In previous installments, I discussed how the essence of IT operations is to guarantee a continuous flow of data in the most secure, efficient, and effective way. Today we will look at how ignio can streamline your IT operations and make them autonomous. And to do that, I’ll first provide a quick review of how IT operations work.
Because every organization large enough to be called an enterprise is now digital by necessity, with IT production environments designed to provide services 24 hours a day, 365 days a year, IT support must also be on the job around the clock. Every day an average IT organization moves trillions of bytes, and in the real world it’s impossible to guarantee 100% success for each byte transmitted.
This means that IT operations must guarantee not only availability (the lights are on, and applications don’t crash) but also reliability (IT solutions always work as expected).
For example, for a consumer banking app, “availability” means it always opens when you tap the button. Of course, that should be as close to 100% as possible. But “reliability” is even more important – that is, your bank transactions are always accurate. To complicate IT issues further, applications may behave perfectly in isolation but develop unforeseen problems when interacting with an unfamiliar program or operating system update.
The IT operations function, or ITOps, handles issues by creating a “ticket” for each work request, whether generated by a human being or by the technology itself. These tickets can be reactive, preventive, or proactive.
Tickets are usually organized by different levels of resolution complexity and therefore the support skill they require. The lowest is Level Zero, where ticket resolution is very simple, ranging up to Level 4, where ticket resolution requires deep technical skills. (For example, the full system goes down.)
Each level (0 to 4) has a queue assigned. Each ticket is assigned to a specific queue, and it is resolved based on a FIFO (first in, first out) approach. To regulate resolution speed, severity is assigned to each ticket.
- Level 0: Activities managed by end users themselves. In many cases a ticket is not even created – for example, changing your password when it is about to expire.
- Level 1: Activities performed by a help desk (also called “command center”). These tasks are highly repetitive; they react to a specific situation. (For example, expanding memory following a monitoring alert or re-enabling a user who got stuck performing some specific transaction.) All tickets usually land first with Level 1 to be resolved or re-routed to Level 2 or 3 resolving groups. Level 1 teams have a generic knowledge of IT solutions and operate mainly following Standard Operating Procedures.
- Levels 2 and 3 are given to more specialized resolving groups. These teams are composed of subject matter experts (SMEs) who have a detailed knowledge of that specific IT solution. They can do both root cause analysis and coding/configuration fixes, if needed. There are multiple different Level 2 and 3 teams grouped by technologies (for example, Oracle databases, SAP manufacturing software, VMware, and the network). What differentiates Level 2 problems from Level 3 can be fuzzy; typically, it is the level of skill needed to resolve them. Level 3 requires an almost architecture level of skill. Most of the Level 3 time is spent gathering information to fully understand the problem to be solved and where a fix needs to be applied.
- Level 4: These urgent problems usually bypass the resident support team. They’re sent directly to software or hardware provider teams, who have the best product knowledge and provide support for very complicated issues.
The majority, more than 80%, of tickets belong to Levels 0, 1, and 2. Over 95% of these low-level tickets can be solved following a standard operating procedure. No diagnosis or creativity is required: Just read the instructions and execute.
Taking the tediousness out of tech support
These tasks are time-consuming and repetitive, not the kind of challenge IT staff could take pride in overcoming, and not a good use of scarce and expensive IT talent. So, they’re an obvious target for automation – allowing your technological components to autonomously diagnose and fix their own problems. And that’s where ignio comes in.
Using ignio to make your IT operations autonomous means that it will take over many of the routine activities to resolve low-level tickets:
- Performing simple tasks such as updating security certificates, installing software patches, or rebooting servers, so humans don’t have to (a reactive, self-healing approach).
- Managing noise: Using its ML algorithms, ignio can also reduce Level 0/1/2 tickets by eliminating noise (redundant machine alerts) and consolidating tickets that all stem from the same issue.
- Anticipating and autonomously fixing issues. For example, ignio can proactively expand memory where it sees that a growing database table is reaching the system’s capacity limit; it predicts that this could eat up all the available memory and successfully prevents it from halting operations.
- Predicting an issue before it happens in the first place, then taking preventive actions (powered by Machine Learning).
- Suggesting improvements to head off future problems. (In this scenario ignio requires human intervention to design and implement a solution.)
And even when ignio can’t autonomously resolve higher-level tickets, it can still help the human experts by offering suggestions from its built-in library of solutions.
