[
https://issues.apache.org/jira/browse/DRILL-8491?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Piyush Shama updated DRILL-8491:
--------------------------------
Priority: Major (was: Critical)
> MongoDB | Queries Conversion optimisation & using various mongoDB features
> --------------------------------------------------------------------------
>
> Key: DRILL-8491
> URL: https://issues.apache.org/jira/browse/DRILL-8491
> Project: Apache Drill
> Issue Type: Improvement
> Reporter: Piyush Shama
> Priority: Major
>
> {*}Title{*}: Inefficient Query Translation and Underutilised Functions in SQL
> to MongoDB Conversion Using Apache Drill
> {*}Description{*}: We have been experiencing significant performance issues
> when using Apache Drill to convert SQL queries for use with MongoDB. It
> appears that the SQL to MongoDB query translation process is not optimally
> executed, leading to inefficient query operations and slow response times.
> {*}Details{*}:
> # {*}Inefficient Query Translation{*}:
> ** The translation of SQL queries into MongoDB-specific queries by Apache
> Drill seems sub optimal. This inefficiency is particularly noticeable with
> complex queries, where the expected execution plan does not align with
> MongoDB's capabilities, resulting in slower query performance.
> # {*}Underutilization of MongoDB Capabilities{*}:
> ** Several MongoDB functionalities are not being fully utilised in the
> translation process:
> *** {*}Aggregation Operations{*}: Functions like {{{}SUM(){}}},
> {{{}AVG(){}}}, {{{}MIN(){}}}, and {{MAX()}} are either poorly translated or
> not utilised, leading to potential performance degradation.
> *** {*}Date Handling{*}: Extraction of date components (e.g., day from an
> ISO date) within queries is not handled efficiently, forcing additional
> processing overhead or client-side computations.
> *** {*}Count Queries{*}: Simple count operations are not optimised, possibly
> translating into more complex query forms than necessary.
> {*}Impact{*}: The current issues significantly affect the performance and
> scalability of applications relying on Apache Drill for interacting with
> MongoDB, particularly in data-heavy environments.
> {*}Expected Behaviour{*}:
> * Queries translated from SQL to MongoDB should utilise MongoDB's native
> query capabilities more effectively, ensuring that operations such as
> aggregations, date extractions, and counts are executed in the most efficient
> manner possible.
> * The translation engine should optimise the query structure to leverage
> MongoDB's strengths, particularly in handling large datasets.
> {*}Steps to Reproduce{*}:
> # Set up Apache Drill with a MongoDB data source.
> # Execute complex SQL queries involving aggregation, date extraction, and
> count operations.
> # Observe the generated MongoDB queries and resulting performance.
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
