Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of GraphSearch.
This lecture covers the challenges of scaling up window aggregation in JIT-compiled engines, the deficiencies of existing stream processing systems, the development of a compilation framework for streaming queries, and the implementation details of a JIT-compiled system for streaming aggregation. It discusses the need for a compilation framework, the modifications made to the storage manager, the implementation of windowing stages, and the performance comparison with other systems. The lecture concludes with the scalability and throughput results, highlighting the linear scaling with the number of threads and the significant improvement in throughput compared to existing systems.