By Dusty Simoni, 2nd Watch
The holy grail of IT operations is to achieve a state where all mundane, repeatable remediations occur without intervention, with a human only being called into service for any action that simply cannot be automated. This allows not only for many restful nights, but it also allows IT operations teams to become more agile while maintaining a proactive and highly optimized enterprise cloud. Getting to that state seems like it can only be found in the greatest online fantasy game, but the growing popularity of “AIOps” gives great hope that this may actually be closer to a reality than once thought.
Skeptics will tell you that automation, orchestration and optimization have been alive and well in the datacenter for more than a decade now. Companies like Microsoft with System Center, IBM with Tivoli and ServiceNow are just a few examples of platforms that harness the ability to collect, analyze and make decisions on how to act against sensor data derived from physical/virtual infrastructure and appliances. But when you couple these capabilities with advancements brought through AIOps, you are able to take advantage of the previously missing components by incorporating Big Data analytics along with artificial intelligence (AI) and Machine Learning (ML).
As you can imagine, these advancements have brought an explosion of new tooling and services from cloud ISVs intended to make the once utopian “autonomic cloud” a reality. Palo Alto Network’s Prisma Public Cloud product is a great example of a technology that functions with autonomic-type capabilities. The security and compliance features of Prisma Public Cloud are pretty impressive, but it also has a component known as User and Entity Behavior Analytics (UEBA).
UEBA analyzes user activity data from logs, network traffic and endpoints and correlates this data with security threat intelligence to identify activities—or behaviors—likely to indicate a malicious presence in your environment. After analyzing the current state of the vulnerability and risk landscape, it reports current risk and vulnerability state and derives a set of guided remediations that can be either performed manually against the infrastructure in question or automated for remediation to ensure a proactive response, hands off, to ensure vulnerabilities and security compliance can always be maintained.
Another ISV focused on AIOps is MoogSoft, which is bringing a next-generation platform for IT incident management to life for the cloud. Moogsoft has purpose-built machine learning algorithms that are designed to better correlate alerts and reduce much of the noise associated with all the data points. When you marry this with their AI capabilities for IT operations, they are helping DevOps teams operate smarter, faster and more effectively in terms of automating traditional IT operations tasks.
As we move forward, expect to see more and more AI and ML-based functionality move into the core cloud management platforms as well. Amazon recently released AWS Control Tower to aide your company’s journey toward AIOps. While coming with some pretty incredible features for new account creation and increased multi-account visibility, it uses service control policies (SCPs) based upon established guardrails (rules and policies). As new resources and accounts come online, Control Tower can force compliance with the policies automatically, preventing “bad behavior” by users and eliminating the need to have IT configure resources after they come online. Once AWS Control Tower is being utilized, these guardrails can apply to multi-account environments and new accounts as they are created.
It is an exciting time for autonomic systems capabilities in the cloud. Every company can now automate, orchestrate and proactively maintain and optimize its core cloud infrastructure.
About The Author
Dusty Simoni is Senior Product Manager for 2nd Watch.