Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Apache Calcite overview

17,573 views

Published on

Brief overview of Apache Calcite.

Presented by Julian Hyde at the Apache Kylin meetup hosted by eBay in San Jose, CA, on 2014/12/04.

Published in: Technology
  • Be the first to comment

Apache Calcite overview

  1. 1. Apache Calcite Overview Julian Hyde Julian Hyde Page 1 © Hortonworks Inc. 2014 Kylin Meetup (eBay, San Jose) December 4th, 2014
  2. 2. Apache Calcite Apache incubator project since May, 2014 Originally named Optiq Query planning framework Relational algebra, rewrite rules, cost model Extensible Packaging Library (JDBC server optional) Community-authored rules, adapters Adoption Embedded: Lingual (SQL interface to Cascading), Apache Drill, Apache Hive, Apache Kylin Adapters: Splunk, Spark, MongoDB, JDBC, CSV, JSON, Web tables, In-memory, Phoenix Page 2 © Hortonworks Inc. 2014
  3. 3. Conventional DB architecture Page 3 © Hortonworks Inc. 2014
  4. 4. Calcite architecture Page 4 © Hortonworks Inc. 2014
  5. 5. Expression tree Splunk Table: splunk MySQL Page 5 © Hortonworks Inc. 2014 SELECT p.“product_name”, COUNT(*) AS c FROM “splunk”.”splunk” AS s JOIN “mysql”.”products” AS p ON s.”product_id” = p.”product_id” WHERE s.“action” = 'purchase' GROUP BY p.”product_name” ORDER BY c DESC Key: product_id join Key: product_name Agg: count group Condition: action = 'purchase' filter Key: c DESC sort scan scan Table: products
  6. 6. Expression tree (optimized) Splunk Table: splunk Page 6 © Hortonworks Inc. 2014 SELECT p.“product_name”, COUNT(*) AS c FROM “splunk”.”splunk” AS s JOIN “mysql”.”products” AS p ON s.”product_id” = p.”product_id” WHERE s.“action” = 'purchase' GROUP BY p.”product_name” ORDER BY c DESC Key: product_id join Key: product_name Agg: count group Condition: action = 'purchase' filter Key: c DESC sort scan MySQL scan Table: products
  7. 7. Defining a rule Page 7 © Hortonworks Inc. 2014 class FilterIntoJoinRule extends RelOptRule { public FilterIntoJoinRule() { super( operand(Filter.class, operand(Join.class, any()))); } public void onMatch(RelOptRuleCall call) { Filter filter = call.rel(0); Join join = call.rel(1); Filter newFilter = ...; Join newJoin = ...; call.transformTo(newJoin); } } Filter Join Filter’ Join’ R1 R2 R1 R2
  8. 8. Calcite – APIs and SPIs Relational algebra RelNode (operator) • Scan • Filter • Project • Union • Aggregate • … RelDataType (type) RexNode (expression) RelTrait (physical property) • RelConvention (calling-convention) • RelCollation (sortedness) • TBD (bucketedness/distribution) JDBC driver Page 8 © Hortonworks Inc. 2014 Cost, statistics RelOptCost RelOptCostFactory RelMetadataProvider • RelMdColumnUniquensss • RelMdDistinctRowCount • RelMdSelectivity SQL parser SqlNode SqlParser SqlValidator Transformation rules RelOptRule • MergeFilterRule • PushAggregateThroughUni onRule • RemoveCorrelationForScal arProjectRule • 100+ more Unification (materialized view) Column trimming Metadata Schema Table Function • TableFunction • TableMacro
  9. 9. Thank you! @julianhyde http://calcite.incubator.apache.org/ Page 9 © Hortonworks Inc. 2014

×