We are not only defining exactly what cloudops are, but also clarifying what technology is needed to solve the core problem.
As with all cloud computing situations, it is useful to disassemble the core components of a running cloudops solution such as AIops. Also, define what the technology needs to do and the value it brings to the table. To this end, we have selected six features provided by the cloudops tool.
Observe and gather data From any number of systems needed to further analyze and find patterns of action. It has several components, including the ability to leverage connectors and agents to communicate with managed systems and to reliably return data to some centralized cloud ops system.
Correlate Large amount of system data (Noise) In a meaningful way. This includes identifying patterns such as the source of the data and grouping the data before further analysis.
analyse pattern To identify the problem and the root cause. This is where AIops or popular cloudops tools make money. You need to be able to find patterns in the data that you collect and associate with to identify patterns that indicate current problems, such as network device failures. More importantly, it is about predicting potential problems. Proactive cloudops help avoid key issues such as identifying cloud storage systems that are initiating I / O errors that may indicate an imminent failure.
share Observable survey results Work with users on your ops team to automate the process of automatically responding and fixing issues. There is only one thing that shows that something is wrong. Another way is to make sure that those processes and those who can fix the problem are notified. Here things are improving rapidly, such as automatic ticketing systems and self-healing processes.
respond To the problem Launch an automated fix or collaboration to get the fix. This means that there is a mechanism in place to fix the problem. Automation has been taken over here as part of the cloudops tool or as another orchestration layer that allows you to define how to fix common problems without human involvement.
Notice Reports and dashboards As a result, cloudops users can see both strategic and tactical data about the effectiveness of the system over time. The dashboard shows current system status and status trends, so you can predict future status. The cloudops team is hesitant to leverage these across the team, but my advice is to improve the situation by allowing everyone involved in cloudops or development to see these metrics in real time. To be able to make the right decisions.
Again, there is no magic to solve the cloudops problem. Many of my recommendations may not be possible for some enterprises without multiple AIops or other cloudops technologies in place. It depends on the type of system and cloud you are running on, and the number and type of applications and data stores.
However, working on these six concepts is a good start to getting where you need them.
Copyright © 2021 IDG Communications, Inc.
6 things your cloudops technology has to do now
Source link 6 things your cloudops technology has to do now