Book Image

Intelligent Automation with VMware

By : Ajit Pratap Kundan
Book Image

Intelligent Automation with VMware

By: Ajit Pratap Kundan

Overview of this book

This book presents an introductory perspective on how machine learning plays an important role in a VMware environment. It offers a basic understanding of how to leverage machine learning primitives, along with a deeper look into integration with the VMware tools used for automation today. This book begins by highlighting how VMware addresses business issues related to its workforce, customers, and partners with emerging technologies such as machine learning to create new, intelligence-driven, end user experiences. You will learn how to apply machine learning techniques incorporated in VMware solutions for data center operations. You will go through management toolsets with a focus on machine learning techniques. At the end of the book, you will learn how the new vSphere Scale-Out edition can be used to ensure that HPC, big data performance, and other requirements can be met (either through development or by fine-tuning guidelines) with mainstream products.
Table of Contents (20 chapters)
Title Page
About Packt
Contributors
Preface
Index

Intelligent monitoring


vSAN environment monitoring is critical for a successful deployment. We have to follow the following monitoring practices:

  • General monitoring practices
  • vSAN Health Check plugin
  • vSAN Observer
  • VMware vRealize Operations Manager Monitoring
  • Monitoring design

General monitoring practices

vSAN supports monitoring datastores through the VMware vSphere Web Client, the HTML 5 client, and the vSAN Management API. The vSphere Web Client monitors different objects, like clusters and datastores.

Without effective control over the infrastructure, a VM or ESXi host sprawl can quickly diminish the return of investment from virtualization. Areas for improvement include assessing workloads to determine performance metrics that can then be used to create VM-specific vSAN policies to better suit the workload. A stripe policy of two nodes for write-intensive workloads that are bigger than the cache disk size may lead to performance improvements. We can minimize the troubleshooting time that is...