Can it help with S3-specific Parquet performance issues?

Yes, it provides specific guidance on row group sizing and retry logic tailored for cloud object stores like Amazon S3 to minimize metadata overhead and LIST operations.

Does it recommend specific compression algorithms?

Yes, it provides a selection guide favoring modern codecs like ZSTD(3) for general production use, Snappy for hot data, and higher ZSTD levels for archival data.

How does it improve Parquet read performance?

By suggesting column projections and row group filtering based on statistics, it helps reduce the amount of data transferred and processed by 10x or more for wide tables.

Can this skill help prevent Out-of-Memory (OOM) errors?

Absolutely. It suggests batch size tuning and streaming patterns instead of full-collection methods to maintain controlled memory usage during large-scale data processing.

Which Rust libraries does this skill support?

It primarily targets the official 'parquet' and 'arrow' crates, including specific support for AsyncArrowWriter and ParquetRecordBatchStreamBuilder.

Rust Parquet Optimization

Name: Rust Parquet Optimization
Author: EmilLindfors

byEmilLindfors

0•

データベース管理

Optimizes Parquet file operations in Rust to improve query performance, reduce storage costs, and prevent memory issues.

This skill provides domain-specific expertise for Rust developers working with Apache Parquet files, specifically targeting the arrow-rs and parquet-arrow ecosystems. It proactively identifies inefficient reading and writing patterns—such as missing compression, suboptimal row group sizing, and lack of column projection—and suggests high-performance alternatives. Whether you are building a data lake on S3 or a local analytics engine, this skill ensures your Parquet implementation utilizes best practices like ZSTD compression, dictionary encoding, and memory-efficient streaming to maximize throughput and minimize resource consumption.

主な機能

01Row group sizing recommendations tailored for cloud storage (S3) scanning

02Automated analysis of WriterProperties for compression and encoding settings

03Memory-efficient streaming patterns to prevent OOM errors in large datasets

04Column projection and predicate pushdown optimization for faster data retrieval

050 GitHub stars

06Column-specific encoding suggestions based on data cardinality

ユースケース

01Troubleshooting slow analytical queries in Rust-based data engines like DataFusion

02Building production-grade data lakes on AWS S3 with optimized ZSTD storage

03Implementing memory-safe Parquet readers for high-throughput data pipelines

主な機能

01Row group sizing recommendations tailored for cloud storage (S3) scanning

02Automated analysis of WriterProperties for compression and encoding settings

03Memory-efficient streaming patterns to prevent OOM errors in large datasets

04Column projection and predicate pushdown optimization for faster data retrieval

050 GitHub stars

06Column-specific encoding suggestions based on data cardinality

ユースケース

01Troubleshooting slow analytical queries in Rust-based data engines like DataFusion

02Building production-grade data lakes on AWS S3 with optimized ZSTD storage

03Implementing memory-safe Parquet readers for high-throughput data pipelines