News for hackers. Stuff that matters.
Re:Do we need a third Apache project for columnar data repre... by polskibus on Tuesday October 31, 2017 @02:40PM (#15595495)
How can I query Apache Arrow without entire Hadoop stack? It seems it could be a great in-memory OLAP engine, if only there was an efficient way to slice and dice it?
Re:Do we need a third Apache project for columnar data repre... by jacquesnadeau on Tuesday October 31, 2017 @03:50PM (#15595984)
We built Dremio (github.com/dremio/dremio-oss, apache licensed) entirely on top of Apache Arrow specifically for the purposes of creating a high speed analytical capabilities including MOLAP like work as well as other forms of caching/acceleration for analytical workloads. Other products/projects are also starting to adopt a similar technical architecture.
Re:Do we need a third Apache project for columnar data repre... by polskibus on Tuesday October 31, 2017 @03:53PM (#15596013)
Dremio looks very interesting indeed. What would you recommend for interacting with Arrow with more control, as a library? I'm interested in creating new Arrow-based data sources, not using it as an intermediary to other data sources.

On a side note - what other products/projects did you mean?

Re:Do we need a third Apache project for columnar data repre... by jacquesnadeau on Tuesday October 31, 2017 @04:15PM (#15596180)
The Arrow project itself is a set of libraries. One of the things we'll do is try to add more algorithms over time to it so if you want say, a fast arrow sort or arrow predicate application. Full SQL is always far more complex and I can't see the project itself .

The engine inside of Dremio is something we call Sabot (a shoe for modern arrows, see sabot round on wikipedia). We hope to make it modular enough one day to use a library but it isn't there yet.

In regards to your other question re projects/products: Arrow contributors are actively trying to get more adoption of Arrow as an interchange format for several systems. We've had discussions around Kudu (no serious work done yet afaik). Parquet-to-Arrow for multiple languages is now available. Arrow committers include committers from several other projects such as HBase, Cassandra, Phoenix, etc. The goal is ultimately to figure integrations with all.

In most cases, these data storage systems are saddled with slow interfaces for data access. (Think row-by-row, cell-by-cell interfaces.) Arrow, among other things, allows them to communicate through a much faster mechanism (shared memory--or at least shared representation if not node local).

Re:Do we need a third Apache project for columnar data repre... by polskibus on Tuesday October 31, 2017 @04:31PM (#15596281)
How does dremio differ from PrestoDB? As far as I know, PrestoDB can also virtualize access to many data sources and join data between them. We didn't go deep with PrestoDB because our basic tests for multi-source joins ran very slowly, and it seemed to pull all data from both joined tables into one place. I'm not a Prestodb expert, so maybe there's a better way to do it (all suggestions welcome).

What's the differentiator? Is dremio smarter somehow and avoids copying all data to perform a simple join? Or does it copy the data the same way but Arrow lets it be faster than Presto? What's on your roadmap?

Re:Do we need a third Apache project for columnar data repre... by jacquesnadeau on Tuesday October 31, 2017 @06:15PM (#15597065)
PrestoDB is similar to Impala, Hive and other SQL Engines. Each is designed to do distributed SQL processing. Dremio does embed an OSS distributed SQL processing engine (Sabot, built natively on Arrow) as well but we see that as only a means to an end. Our focus is much more on being a bi & data fabric/service.

At the core of this vision are: very advanced pushdowns (far beyond other OSS systems), a powerful self-service UI for managing, curating and sharing data (designed for analysts, not just engineers) and--most importantly--the first open source implementation of distributed relational caching for all types of data. You can see more details about this last part in a deck I presented at DataEngConf early today: https://www.slideshare.net/dremio/using-apache-arrow-calcite...

Re:Do we need a third Apache project for columnar data repre... by polskibus on Tuesday October 31, 2017 @04:27PM (#15596254)
Thank you very much for a thorough response. I think I would be happy with a library without SQL support, as long as filtering, grouping would be supported. Seems like that would be Sabot :) Maybe one day I'll be able to use it.
Re:Do we need a third Apache project for columnar data repre... by jacquesnadeau on Tuesday October 31, 2017 @05:01PM (#15596529)
For those level of operations, the Arrow library itself will probably have something fairly soon.
Re:Do we need a third Apache project for columnar data repre... by polskibus on Tuesday October 31, 2017 @03:58PM (#15596044)
How does Dremio compare to non-open source Querona? Is it a similar service to Dremios?
Re:Do we need a third Apache project for columnar data repre... by jacquesnadeau on Tuesday October 31, 2017 @04:43PM (#15596370)
Dremio is focused on a combination of data access, acceleration and a self-service analyst experience (Tableau/Qlik one-click integration, data curation and data management).

We also invest heavily in our pushdowns. For example, we invested heavily in our Elasticsearch capabilities (support for CONTAINS and Lucene syntax, Painless and Groovy pushdowns, index leveraging, etc) and I'm pretty comfortable saying we're the best in world at exposing Elastic in a SQL context. Similarly for our other sources (join & window pushdowns in Oracle, SqlServer, etc, high speed predicate application in Hadoop, Mongo aggregation pipeline pushdowns including support unwind, etc).

I don't know Querona well (just a cursory site review). They seem much more focused on classic federation. It would also be important to understand when they repackage the Spark source connectors versus where they have built something that pushes down more powerfully (which you typically need for the best performance with these newer data source systems).

Re:Do we need a third Apache project for columnar data repre... by zten on Tuesday October 31, 2017 @02:51PM (#15595595)
I would try Apache Spark. It's fairly easy to get it running locally, and does not require Hadoop. It provides the query engine.
Re:Do we need a third Apache project for columnar data repre... by polskibus on Tuesday October 31, 2017 @03:04PM (#15595694)
I was looking for something local, a library or sth like that. I don't want to run other processes just to query that memory. Spark is enormous and slow in comparison with local in memory solutions.
Re:Do we need a third Apache project for columnar data repre... by zten on Tuesday October 31, 2017 @03:53PM (#15596008)
Gotcha. Just some confusion around what you meant by in-memory OLAP engine. It seems that many storage format libraries get integrated into other platforms that bring the other supporting features (and the other undesirable heavyweight things you didn't want) with them.

With a little bit more poking: Apache Calcite?

Re:Do we need a third Apache project for columnar data repre... by sixdimensional on Tuesday October 31, 2017 @04:25PM (#15596247)
Calcite is more a query optimization framework. Actually, I think parent has a good question, which is, how to use these tools, formats, etc. from something simpler than an entire full blown distributed technology - and I think the answer to that is, there aren't too many technologies out of the box that do that right now.

I would think maybe extending SQLite to use columnar storage would be one possible path that might make sense.

There are columnar DBMS engines you can embed like MonetDB, but I personally haven't seen a lightweight query engine sitting right on top of Parquet/ORC, or implementing Arrow for in-memory columnar. You can get Apache Spark up and running pretty quickly on a single node, but I haven't seen anything even easier than that.

Re:Do we need a third Apache project for columnar data repre... by jankotek on Tuesday October 31, 2017 @03:52PM (#15596001)
mapdb has in roadmap to provide query api.
Re:Do we need a third Apache project for columnar data repre... by polskibus on Tuesday October 31, 2017 @03:57PM (#15596037)
Thanks, I more interested in something that is easier to embed in platforms other than Java.