Computing accurate percentiles with DDSketch

** We recently published a paper, DDSketch: A Fast and Fully-Mergeable Quantile Sketch with Relative-Error Guarantees, in PVLDB on August 28th. This paper introduces DDSketch, a new sketch algorithm designed to efficiently compute percentiles from large-scale monitoring data while maintaining high accuracy and low memory usage. Unlike existing algorithms, DDSketch provides a relative-error guarantee, ensuring that the computed percentile values are within a specified fraction of the actual value. The algorithm achieves this by using representative values and bucketization, allowing for efficient merging and minimizing memory footprint. Our benchmarks demonstrate that DDSketch outperforms other sketch algorithms in terms of performance, memory usage, and accuracy, making it an attractive solution for distributed data processing applications.

No tracked trend matches for this post yet.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.