Need advice about which tool to choose?Ask the StackShare community!
Apache Impala vs Presto: What are the differences?
Introduction:
Apache Impala and Presto are both open-source distributed SQL query engines designed for querying and analyzing large datasets in real-time. While they share similar goals and functionalities, there are significant differences between the two.
Architecture: Apache Impala follows a massively parallel processing (MPP) architecture, where queries are executed using a distributed processing model. On the other hand, Presto follows a distributed SQL query engine architecture and uses a query coordinator and workers to process queries.
Query Execution Engine: Impala utilizes a specialized query execution engine that is optimized for interactive analytics, enabling faster query execution. Presto, on the other hand, uses a code generation-based query execution engine that is designed to handle complex SQL queries and supports a wide range of data sources.
Metadata Caching: Impala features an integrated metadata catalog called the Apache Hive Metastore, which enables caching of the metadata for better query performance. Presto, on the other hand, does not have an integrated metadata catalog and relies on connecting directly to the underlying data sources for metadata.
Supported Data Sources: Impala is primarily designed to work with Apache Hadoop-based data sources such as HDFS, Apache HBase, and Apache Kudu. Presto, on the other hand, supports a broader range of data sources including various file systems, SQL databases, NoSQL databases, and cloud storage systems.
Concurrency and Resource Management: Impala has built-in support for concurrency control and resource management, which allows for efficient allocation of resources among multiple users and queries. Presto also supports concurrency and resource management but requires the use of an external tool like the YARN or Kubernetes cluster manager for resource allocation.
Maturity and Adoption: Impala has been around for a longer time and is widely adopted in the Apache Hadoop ecosystem. Presto, although relatively newer, has been gaining popularity and is employed by several large tech companies.
In summary, Apache Impala and Presto differ in their architecture, query execution engines, metadata caching, supported data sources, concurrency and resource management, and maturity and adoption in the industry.
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 Impala
- Super fast11
- Massively Parallel Processing1
- Load Balancing1
- Replication1
- Scalability1
- Distributed1
- High Performance1
- Open Sourse1
Pros of Presto
- Works directly on files in s3 (no ETL)18
- Open-source13
- Join multiple databases12
- Scalable10
- Gets ready in minutes7
- MPP6