This document introduces Apache Drill, an open source interactive analysis engine for big data. It was inspired by Google's Dremel and supports standard SQL queries over various data sources like Hadoop and NoSQL databases. Drill provides low-latency interactive queries at scale through its distributed, schema-optional architecture and support for nested data formats. The talk outlines Drill's capabilities and status as a community-driven project under active development.
5. Use Case
• Jane, a marketing analyst
• Determine target segments
• Data from different sources
6. Today’s Solutions
• RDBMS-focused
– ETL data from MongoDB/Hadoop
– Query with SQL
• MapReduce-focused
– ETL from RDBMS/MongoDB
– Use Hive
7. Requirements
• Support for different data sources
• Support for different query interfaces
• Low-latency/real-time
• Ad-hoc queries
• Scalable and fast
• Reliable
9. Apache Drill Overview
• Inspired by Google Dremel
• Standard SQL2003 support
• …. other QL (DSL, etc.) possible
• Plug-able data sources
• Support for nested data (JSON, etc.)
• Schema is optional
• Community driven, open, 100’s involved
14. Key Features
• Full SQL
• Nested data
• Optional schema
• Extensibility points
15. Full SQL – ANSI SQL2003
• SQL-like is often not enough
• Integration with existing tools
– Tableau, Excel, SAP Crystal Reports
– Use standard ODBC/JDBC driver
16. Nested Data
• Nested data becoming prevalent
– JSON/BSON, XML, ProtoBuf, Avro
– Some data sources support it natively
(MongoDB, etc.)
– Innovation through Dremel
• Flattening nested data is error-prone
• Apache Drill supports nested data,
extension to ANSI SQL2003
17. Optional Schema
• Many data sources don’t have rigid schemas
– Schema changes rapidly
– Different schema per record (e.g. HBase)
• Apache Drill supports queries against
unknown schema
• user can define schema or via discovery
18. Extensibility Points
• Query language (parser) - UDFs
• Data sources/formats (scanner)
• Optimizer
• Custom operators (logical plan)
Source Logical Physical
Query Parser Plan Optimizer Plan Execution
22. Engage!
• Follow @ApacheDrill on Twitter
• Sign up at mailing lists (user|dev)
http://incubator.apache.org/drill/mailing-lists.html
• Keep an eye on http://drill-user.org/
• Ping me: mhausenblas@maprtech.com
Notes de l'éditeur
Hive: compile to MR, Aster: external tables in MPP, Oracle/MySQL: export MR results to RDBMSDrill, Impala, CitusDB: real-time
Suppose a marketing analyst trying to experiment with ways to do targeting of user segments for next campaign. Needs access to web logs stored in Hadoop, and also needs to access user profiles stored in MongoDB as well as access to transaction data stored in a conventional database.
Re ad-hoc:You might not know ahead of time what queries you will want to make. You may need to react to changing circumstances.
Two innovations: handle nested-data column style (column-striped representation) and query push-down
Source query is parsed and transformed to produce the logical planTypically, the logical plan lives in memory in the form of Java objects, but it also has a textual form. The logical query is then transformed and optimized into the physical plan.The physical plan represents the actual structure of computation as it is done by the system. One of the most important things the optimizer does is the introduction of parallel computation (other: columnar data to improve processing speed)