Run IT like a factory, using operations control center

By : |February 19, 2014 0

BANGALORE,INDIA: In today’s technology driven world, most business processes of a company are software dependent, be it a sophisticated product from a large vendor or processes captured in Excel and Word documents. In many large companies, if the software application fails to work, the company comes to a standstill.

This has increased the expectation from the IT department to keep software applications up and running all the time. The Service Level Agreements (SLA) at many companies requires greater than 99 per cent availability of the software for end-users!

IT landscapes have become very heterogeneous. Servers from different vendors, multiple databases and operating systems and applications from different products coexist to drive business processes. Business processes themselves are complex; requiring data exchange between different parties (vendors, customers, and partners). Some customers use an on-demand model for certain operations while others opt for an on-premise solution. This plethora of technologies and deployment options is a nightmare for the IT department and warrant experts with multiple skills.

As a result, the time for problem resolution escalates while driving up the Total Cost of Operation (TCO). But most importantly, it hinders the attainment of the 99 per cent availability SLA, which reduces the business continuity of the company.

Operations Control Center for IT
Just like how assembly lines function effectively with the help of an operations control center, this concept can be adapted to IT companies to gain much needed efficiency and reduction in TCO.

An Operations Control Center for IT consists of the following parts:
Monitoring Infrastructure
Central Monitors
Standard Operating Procedures
Event Management
Continuous Improvement
Monitoring Infrastructure

Multiple components in an IT Landscape make it critical to have a tool or a set of tools that monitor technical components and business processes 24×7 without manual effort. The monitoring infrastructure should be capable of surveying infrastructure components (like network routers, storage devices), application components (like application server, database, operating systems, end-user activity) and, business process components (like document throughput, process availability). Further, the monitoring infrastructure should be able to create alerts or exceptions when a set threshold is breached.

Central Monitors
This is a set of monitors/dashboards that show critical information about business processes (e.g., order throughput, dataflow with vendor/customer) or the IT landscape (e.g., availability of production systems, interface health, system exceptions). These display units are set up in control rooms to create the much needed transparency on the functions of an IT landscape.

Standard Operating Procedures
With the monitoring infrastructure and central monitors in place, the next step is to resolve any issues that these monitors might have caused to business users’ deliverables. This is where an identified Standard Operating Procedures steps in. It is important to document every critical alert resolution into an operating procedure. With this, even an IT operator without specific expertise will be able to go through the outlined operating procedure and understand how to resolve an alert. Drawing an analogy to assembly lines, this standard operating procedure minimizes the requirement for an expert for every alert. The time and effort required to acquire and train the personnel will also be reduced significantly.

Event Management
Event Management defines a structured process on what happens between alert arrival and alert closure. A small team of IT Operators positioned in the Operations Control Center resolve alerts using standard operating procedures. The IT Operator’s action starts from the individual alert. Contextual information about where the issue is, documentation and guidance is present in the alert details. However, there is a chance that some alerts can’t be resolved by this team of IT Operators, or that the resolution requires some change to the software (like a patch or upgrade). In such cases, the alert needs to be transferred as an Incident to the next level support team or Change.

Continuous Improvement
As the IT landscape evolves, so will the issues and problems. A static setup of central monitors and IT support organization are not enough to address these new problems and challenges. However, the setting up of Monitoring Infrastructure, Central Monitors, Standard Operating Procedures and, Event Management run the IT landscape in a stable manner. But a small change in Business Process or IT component could derail established SLAs. This makes it necessary to implement a Continuous Improvement model, so that it looks into optimizing the next critical business process in a structured way.

In conclusion, improved business continuity, better business performance, higher degree of automation, and reduced operations cost, are only some of the benefits of using the Operations Control Center to run IT.

(The author is vice president, product development, SAP Labs India. The views expressed in this article are those of the author and may not reflect those of SAP Labs India.)

No Comments so fars

Jump into a conversation

No Comments Yet!

You can be the one to start a conversation.