Videos Archive - PrestoDB

Ending DAG Distress: Building Self-Orchestrating Pipelines for Presto – Roy Hasson, Upsolver

Ending DAG Distress: Building Self-Orchestrating Pipelines for Presto – Roy Hasson, Upsolver dbt and Airflow is a popular combination for creating and scheduling batch data modeling and transformation jobs that execute in a data warehouse like Snowflake. Presto users querying the data lake need a similar solution that is simple to use and makes it easy to ingest, model, transform and maintain datasets, without having to write or manage complex DAGs. In this session you will learn how Upsolver built a tool that allows engineers, developers and analysts to write data pipelines using SQL. Pipelines are automatically orchestrated, are data-aware and maintain a consistent data contract between each stage of the pipeline. You will also learn how to introduce the idea of data products into your company to enable more self-service for your Presto users.

Querying streaming data with Presto, Amazon Athena and Upsolver

In this session, Yoni will present on querying streaming data with Presto and Amazon Athena including performance, data partitioning and compaction. In addition, we will demo using the Upsolver platform with Amazon Athena. In addition, he will share what they are working on with Prestodb.

Disaggregated Coordinator Production Rollout – Swapnil Tailor, Meta

We would like to talk about disaggregated coordinator production rollout with some stats. Also we can inform community about changes needed to support deployment of disaggregated coordinator.

Benchmarking Continuous Data Processing on Snowflake with Upsolver – Sean Spediacci, Upsolver

In this talk, we provide a glimpse at the results of our latest benchmark test which compares the speed and cost of processing data inside Snowflake (ELT) vs. processing and serving prepared live tables from a data lake using Upsolver (ETL).

Disaggregated Coordinator – Swapnil Tailor, Facebook

In the existing Presto architecture, single coordinator has become a bottleneck in a number of ways for cluster scalability. – With an increasing number of workers, the coordinator has the potential of slow down due to a high number of tasks. – In high QPS use cases, we have found workers can become starved of splits by excessive CPU being spend on task updates in coordinator. – Also with single coordinator, we have an upper limit on the worker pool because of above-mentioned reasons. To overcome with this challenges, we are coming up with a new architecture which supports multiple coordinators in a single cluster.

The Data Lake House: Powering Open Architecture in the Cloud – Ori Rafael, Upsolver

Ori Rafael, Co-founder and CEO of Upsolver, will present the Cloud Lake House as the foundation of an open data lake architecture built on Apache Parquet. Ori will explain how this architecture supports diverse analytic consumers and use cases, from open-source Presto to proprietary data warehouses.