Need advice about which tool to choose?Ask the StackShare community!
Apache Drill vs Presto: What are the differences?
Introduction:
Apache Drill and Presto are both open-source distributed SQL query engines that support querying and analyzing large datasets in real-time. While they have some similarities, there are key differences between them that make them suitable for different use cases.
-
Architecture and Data Sources:
- Apache Drill is based on a schema-free JSON-like model, allowing it to explore and query various types of data sources, including nested data and self-describing data formats like Parquet, Avro, and JSON.
- Presto, on the other hand, follows a traditional query engine model where data is organized into tables and columns. It primarily focuses on querying structured data stored in sources like Hive, MySQL, PostgreSQL, and others.
-
Query Optimization and Execution:
- Apache Drill utilizes a "schema-free" execution model, where it dynamically determines the query plan based on the data at hand during runtime. It leverages distributed query execution for faster processing.
- Presto takes a different approach by leveraging a cost-based query optimizer that uses statistics to estimate costs and plan the most efficient query execution path. This optimizer allows Presto to optimize queries more effectively based on the characteristics of the underlying data.
-
Ecosystem Integration:
- Apache Drill offers native integration with Apache Hadoop, YARN, and HBase, making it well-suited for big data environments that require integration with these technologies. It also provides easy integration with BI tools like Tableau and Qlik.
- Presto, on the other hand, has extensive ecosystem integration and supports various data sources like MySQL, PostgreSQL, Cassandra, and more. It provides a pluggable connector architecture that allows users to add support for new data sources.
-
SQL Support and ANSI Compliance:
- Apache Drill supports a broad range of SQL functionality but may have limitations in complex joins, subqueries, and optimization for specific databases. It aims to provide a consistent SQL interface across different data sources.
- Presto, on the other hand, places a higher emphasis on SQL compliance and aims to support most SQL standards. It offers advanced SQL features like window functions, query optimization, and joins optimization.
-
Community and Development:
- Apache Drill has a relatively smaller community compared to Presto. It is primarily developed and maintained by the Apache Software Foundation (ASF) and benefits from the resources and support provided by ASF.
- Presto, on the other hand, has a larger and more active community of contributors, including Facebook, where it originated. The larger community ensures the continuous development and improvement of Presto.
-
Maturity and Production Readiness:
- Apache Drill has been in development since 2012 and has reached a stable version. However, some users may still consider it less mature compared to other query engines due to its relatively shorter history.
- Presto is more established and widely used in production environments by companies like Facebook, Airbnb, and Netflix. It has been battle-tested and proven to be scalable and robust in handling large datasets.
In summary, Apache Drill and Presto are both powerful SQL query engines with their own strengths and use cases. Apache Drill is a versatile choice for exploring diverse data sources, while Presto offers advanced SQL features, extensive ecosystem integration, and a mature production-ready solution.
To provide employees with the critical need of interactive querying, we’ve worked with Presto, an open-source distributed SQL query engine, over the years. Operating Presto at Pinterest’s scale has involved resolving quite a few challenges like, supporting deeply nested and huge thrift schemas, slow/ bad worker detection and remediation, auto-scaling cluster, graceful cluster shutdown and impersonation support for ldap authenticator.
Our infrastructure is built on top of Amazon EC2 and we leverage Amazon S3 for storing our data. This separates compute and storage layers, and allows multiple compute clusters to share the S3 data.
We have hundreds of petabytes of data and tens of thousands of Apache Hive tables. Our Presto clusters are comprised of a fleet of 450 r4.8xl EC2 instances. Presto clusters together have over 100 TBs of memory and 14K vcpu cores. Within Pinterest, we have close to more than 1,000 monthly active users (out of total 1,600+ Pinterest employees) using Presto, who run about 400K queries on these clusters per month.
Each query submitted to Presto cluster is logged to a Kafka topic via Singer. Singer is a logging agent built at Pinterest and we talked about it in a previous post. Each query is logged when it is submitted and when it finishes. When a Presto cluster crashes, we will have query submitted events without corresponding query finished events. These events enable us to capture the effect of cluster crashes over time.
Each Presto cluster at Pinterest has workers on a mix of dedicated AWS EC2 instances and Kubernetes pods. Kubernetes platform provides us with the capability to add and remove workers from a Presto cluster very quickly. The best-case latency on bringing up a new worker on Kubernetes is less than a minute. However, when the Kubernetes cluster itself is out of resources and needs to scale up, it can take up to ten minutes. Some other advantages of deploying on Kubernetes platform is that our Presto deployment becomes agnostic of cloud vendor, instance types, OS, etc.
#BigData #AWS #DataScience #DataEngineering
The platform deals with time series data from sensors aggregated against things( event data that originates at periodic intervals). We use Cassandra as our distributed database to store time series data. Aggregated data insights from Cassandra is delivered as web API for consumption from other applications. Presto as a distributed sql querying engine, can provide a faster execution time provided the queries are tuned for proper distribution across the cluster. Another objective that we had was to combine Cassandra table data with other business data from RDBMS or other big data systems where presto through its connector architecture would have opened up a whole lot of options for us.
Pros of Apache Drill
- NoSQL and Hadoop4
- Free3
- Lightning speed and simplicity in face of data jungle3
- Well documented for fast install2
- SQL interface to multiple datasources1
- Nested Data support1
- Read Structured and unstructured data1
- V1.10 released - https://drill.apache.org/1
Pros of Presto
- Works directly on files in s3 (no ETL)18
- Open-source13
- Join multiple databases12
- Scalable10
- Gets ready in minutes7
- MPP6