Data Processing

Zachary Chapman

Follow us on LinkedIn

9 Dec 2019
  • Technology and Cyber Risk

This article is a part of Risk Update 2.


Technology functions are increasing their use of automation for data processing. Automated data processing requires strong systems and supporting infrastructure in order to be leveraged effectively in keeping up with technology transformation.

Companies must ensure processes are being continually streamlined, and avoid the risk of falling behind their competitors due to reliance on legacy systems preventing complete automation integration into business operations.

Automation through Scripting

Traditionally, scripting has been used to automate processes using pre-built commands to run a list of tasks. Scripts, however, require manual initiation, and the list of commands included in the script cannot be deviated from, restricting  customisation and the potential for large scale automation.

Job Scheduler Automation

The limitations in the design of scripts can be overcome through the use of a job scheduler, where sequences of scripts are automatically run on a schedule. Job schedulers enable configuration of jobs to run at specific times or when an event is triggered, which can be set up to run in batches after hours to avoid using processing capacity of users during business hours.

Examples of job schedulers include Windows Task Scheduler and the CRON utility for Unix, which come pre-installed on systems supporting the associated operating system. More advanced software such as the Advanced Task Scheduler designed for IBM systems, allows for additional customisation capability for task automation.

Drawbacks of Job Scheduling Systems

Job Schedulers provide powerful process automation that increases productivity in business operations, however there are limitations to the effectiveness of automation between systems, preventing complete integration across platforms in a workflow. Examples are listed below.

  • Siloed job schedulers that are incompatible with other platforms prevent complete synchronisation of end to end process automation, often requiring manual intervention between processes in a workflow.
  • Processing errors between platforms are likely to occur due to the lack of a centralised workflow management tool. The risk of human error with the reliance on manual identification and remediation of issues.
  • Security vulnerabilities on legacy job schedulers reaching end of life that are exempt from patches, provide hackers opportunity to perform malicious activity.
  • Compliance gaps are likely to emerge in the ability for job schedulers to meet the changing criteria for systems to be considered compliant to the appropriate regulations.

Use of Enterprise Workload Automation

Workload automation tools provide a solution to the limitations of job scheduling systems, by enabling integration of job processing across all platforms supporting business functions. Automated workflows have the following benefits:

  • Visibility is achieved over the end to end automated process, ensuring issues are  highlighted in a centralised system, such as the dashboards used in Control-M, to highlight the success and failures of job completion, with tasks not meeting SLAs being flagged.
  • Remediation of jobs with issues identified is improved, with manual identification and remediation of issues not required as a result of implementing a centralised monitoring interface to identify areas of concern in real time.
  • Maintenance of workflow tasks is simplified, in that jobs can be modified in a single, centralised tool, removing the complexity and potential for error that would otherwise be required to update several job schedulers.
  • Auditing and regulatory compliance is improved with tools such as AutoSys Workload Automation allowing role-based access control for accessing job schedulers, with tracking of changes made to the workflows and reporting available for review and identification of areas of improvement.
  • User education on the operation of the workload automation tool is simplified, with users supporting the system only requiring education on a single tool. This eliminates the need for companies to employ subject matter experts across a multitude of systems on separate platforms, and saves time and cost associated with the reissuance of education to each.

The unfortunate downside to workload automation is that well established organisations have business processes integrated into existing infrastructure that is outdated.

Organisations should weigh up the costs, risks and benefits associated with a large scale uplift to the current automation tools, and if a parallel or phased implementation of an improved workload automation tool is best suited in integrating current systems to the new automation tool upon refreshment or replacement after reaching end of life.

The Future of Workload Automation

Workload automation tools have provided a necessary uplift to existing automation processing systems, leveraging technologies such as artificial intelligence to provide advanced methods of automation into business environments.

Workload Automation Simplification is emerging with the use of user friendly interfaces removing the requirement for job scheduler experts. Applications, such as Microsoft Flow, allow for pre-built templates or customised processes to automate workflows on local and cloud based applications. Users must ensure compliance is maintained to security policies, even with a simplified interface.

Adaptation to Emerging Technology is vital to maintain the pace within the ever-changing technological landscape. Systems, such as IBM’s Workload Automation Tool, enable automation capability across Internet of Things devices over hybrid cloud environments, and integration of Big Data analytics through Hadoop into the existing infrastructure. Risk management must evolve to accommodate this transformation, ensuring  controls are continually updated to ensure sufficient protection is in place for emerging trends.

Machine Learning is being adapted to further automate error detection, correction and prevention. Continuous learning enables mitigation of operational risk of workload automation through trend analysis of systemic issues and identification of trends, removing reliance on human intervention.

Robotic Process Automation allows for “Robot” accounts to be configured and trained by humans and, through machine learning, carry out repetitive tasks and interact with interface objects, such as buttons and fields. Operational risk is prevalent if Robot accounts are configured incorrectly, and account privileges should be managed through tools such as CyberArk’s check-in/check-out authentication through encrypted sessions. Failure to do so opens up the risks of using Robot account privileges and access to perform malicious activity.

Closing Thoughts

Workload automation must remain adaptive to the rapidly transforming digital landscape in order to remain effective. Integrity must be maintained, and controls around error detection, user access and logging must be adapted to the new ways of working. The potential for workload automation is limitless if companies embrace change and are willing to sacrifice the time, effort and costs associated with migration away from legacy systems.

Amstelveen Risk Update: Edition 2, December 2019
Download the article

You may also like

Let us tell you more

Risk management expectations are evolving rapidly. How well is your organisation equipped to respond?