queries are based on a real data set.
>>   * The data set is publicly available.
>>   * The benchmark details will be available as an open source project.
>>   * Scale out and speed up will have a default test case defined.
>>   
>> Initial suggested scenarios.
>>   * Filtering results.
>>   * Computing an aggregate result.
>>   * Sorting the result.
>> Side Note: While this benchmark and framework is new, VXQuery has a test suite that runs thousands of queries against the system. This test suite could form the basis of the new benchmark and framework.
>
>
> --
> This message was sent by Atlassian JIRA
> (v6.3.4#6332)