Hi Harald,
Do you ever triage a problem, only to find after hours/days of searching you finally find the root cause of the problem buried deep in the application log files? Then AI Manager can help do that task for you, in a fully automated manner, without the need for data scientists.
AI Manager is the newer capability and part of our Cloud Pak for Watson AIOps. It does something I don't believe exists elsewhere. It takes normal log file data as input and automatically;
learns what is normal
creates models to monitor the ongoing log data 24x7
alerts when things are abnormal.
I see it as akin to Metric Manager which does that by ingesting metrics, numbers from one or more monitoring/APM/observability tools and other sources, but for logfile data. The later being much harder to automate as you are not dealing with numerics but the subtleties of understanding text and being able to process that at scale, automatically, with NLP. Something IBM has quite a market reputation for using Watson.
AI Manager is also the capability that lets us rank and score the risk of changes*, learn and suggest next best actions based on past ServiceNow incidents and more.
Ingo has a good article here, as well as more on change risk advisor using the second link.
https://www.ibm.com/cloud/architecture/architectures/serviceManagementArchitecture?cm_sp=Blog-_-blogcta-_-ArchCenter
* https://www.ibm.com/cloud/blog/steps-to-build-an-automated-system-for-change-risk-assessment
------------------------------
Angus Jamieson
IT Service Management Solutions Architect
IBM
Edinburgh
------------------------------
Original Message:
Sent: Mon September 13, 2021 01:17 AM
From: Harald Biedermann
Subject: Use cases of AI Manager
Hi,
I am thinking of which use cases does the AI Manager address? What are your experiences?
Best regards, Harry
------------------------------
Harald Biedermann
Teamleader Monitoring & Eventmanagement
OEBB-BCC GmbH
1030 Vienna
------------------------------