Support Questions

Find answers, ask questions, and share your expertise
Announcements
Celebrating as our community reaches 100,000 members! Thank you!

Unified BI Semantic Layer

avatar
New Contributor

Hi,

I am looking to use an Open Source unified BI Semantic Layer to sit on top of Hadoop from which multiple BI Tools (e.g. Qlik, Tableau) will Report. Kylin looks like an option, but as far as I understand you cannot create Calculated Measures in Kylin e.g. Ratios that need to be calculated on the fly. How does Kylin cater for this? If this not achievable in Kylin is there any other suggestions e.g. I have read some promising reviews on Apache Lens.

Regards

Warren

1 ACCEPTED SOLUTION

avatar

@Warren Tracey

Although Kylin is an Apache project, it is not currently part of the Hortonworks Data Platform. It is important to mention that Kylin does limit your BI options somewhat, as it is oriented at Microsoft PowerPivot and Tableau. Alternatively, we at Hortonworks are engaging AtScale quite frequently with our customers in the field.

AtScale provide a unified Semantic layer, based on "Virtual Cubes". Virtual Cubes allow the user to create models with measures and dimensions, just like OLAP, but on large volumes of data stored in Hadoop. Users can ‘scale-out’ their BI, because the cube is ‘virtual’. Users can query multiple years, lines of business, brands, etc. all from 1 ‘virtual cube’, scaling out to millions and billions of rows of data available to query.

AtScale Adaptive Cache generates automatic ‘smart aggregates’ that learn to anticipate user BI and OLAP queries so you maintain scale, performance and control across your Hadoop cluster. AtScale can leverage any SQL engine under the covers (including Hive or Spark), as well as any BI tool (including those that require an MDX interface, rather than a SQL interface).

10625-atscale.jpg

If not interested, I can dig further into Apache Kylin or Apache Lens for you - but your support options may be limited going forward until the respective communities around those projects begin to grow.

View solution in original post

3 REPLIES 3

avatar

@Warren Tracey

Although Kylin is an Apache project, it is not currently part of the Hortonworks Data Platform. It is important to mention that Kylin does limit your BI options somewhat, as it is oriented at Microsoft PowerPivot and Tableau. Alternatively, we at Hortonworks are engaging AtScale quite frequently with our customers in the field.

AtScale provide a unified Semantic layer, based on "Virtual Cubes". Virtual Cubes allow the user to create models with measures and dimensions, just like OLAP, but on large volumes of data stored in Hadoop. Users can ‘scale-out’ their BI, because the cube is ‘virtual’. Users can query multiple years, lines of business, brands, etc. all from 1 ‘virtual cube’, scaling out to millions and billions of rows of data available to query.

AtScale Adaptive Cache generates automatic ‘smart aggregates’ that learn to anticipate user BI and OLAP queries so you maintain scale, performance and control across your Hadoop cluster. AtScale can leverage any SQL engine under the covers (including Hive or Spark), as well as any BI tool (including those that require an MDX interface, rather than a SQL interface).

10625-atscale.jpg

If not interested, I can dig further into Apache Kylin or Apache Lens for you - but your support options may be limited going forward until the respective communities around those projects begin to grow.

avatar

@Warren Tracey

Did this answer your question? If so, please accept the answer. If not, I'll be happy to answer any other questions you have.

Thanks!

_Tom

avatar
New Contributor

Thank you for your response Tom, we have used Mondrian 4 for the semantic layer and adapted it so that front end tools like Excel, Tableau & Qlik can read it. We are still in development phase, but we really like what we see thus far.

@Grant Krieger