Release 0.3.0

Released September 3rd, 2016

Maven

<dependency>
    <groupId>com.sparklinedata</groupId>
    <artifactId>spl-accelerator_2.10</artifactId>
    <version>0.3.0</version>
</dependency>

Downloads

• Sparkline Accelerator jar
• Source Code

ChangeLog

• Changes

Feature/Bug Fixes

• Support pushdown of non-aggregation query patterns; controlled by the nonAggQueryHandling option
• Support for TopN Druid Query optimization, controlled by the allowTopN option
• DruidQueries UI tab now shows the associated SQL query
• sparkline properties are now settable as spark properties(in properties file or env, just like any spark sql property)
• Queries issued to Druid can be tracked by a generated queryId
• fix bugs in Metadata caching: react to Druid Segment changes
• relay Spark Task cancellations to Druid Queries in progress, leading much proper cleanup of cancelled queries.
• support for Druid 0.9.1, Spark 1.6.1
• Cost Model fixes and support for SelectQuery. Introduce queryGranularity option for datasources