The diagram below shows specific ways ignio acts autonomously to reduce a human IT team’s workload for each level of ticket. It could almost be considered a member of the team in its own right!
How ignio reduces ticket workloads at each level
A single AIOps platform with multiple options
The picture below summarizes ignio’s taxonomy. At the core are the ML-driven functions of blueprinting and profiling a customer’s IT system, and managing events, workload, and capacity. From there, you can build in sophisticated AIOps functionality and tools to optimize digital workspaces and ERP. Finally, you can add business health monitoring and observability.
Note that all these ignio products are part of a single, unified AIOps platform. There is no need to integrate different third-party components for each of the individual functions. ignio has multiple certified integration points which let it fit perfectly in many production environments.
Basic ignio taxonomy by underlying technology
Deploying in sensible stages
There are multiple approaches to deploying ignio. If you want to maximize its potential, one best practice is to start by implementing its event management and capacity management functions, a relatively easy way to address the purely reactive problems at Levels 0, 1, 2, and even 3. This first stage will help to reduce noise and provide some early wins.
The next stage is to leverage ignio’s ML capabilities for autonomously solving problems in event and capacity management. These are the “complex reactive” problems illustrated in Diagram 1.
Of course, additional AIOps use cases can be added. Once an ITOps team has deployed ignio to solve event management, they’ll often find further efficiency gains in adding more functionality, creating full self-healing use cases. Such an implementation footprint will look like the diagram below:
First stage of ignio deployment, with basic AIOps functions including event and capacity management
In the next phase of deployment, add ignio AI.Workload Management. This is a sequential approach. However, if your organization has the budget and wants a fast transformation, it is possible to implement all these functionalities in parallel. (A good example is Walgreens-Boots Alliance, as described in the first two customer videos here.)
At the end, this is how your implementation will look:
Second phase of ignio deployment, adding ignio AI.Workload Management
I have also met customers that implemented only ignio AI.ERPOps or ignio AI.Digital Workspace, as shown in the diagrams below. While this choice immediately addresses SAP–specific concerns, it will only cover a specific set of your problems.
Alternative starter implementation, ignio AI.ERPOps only
The final consideration is whether you have SAP as part of the IT landscape. If so, you’ll want ignio AI.ERPOps. We designed this product with specialized functions to optimize Intelligent Automation in SAP landscapes, especially optimizing the data flows that traverse any SAP system. ignio can support a constant flow of data in and out of SAP systems, as I discuss here.
One further option is to add focused monitoring functionality with ignio Observe, a module of ignio AIOPS that we recently introduced to optimize your third-party monitoring tools. It isn’t required, since ignio has many integration points with other monitoring solutions. However, by consolidating those monitoring tools into ignio, you can reduce the overall annual license cost.
A full complement of ignio products
Transforming IT from a cost center to a profit center
Looking at all these diagrams, I hope you can see ignio’s potential. With full implementation, my experience shows these transformational KPIs:
- Overall MTTR improvement (often dropping from days to hours/minutes)
- 30% to 50% of total incidents managed by ignio (this range indicates a solid base of use cases deployed).
- 95% of alerts are managed by ignio.
- At least 25% elimination of repetitive and manual IT activities.
- 60 to 80% of SOPs are managed by ignio.
And that creates this kind of business value:
- Customer satisfaction increases (MTTR improves, incident management improves, better coverage).
- Operating cost decreases (better utilization of ITOps resources).
- IT workers’ satisfaction increases when they lose many tiring, routine tasks and gain greater control of operations.
- Negotiating power with System Integrators increases (because tribal knowledge is now in ignio).
- Ability to create value via ITOps increases (because now you can focus on what the business needs).
Overall, adopting ignio as your autonomous ITOps digital team not only helps you to optimize your IT operations but also to create business value from IT operations. This means finally your IT department is no longer perceived as a cost center but as a profit center. For example, if ignio can reduce errors in master data, pricing, or billing, your company gets an immediate revenue boost.
If I didn’t see you at any of our 2022 events — most recently the Gartner IOCS show in Las Vegas — my apologies. I look forward to more face-to-face interactions with the Digitate and AIOps community in the coming year. We’ll start announcing the Digitate 2023 event schedule in the next month or two. In the meantime, if you have questions about Digitate’s AI-driven solutions and how they can help your digital enterprise run more smoothly, feel free to reach out.