Feature

Automating workloads to improve business efficiency

In the age of the customer, more and more pressure is being put on workload automation. We look at where workload automation is heading.

Customers expect to be able to do things like consult and manage their bank account online and receive accurate information about their next flight.

140606_0593.jpg

Good workload automation helps businesses align their processes effectively with those customer expectations.

Business users are generally not aware of the power behind workload automation. Self-service is a golden opportunity to bring technology management professionals and business users together to automate data-intensive or computing-intensive processes, which some users are painstakingly cobbling together to perform part of a business process. It is also a prime opening into business process automation.

Data volume, time and accuracy challenges determine the framework in which workload automation must operate. The production of accurate information on time means that workload automation has to be able to transfer data securely and reliably between systems. It also means that a single point of control for orchestrating the use of resources is required.

For the variety of custom and packaged applications in the enterprise, this means that different batch schedules have to be integrated and consolidated to avoid resource conflicts. Forecasting when jobs will be complete and using automated provisioning to respond to incidents and resource shortages ensures that jobs will conclude on time. All of these constraints are now critical components that define today’s workload automation solution.

Extracting and mining the right data means that more and more business users are using workload automation through self-service interfaces to perform ad hoc business tasks. This implies not only a tight coordination between IT operations and the business but also a level of abstraction that allows non-technical users to define a workflow without using the dreaded scripts that were a common plug-in substitute a few years ago

Key requirements

Today’s requirements vary from the enterprise-wide evolution to job scheduling products to the most advanced business process automation. But eventually it all boils down to improving the control of application execution over the available infrastructure. Why? Simply because batch processing does not run at night any more, and is not really batch processing anyway but rather asynchronous processing: the execution of tasks that take too long to qualify as online transactional processing and can be executed at any time, resources permitting. This leads to three requirements.

Central control

First, you must have centralised control to manage multiple activities. Because process scheduling is needed by everyone in the enterprise, from the developers compiling code to business intelligence users to the typical asynchronous process runs, and because these activities actually run on the same infrastructures, workload automation is like spinning plates on sticks. Fundamentally, someone has to make sure that they keep spinning without colliding with each other.

Workload automation has to be enterprise-wide, aware of resource usage, and connected to IT systems monitoring

Workload automation therefore has to be enterprise-wide, aware of resource usage, and connected to IT systems monitoring. This could be done by integrating the automation solution with existing solutions or by including a monitoring capability within it. The latter is far less common, but fuller automation suites with such capability are emerging. This is a common feature among vendors, although the richer product families and third-party ecosystems of the large vendors – BMC, CA Technologies, and IBM – tend to give them an edge with such integration. But other vendors are on the right path: ASG integrates its automation capabilities with its IT management solutions, and Orsyp’s acquisition of Sysload adds performance and capacity management capabilities to its automation suite.

Timely completion

The second requirement from any solution is that processes have to complete on time. Whether processes are scheduled from calendar, event, or self-service triggers, many of them are critical to delivering accurate information to clients and consumers. This implies: first, the ability to forecast job processing time; second, the ability to control the actual processing; and third, the provision of alternative processing platforms in case of failure or capacity shortage.

To support this, performance, availability and capacity factors must govern the execution of any process by analysing service behaviours and then triggering the appropriate actions. These decision triggers are most effectively tied to service objectives as defined by business requirements. This capability is used to plan execution resources, based on a historical perspective (for example, resource usage for this job at this time of the year), a resource forecast (based on capacity management) or any form of predictive analysis.

Automated self-service

The third requirement is that users increasingly require an automation self-service capability. This is the first stage, where IT process automation meets business process automation. The end users of the process can choose from a service catalogue and initiate process execution themselves – without having to involve someone from IT operations. At this stage, automation presents either complete process suites or exposes reusable components that can be stitched together into new processes. Eventually, end users can plan the execution of business process steps that will self-provision resources and load applications to achieve business objectives.

Next steps in workload automation

Workload automation’s role in the enterprise is growing. The replacement of multiple schedulers by one central solution is one growth area, and will be facilitated by the move to software as a service (SaaS) solutions by the major vendors. This is still a work in progress in many instances. Redwood Software, with its three years of experience in the SaaS market, has probably the most advanced solution to date.

But this is not the only source of potential growth for the workload automation market, as the solution can be applied to control other activities in the enterprise, such as big data analysis or supervising the automation of ad hoc processes such as release management and change management. The convergence of workload and process automation is taking shape by running the specialised modules of software solutions, such as release management, under the control and orchestration of workload automation.

The tools you use to automate your IT processes today will extend to automating your business processes tomorrow. To shift your tool decision-making from tactical to strategic, you need to select vendors that address today’s needs while satisfying an architectural vision for the future. Selecting a product in this market requires a clear understanding of the enterprise’s strategic evolution in terms of infrastructure, applications and organisation. A list of major requirements must be drafted, prioritised and compared to features offered by the products available on the market today and in the next 12 to 24 months.

Jean-Pierre Garbani is a principal analyst at Forrester.


Email Alerts

Register now to receive ComputerWeekly.com IT-related news, guides and more, delivered to your inbox.
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy

This was first published in June 2014

 

COMMENTS powered by Disqus  //  Commenting policy