# Chapter 2 Load Balancing Algorithm - LoadSheddingStrategy

In Chapter 1, we introduced the basic principles of load balancing and compared the performance of different algorithms to obtain an important scoring table. In the next two chapters, we will delve into the implementation principles of different algorithms and elaborate on the process of obtaining this scoring table. This analysis process is also the cornerstone for the subsequent design of the new algorithm `AvgShedder`. Only by accurately identifying the root causes of defects in the old algorithms and avoiding these issues in the design of the new algorithm can we scientifically and reasonably design a new algorithm. 

We previously introduced that the execution of load balancing is divided into `LoadSheddingStrategy` and `ModularLoadManagerStrategy`. Therefore, in this chapter, we will first introduce the `LoadSheddingStrategy` and then introduce the `ModularLoadManagerStrategy` in the next chapter.

<figure><img src="/files/FSSZsctxa9HbB3jmVrYW" alt=""><figcaption></figcaption></figure>

The load balancer calls `LoadSheddingStrategy#findBundlesForUnloading` to find the bundles that need to be unloaded. It passes the load data (`LoadData`) it collects and the broker configuration (configured by the `broker.conf`file) and obtains a `MultiMap` that maps brokers to bundles.

The load data `LoadData` includes the throughput, message rate, and resource usage (including CPU usage, network card in/out usage, direct memory usage, and memory usage) of each broker, as well as the throughput and message rate information of each bundle, for the use of the load balancing algorithm.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://tumbleds-library.gitbook.io/thetumbleds-library/chapter-2-load-balancing-algorithm-principles-and-analysis-load-shedding-strategy.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.