Book Image

Getting Started with Elastic Stack 8.0

By : Asjad Athick
5 (1)
Book Image

Getting Started with Elastic Stack 8.0

5 (1)
By: Asjad Athick

Overview of this book

The Elastic Stack helps you work with massive volumes of data to power use cases in the search, observability, and security solution areas. This three-part book starts with an introduction to the Elastic Stack with high-level commentary on the solutions the stack can be leveraged for. The second section focuses on each core component, giving you a detailed understanding of the component and the role it plays. You’ll start by working with Elasticsearch to ingest, search, analyze, and store data for your use cases. Next, you’ll look at Logstash, Beats, and Elastic Agent as components that can collect, transform, and load data. Later chapters help you use Kibana as an interface to consume Elastic solutions and interact with data on Elasticsearch. The last section explores the three main use cases offered on top of the Elastic Stack. You’ll start with a full-text search and look at real-world outcomes powered by search capabilities. Furthermore, you’ll learn how the stack can be used to monitor and observe large and complex IT environments. Finally, you’ll understand how to detect, prevent, and respond to security threats across your environment. The book ends by highlighting architecture best practices for successful Elastic Stack deployments. By the end of this book, you’ll be able to implement the Elastic Stack and derive value from it.
Table of Contents (18 chapters)
1
Section 1: Core Components
4
Section 2: Working with the Elastic Stack
12
Section 3: Building Solutions with the Elastic Stack

Inferring against incoming data using machine learning

As we learned in Chapter 4, Leveraging Insights and Managing Data on Elasticsearch, ingest pipelines can be used to transform, process, and enrich incoming documents before indexing. Ingest pipelines provide an inference processor to run new documents through a trained machine learning model to infer classification or regression results.

Follow these instructions to create and test an ingest pipeline to run inference using the trained machine learning model:

  1. Create a new ingest pipeline as follows. model_id will defer across Kibana instances and can be retrieved from the model pane in the Data Frame Analytics tab on Kibana. model_id in this case is classification-request-payloads-1615680927179:
    PUT _ingest/pipeline/ml-malicious-request
    {
      "processors": [
        {
          "inference": {
            "model_id...