"FFA2019分论坛：开源大数据及新型实时流基准介绍"

This document provides an introduction to a new streaming benchmark, focusing on the design and application of the benchmark. The benchmark is designed to assess the performance of streaming data processing systems and is intended to be used in the context of open source big data platforms. The introduction section provides an overview of the benchmark, including the key points such as the data set, workload, metrics, and classical benchmarks. The data set refers to the set of data used for testing the streaming data processing systems. The workload represents the specific tasks or operations performed on the data set, and the metrics measure the performance of the systems in terms of speed, throughput, and latency. The design of the streaming benchmark is discussed in detail, including the considerations and requirements for creating an effective benchmark. This includes the need for a representative and realistic data set, a diverse set of workloads to simulate real-world scenarios, and a set of comprehensive and meaningful metrics to evaluate the performance of the systems. The application of the streaming benchmark is also explored, highlighting the potential use cases and benefits of using the benchmark in evaluating and comparing different streaming data processing systems. This includes the ability to identify performance bottlenecks, optimize resource allocation, and make informed decisions when selecting and configuring streaming data processing systems. Overall, the introduction to the new streaming benchmark provides a comprehensive overview of the key components and considerations for designing and applying the benchmark in the context of open source big data platforms. This benchmark has the potential to be a valuable tool for researchers, developers, and organizations working with streaming data processing systems, as it enables them to make informed decisions and improvements in their data processing workflows.