Announcing Imply 2.1.0

by Vadim Ogievetsky · April 19, 2017

We are extremely excited to announce Imply 2.1.0, a feature-packed release, available now for download. Native SQL comes to Druid and dashboarding is now possible in Pivot. You can find out more below:

Druid 0.10.0

Imply 2.1.0 includes Druid 0.10.0, the latest release from the Druid community. Full release notes for Druid 0.10.0 are at: https://github.com/druid-io/druid/releases/tag/druid-0.10.0. More information about Druid can be found at: http://druid.io/.

Below, we’re excited to share some highlights from this Druid release!

Built-in SQL

Druid now includes a built-in SQL server powered by Apache Calcite. Druid provides two SQL APIs: HTTP POST and JDBC. This provides an alternative to Druid’s native JSON API which is more familiar to new developers, and which makes it easier to integrate pre-existing applications that natively speak SQL. Not all Druid and SQL features are supported by the SQL layer in this initial release, but we intend to expand both in future releases.

Numeric dimensions

Druid now supports numeric dimensions at ingestion and query time. Users can ingest long and float columns as dimensions (i.e., treating the numeric columns as part of the ingestion-time grouping key instead of as aggregators, if rollup is enabled). Additionally, Druid queries can now accept any long or float column as a dimension for grouping or for filtering.

There are performance tradeoffs between string and numeric columns. Numeric columns are generally faster to group on than string columns. Numeric columns don’t have indexes, so they are generally slower to filter on than string columns.

Other highlights

Druid 0.10.0 contains hundreds of other performance improvements, stability improvements, and bug fixes from over 40 contributors. In addition to the built-in SQL layer and numeric dimensions, this release also includes:

  • A revamp of the “index” task
  • A new “like” filter
  • Large column support
  • Ability to run the coordinator and overlord as a single service
  • Better performing defaults
  • Kerberos authentication support
  • Eight new extensions

See the Druid release notes for full details on these new features!

Imply Pivot 0.11

In addition to the Druid update, Imply 2.1.0 also includes an all-new version of Pivot which supports interactive dashboards along with lots of other new features.

Interactive dashboards

We have extended the collection functionality from our last release to allow for more customizable dashboard layouts. You can turn on “shared filters”, which means that interacting with one visualization applies a filter across the dashboard.

Dashboard with filter bar and highlighted segment

Easy data cube editing

Data cube editing is now simpler than ever. Just click the pencil icon Pencil to go to the edit mode, where you can change dimensions, measures and general preferences.

Data cube edit view in General tab

Simple measure editor

You can now create simple measures without writing any code.

Measure edit dialog

You can still create complex measures by going to the Advanced tab. For example, to create a measure expressing the percentage of revenue form the United State enter the following Plywood formula: $main.filter($country == 'United States').sum($revenue) / $main.sum($revenue) from the advanced tab.

Grouped dimensions and measures

You can now put dimensions and measures into groups. The groups allow you to save space and express closeness of relationship.

Data cube view with grouped dimensions and measures

You simply create a dimension and measure groups from the data cube edit mode, and drag items to add them to groups.

Data edit view dimension list with groups

Other improvements

In addition to all of the above we also made Pivot leverage all of the new query features of Druid 0.10.0. In particular Pivot will minimize the use of JavaScript aggregators and post-aggregators to optimize performance and increase security.

Getting started

You can get the latest version of Imply on our download page. To learn more, please see our documentation. Any feedback, bug reports and feature requests are always welcome - you can post them in our user group or contact us.

Back to blog

How can we help?