hap： 多流动态实时分析系统.pdf

HAP: 多流动态实时分析系统 (A multi-streams dynamic real-time analytic system) 张李晔 新氦科技 he2.io Agenda • Big data environment • What is HAP • HAP Arch • Streaming underlying • Case study • Conclusion he2.io Big data environment • Batch • Streaming • Lambda • Kappa he2.io 4 questions for unbounded data processing* • What results are calculated? • Where in event time are results calculated? • When in processing time are results materialized? • How do refinements of results relate? he2.io What is HAP • K8s based software stack • HAP philosophy • One query for everything • “Eliminating” window in streaming • Multi streaming support • Dynamic job creation/deletion • Low latency • Sub-second analyzing he2.io Computing model Query Engine HAN Streaming Data Store Driver Driver Streaming Data Store Djoyn Druid Framework he2.io HAP Arch