[Compilation] Getting started with streaminsight application optimization (1) -- Overview

Refer to blog:Http://blogs.msdn.com/ B /domain_connect4

Overview

Learning how to optimize is an essential part of streaminsight development. How can streaminsight maintain a high throughput, a low latency, and as little memory as possible in complex event stream processing? For streaminsight application developers, understanding the basic tuning knowledge is the key to solving the above problems.

In general, streaminsight optimization can be divided into three categories:

Throughput (throughput)

• Input throughput (input throughput)
• Output Throughput)
• Total Throughput (total throughput)

Latency)

• End-to-End latency)
• System latency)
• Informational latency)

Memory usage (memory)

• Index memory usage)
• Event Memory usage)
• Total memory usage (total memory)

The above three indicators are key elements for identifying the performance of streaminsight applications and are indispensable. It is difficult to determine whether a system has good performance or is faulty if it is based on a specific indicator.

In the next series of blog posts, we will introduce these streaminsight Performance Elements one by one.