Xiang Fu
CoFounder & Chief Architect @ StarTree
San Francisco, California, United States
Actions
Xiang Fu is a seasoned software engineer and entrepreneur, currently serving as the co-founder of StarTree, a leading SaaS data analytics company. Prior to launching StarTree, Xiang made significant contributions to the tech industry as the architect of Uber's streaming data platform, where he tackled complex challenges in data serving, processing, and analytics at an unprecedented scale, supporting diverse applications such as Uber Rides, Uber Eats, and Uber Freight.
A distinguished member of the Apache Software Foundation, Xiang is the founder and a Project Management Committee (PMC) member of Apache Pinot, an open-source, real-time, distributed OLAP datastore. His work on Apache Pinot, which began during his tenure at LinkedIn, has left a lasting impact on the real-time data analytics landscape.
Beyond his professional endeavors, Xiang is an avid aficionado of fine red wines and whiskies, enjoying the nuanced tastes and rich histories behind each bottle. His journey from engineering innovative software solutions to shaping the future of data analytics, coupled with his personal passions, exemplifies a profound commitment to excellence and a deep appreciation for the finer things in life.
Area of Expertise
Topics
High-Performance Full-Text Search on Apache Iceberg Using Lucene-Integrated Indexing
Modern data lakehouses built on Apache Iceberg deliver massive scalability and cost efficiency, but they are fundamentally optimized for scan-heavy analytics—not for sub-second, full-text search over terabytes of cold data. This creates a growing gap for teams who want search-engine-like experiences directly on their data lake without duplicating data into a separate system.
In this session, we explore how Apache Pinot bridges that gap by integrating Apache Lucene indexing directly into its distributed serving layer, while keeping Iceberg (Parquet on object storage) as the system of record. By virtualizing Pinot segments over Iceberg files and selectively pinning Lucene indexes to local NVMe storage, Pinot enables high-performance inverted-index search without large-scale data movement.
Attendees will gain a deep understanding of the architecture, indexing strategy, and I/O optimizations that make interactive full-text search possible on multi-terabyte Iceberg tables.
High-Performance Full Text Search Directly on Iceberg via Lucene-Integrated Indexing
While Data Lakehouses like Apache Iceberg provide massive, cost-effective scalability, they are fundamentally designed as scan-heavy engines.
They lack the sub-second, "needle-in-a-haystack" search capabilities provided by inverted indices found in traditional search engines.
This session explores how Apache Pinot fills this gap by integrating Apache Lucene segments directly into its distributed serving layer while maintaining the source of truth in Iceberg's Parquet format.
We will conduct a technical deep-dive into:
Segment-to-Parquet Virtualization: Pinot’s segment abstraction onto remote Iceberg/Parquet files without data duplication or heavy re-ingestion.
Hybrid Index Pinning: The mechanics of pinning Lucene Inverted and Text Indexes to local NVMe storage on Pinot servers while leaving the raw data blobs on S3.
Lucene I/O Orchestration: How the Pinot optimizes query plans to minimize S3 "Time to First Byte" by leveraging metadata-heavy index structures.
Iceberg Summit 2026 Sessionize Event
OpenSearchCon China 2026 Sessionize Event
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top