[
https://issues.apache.org/jira/browse/HIVE-4160?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Eric Hanson updated HIVE-4160:
------------------------------
Description:
The Hive query execution engine currently processes one row at a time. A single
row of data goes through all the operators before the next row can be
processed. This mode of processing is very inefficient in terms of CPU usage.
Research has demonstrated that this yields very low instructions per cycle
[MonetDB X100]. Also currently Hive heavily relies on lazy deserialization and
data columns go through a layer of object inspectors that identify column type,
deserialize data and determine appropriate expression routines in the inner
loop. These layers of virtual method calls further slow down the processing.
This work will add support for vectorized query execution to Hive, where,
instead of individual rows, batches of about a thousand rows at a time are
processed. Each column in the batch is represented as a vector of a primitive
data type. The inner loop of execution scans these vectors very fast, avoiding
method calls, deserialization, unnecessary if-then-else, etc. This
substantially reduces CPU time used, and gives excellent instructions per cycle
(i.e. improved processor pipeline utilization). See the attached design
specification for more details.
was:
Hive query execution engine currently processes one row at a time. A single
row of data goes through all the operators before next row can be processed.
This mode of processing is very inefficient in terms of CPU usage. Research has
demonstrated that this yields very low instructions per cycle [MonetDBX100].
Also currently hive heavily relies on lazy deserialization and data columns go
through a layer of object inspectors that identify column type, de-serialize
data and determine appropriate expression routines in the inner loop. These
layers of virtual method calls further slow down the processing.
This work will add support for vectorized query execution to Hive, where,
instead of individual rows, batches of about a thousand rows at a time are
processed. Each column in the batch is represented as a vector of a primitive
data type. The inner loop of execution scans these vectors very fast, avoiding
method calls, deserialization, unnecessary if-then-else, etc. This
substantially reduces CPU time used, and gives excellent cycles per instruction
(i.e. improved processor pipeline utilization). See the attached design
specification for more details.
> Vectorized Query Execution in Hive
> ----------------------------------
>
> Key: HIVE-4160
> URL: https://issues.apache.org/jira/browse/HIVE-4160
> Project: Hive
> Issue Type: New Feature
> Reporter: Jitendra Nath Pandey
> Assignee: Jitendra Nath Pandey
> Attachments: Hive-Vectorized-Query-Execution-Design.docx,
> Hive-Vectorized-Query-Execution-Design-rev2.docx,
> Hive-Vectorized-Query-Execution-Design-rev3.docx,
> Hive-Vectorized-Query-Execution-Design-rev3.docx,
> Hive-Vectorized-Query-Execution-Design-rev3.pdf
>
>
> The Hive query execution engine currently processes one row at a time. A
> single row of data goes through all the operators before the next row can be
> processed. This mode of processing is very inefficient in terms of CPU usage.
> Research has demonstrated that this yields very low instructions per cycle
> [MonetDB X100]. Also currently Hive heavily relies on lazy deserialization
> and data columns go through a layer of object inspectors that identify column
> type, deserialize data and determine appropriate expression routines in the
> inner loop. These layers of virtual method calls further slow down the
> processing.
> This work will add support for vectorized query execution to Hive, where,
> instead of individual rows, batches of about a thousand rows at a time are
> processed. Each column in the batch is represented as a vector of a primitive
> data type. The inner loop of execution scans these vectors very fast,
> avoiding method calls, deserialization, unnecessary if-then-else, etc. This
> substantially reduces CPU time used, and gives excellent instructions per
> cycle (i.e. improved processor pipeline utilization). See the attached design
> specification for more details.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
