Essentials of Reliable Stream Processing: A Visual Demo using Apache Flink

摘要：

這場演講的主目的是傳達我認為一個可靠串流運算系統必須要有的三大關
鍵因素：Correctness, Robustness, 與 Maintainability。
不紙上談兵，演講中會直接 Demo 這三大關鍵因素的代表意義以及若缺
少了會造成的影響。Demo 將會建立在 Apache Flink 上，但觀念實
際上通用到任何分散式串流運算系統的選擇和考量，希望讓聽眾未來對於
導入串流運算到架構中有所幫助。

The aim of this talk is to go through the 3 key essentials
I deem a must for a reliable stream processing framework:
Correctness, Robustness, and Maintainability.
Instead of only theory talking, I will demonstrate by
letting attendees "see" the essentials at work, and what
will happen without them. The demonstration will be built
on Apache Flink with Flink's out-of-the-box support for
these essentials. However, the general take-away for
attendees is to be able to have the 3 key essentials in
mind to help them with deciding on a stream processing
system for their data infrastructures.

講者簡介：

Tzu-Li (Gordon) Tai is a Committer member at the Apache
Flink project. He works at VMFive as a software engineer,
building scalable backends and data platforms to handle
streaming data for mobile advertisement. He is mostly
interested in the development and design of distributed
computing software systems.

• ← Achieve Big Data Analytic Platform with Lambda Architecture on Cloud
• 如何建置關鍵字精靈 →