Stream Processing Articles

  • Unified Streaming And Batch Pipelines At LinkedIn: Reducing Processing time by 94% with Apache Beam

    March 23, 2023

    Co-Authors: Yuhong Cheng, Shangjin Zhang, Xinyu Liu, and Yi Pan Efficient data processing is crucial in reducing learning curves, simplifying maintenance efforts, and decreasing operational complexity. This, in turn, helps engineers to develop and deploy data processing applications quickly and easily, powering various business requirements, and enhancing member...

  • FastIngest: Low-latency Gobblin with Apache Iceberg and ORC format

    January 6, 2021

    Co-authors: Zihan Li, Sudarshan Vasudevan, Lei Sun, and Shirshanka Das Data analytics and AI power many business-critical use cases at LinkedIn. We need to ingest data in a timely and reliable way from a variety of sources, including Kafka, Oracle, and Espresso, bringing it into our Hadoop data lake for subsequent processing by AI and data science pipelines. We...

  • from-lambda-to-lambdaless-lessons-learned

    From Lambda to Lambda-less: Lessons learned

    December 1, 2020

    Co-authors: Xiang Zhang and Jingyu Zhu Introduction The Lambda architecture has become a popular architectural style that promises both speed and accuracy in data processing by using a hybrid approach of both batch processing and stream processing methods. But it also has some drawbacks, such as complexity and additional development/operational overheads. One of...

  • table-comparing-the-nexmark-benchmark-results

    Building a better and faster Beam Samza runner

    October 1, 2020

    Co-authors: Yixing Zhang, Bingfeng Xia, Ke Wu, and Xinyu Liu Since Beam Samza runner was developed in 2018 at LinkedIn, we now have...

  • mock-screenshot-of-the-recruiter-usage-dashboard

    Bridging batch and stream processing for the Recruiter...

    July 14, 2020

    Co-authors: Khai Tran and Steve Weiss Batch and streaming computations are often combined together in the Lambda architecture, but...

  • change-data-capture

    Open sourcing Brooklin: Near real-time data streaming at...

    July 16, 2019

    Editor's note: This blog has been updated. Brooklin—a distributed service for streaming data in near real-time and at scale—has been...