What if there was a way to see what was happening inside your IT operation while also seeing what’s happening to every associated environment? IPsoft has created a solution that’s able to do just that. Read this article to find out more.
One of the questions we ask ourselves here at IPsoft is how we can improve IT operations management. We have been on a quest to intelligently automate IT management via our virtual engineers for years, and they have learned how to fix almost anything in IT — from Oracle databases to bespoke applications to Amazon cloud environments.
Continuous improvements require insight, and insight is generally derived from real-time dashboards, reports, and analytics, all of which are plentiful throughout our solutions. Still, we wanted to take these insights one step further.
We created a visualization solution within our Amelia AIOps (formerly 1Desk) platform that allows users to see the end-to-end automation functionality in progress. The solution depicts live and historical flows of activities within the platform and associated environments, i.e. environments AIOps is integrated with from either an input or output perspective. The type of answers we’re trying to address, and they type of insight we’re trying to derive, include the following:
- What is happening in our IT environment right now? Is that a normal load or something unique?
- Is there a sudden influx of requests which will clog the system later?
- Where are the bottlenecks? Where are the hot-spots of issues originating?
- What channels are being utilized the most and can we prioritize some channels over others?
This at-a-glance view (see below) shows where demand is coming from, and it is divided into user-generated requests and system-generated requests. End-to-End Automation details how this demand flows through the system -- think of it like a control room for a power plant.
System-generated events are categorized via their system of origin, i.e. network, VMware, AWS etc. AIOps can visualize how various engines are analyzing the incoming events and metrics into anomalies and correlated events, and eventually determine that an event is what we call a “Fully Qualified Task” or FQT. User requests are categorized by their origination channel, i.e. web, chat, phone, email or external applications such as third-party ticketing systems like ServiceNow. Amelia in AIOps works with the requester through all these channels to establish an FQT for the end user’s issue if this was not derived from the initial contact.
The FQTs then enter the traditional ITSM flow, i.e. incident, problem, change, request, lookup -- based on the FQT. In End-to-End Automation, AIOps can now track the actual remediation of the issue or request. Is it being fixed through automation, RPA or escalated to human engineers? The platform shows the queues and bottlenecks (i.e. how many are waiting to be dispatched and assigned and, once assigned, how many are still waiting to get picked by engineers for remediation).
- For Further Reading: AIOps is the Immune System for the Enterprise
Once a human engineer works on the issue or request, AIOps can even see some of the tools engineers use within our own system to remediate or complete the request. At any point, managers can drill into the various work stages to view individual items and even select them to see the progression path through the system.
Finally, any remediation performed by a human engineer in AIOps always results in a new task for our Digital Labor Studio team, so End-to-End Automation also details the learnings that can be extracted from all the manual tasks still being performed within the IT organization.
The more data we’re able to provide to our users through AIOps features, the better.
This active visualization solution within AIOps allows managers to obtain deeper insight into the anatomy of their IT operations -- how issues flow through the system, and just as importantly how they’re resolved, among other things. This information allows managers to better determine how to prioritize channels, tools, and staffing based on volume, and it provides a visual means for spotting interesting or troubling trends.
The more data we’re able to provide to our users through AIOps features, the better. Soon, we will combine the End-to-End Automation view with more analytics, such as the cost of various channels, and alerts designed to warn users of upcoming slow-downs in resolutions. However, even before more analytics are added, we know that AIOps can provide IT operations managers with insights and a visual overview that can help them do their job faster and more effectively.
