[
https://issues.apache.org/jira/browse/SPARK-17626?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15564169#comment-15564169
]
Ioana Delaney commented on SPARK-17626:
---------------------------------------
[~ron8hu] Thank you for your comments. Our current star schema detection uses
simple, basic heuristics to identify the star join with the largest fact table
and places it on the driving arm of the join. Even with this simple, intuitive
join reordering, the performance results show very good improvement. The next
step would be to improve the schema detection logic based on cardinality
heuristics and then with more reliable informational RI constraints. When CBO
implements the planning rules, the two algorithms can be integrated. Regarding
predicate selectivity hint, it can also be used to diagnose query
planning/performance. By changing predicate selectivity, we can influence the
choice of a plan and thus test various planning options.
> TPC-DS performance improvements using star-schema heuristics
> ------------------------------------------------------------
>
> Key: SPARK-17626
> URL: https://issues.apache.org/jira/browse/SPARK-17626
> Project: Spark
> Issue Type: Umbrella
> Components: SQL
> Affects Versions: 2.1.0
> Reporter: Ioana Delaney
> Priority: Critical
> Attachments: StarSchemaJoinReordering.pptx
>
>
> *TPC-DS performance improvements using star-schema heuristics*
> \\
> \\
> TPC-DS consists of multiple snowflake schema, which are multiple star schema
> with dimensions linking to dimensions. A star schema consists of a fact table
> referencing a number of dimension tables. Fact table holds the main data
> about a business. Dimension table, a usually smaller table, describes data
> reflecting the dimension/attribute of a business.
> \\
> \\
> As part of the benchmark performance investigation, we observed a pattern of
> sub-optimal execution plans of large fact tables joins. Manual rewrite of
> some of the queries into selective fact-dimensions joins resulted in
> significant performance improvement. This prompted us to develop a simple
> join reordering algorithm based on star schema detection. The performance
> testing using *1TB TPC-DS workload* shows an overall improvement of *19%*.
> \\
> \\
> *Summary of the results:*
> {code}
> Passed 99
> Failed 0
> Total q time (s) 14,962
> Max time 1,467
> Min time 3
> Mean time 145
> Geomean 44
> {code}
> *Compared to baseline* (Negative = improvement; Positive = Degradation):
> {code}
> End to end improved (%) -19%
> Mean time improved (%) -19%
> Geomean improved (%) -24%
> End to end improved (seconds) -3,603
> Number of queries improved (>10%) 45
> Number of queries degraded (>10%) 6
> Number of queries unchanged 48
> Top 10 queries improved (%) -20%
> {code}
> Cluster: 20-node cluster with each node having:
> * 10 2TB hard disks in a JBOD configuration, 2 Intel(R) Xeon(R) CPU E5-2680
> v2 @ 2.80GHz processors, 128 GB RAM, 10Gigabit Ethernet.
> * Total memory for the cluster: 2.5TB
> * Total storage: 400TB
> * Total CPU cores: 480
> Hadoop stack: IBM Open Platform with Apache Hadoop v4.2. Apache Spark 2.0 GA
> Database info:
> * Schema: TPCDS
> * Scale factor: 1TB total space
> * Storage format: Parquet with Snappy compression
> Our investigation and results are included in the attached document.
> There are two parts to this improvement:
> # Join reordering using star schema detection
> # New selectivity hint to specify the selectivity of the predicates over base
> tables. Selectivity hint is optional and it was not used in the above TPC-DS
> tests.
> \\
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
