Videos Archive - PrestoDB

Velociraptor – The Next Generation of RaptorX – Vladimir Rodionov, Carrot Cache

Vladimir Rodionov, founder of Carrot Cache will present the Velociraptor – the next evolution of PrestoDB hierarchical caching framework RaptorX. Velociraptor enables efficient data and meta-data caching well beyond RaptorX limits in terms of number of data files (multi-billions), number of table partitions (multi-millions) and number of table columns (multi-thousands). Velociraptor replaces all five RaptorX caches (Hive meta-data, file list, query result fragments, ORC/Parquet meta-data and data I/O) with a scalable solution, based on Carrot Cache, which does not pollute JVM heap memory, does not affect Java Garbage Collector, keeps all data and meta-data off Java heap memory or on disk and can scale well beyond server’s physical RAM limit. Velociraptor supports server restart, by quickly saving and loading data to/from disk for automatic cache warm up.

Parquet Column Level Access Control with Presto

Apache Parquet is the major columnar file storage format used by Apache Presto and several other query engines in many big data analytic frameworks today. In a lot of use cases, a portion of the column data is highly sensitive and must be protected. Column encryption at the file format level is supported in the Parquet community. Due to the rewritten code of Parquet in Presto, Parquet column encryption at Presto needs to be ported with modifications to the Presto code page. And the integration with Key Management Service (KMS) and other query engines like Hive and Spark is another challenge. In this talk, we will show the work we have done for enabling Presto for Parquet column decryption including challenges, solutions, integration with Hive/Spark Parquet column encryption and look forward to the next step of encryption work.

Speed Up Presto Reading with Paquet Column Indexes – Xinli Shang, & Chen Liang, Uber

Data analytic tables in the big data ecosystem are usually large and some of them can reach petabytes in size. Presto as a fast query engine needs to be intelligent to skip reading unnecessary data based on filters. In addition to the existing filtering to skip partitions, files, and row groups, Apache Parquet Column Index provides further filtering to pages, which is the I/O unit for the Parquet data source. In this presentation, we will show the work that we integrated Parquet Column Index to Presto code base, the performance gains, etc. We will also talk about our effort to open-source this project to PrestoDB and look forward to collaborating with the community to merge!

After RaptorX: Improve Performance Understanding and Workload Analysis in Presto – Ke Wang & Bin Fan

RaptorX, an umbrella project presented in PrestoCon Day in March, enabled the Presto interactive fleet in Facebook to reduce latency by 10x, based on a set of architectural improvements and optimizations with hierarchical caching. This presentation provides an update on the follow-up enhancement. Bin Fan from Alluxio will talk about the exploration of a probabilistic algorithm in Alluxio caching to estimate cache working set and the implementation of shadow cache Ke Wang from Facebook will talk about how shadow cache is used to understand the system bottleneck for better resource allocation and query routing decisions. She will also cover a recent improvement in collecting and aggregating per-query runtime statistics on the Presto engine to better understand the time breakdown, resource usage breakdown and cache hit rate on a per-query basis, which can help identify areas of improvement.

RaptorX: Building a 10X Faster Presto – James Sun, Facebook, Inc

RaptorX is an internal project name aiming to boost query latency significantly beyond what vanilla Presto is capable of. For this session, we introduce the hierarchical cache work including Alluxio data cache, fragment result cache, etc. Cache is the key building block for RaptorX. With the support of the cache, we are able to boost query performance by 10X. This new architecture can beat performance oriented connectors like Raptor with the added benefit of continuing to work with disaggregated storage.