For years, search engines have been the first place to go to look for information, from who batted cleanup for the 1979 Pittsburgh Pirates (Willie Stargell) to how to optimize WebSphere Application Servers.
While search engines have always been a great source of general information, they were not personalized for you and your situation. Until now.
The cognitive era is here, which means you can apply natural language interaction, moving beyond generic trends and standard benchmarks to get real-time, specifically targeted data and insights. It was this concept that drove our third Connect to Cloud Cognitive Build team to create the new Cognibot application.
IT operations managers around the world deal daily with the realities of trying to keep their environment not only up and running, but also optimized to the best of their ability. Industry experts have studied the four phases of IT service management when it came to repairing problems:
- Mean time to identify
- Mean time to know
- Mean time to fix
- Mean time to verify
The team noticed that while there was a great deal of information available for identifying a problem, there was significantly less available to recommend what to do to fix it.
That brings us back to search engines. In many cases, IT managers look at system reports and attempt to search online for possible causes and resolutions. Not only is that approach time consuming, but what they find is not tailored for their specific issue.
What if someone applied cognitive services to the problem?
The Cognibot project does just that. The team has created an interactive service that combines the knowledge of what one has already done to fix a problem, with the real data from the specific IT environment. It then adds Watson capabilities so users can ask natural-language questions and get response customized for their specific issue.
Here’s an example: suddenly your IT department gets flagged with an issue in your WebSphere Application Server deployment. Normally, an IT subject matter expert will get called in to identify the issue, search for the solution, the execution the recovery plan. What if we could streamline that process? Your IT department still gets the notification, but instead of searching for the answer your Cognibot interface has already analyzed your real data, researched fixes that have worked in the past, and recommended a solution that will work in this case. All you have to do is click “accept” to enact the fix, helping reduce your mean time to repair drastically. Now that’s a cognitive solution.
Interested in learning more? Join us over the next few weeks as we track the team’s progress toward creating the first IT Operations consultant.
Share this post: