Consumers of IT services want speed and dependability above all else. When they envision a new service, they want it built and deployed in record time. While there has always been time pressure, it has become more pressing with the consumerization of IT. Consumerization describes the emergence of IT in the consumer market which then spread to business and government organizations. Users carry their devices and expectations to the workplace and want the same instant gratification, self-service mobile access, and constant stream of new apps they experience in their personal use of technology. Combine this with business intelligence capabilities advancing big data and analytics, and you end up with digital transformation: using data, pervasive connectivity, and easy to use devices to change the way businesses interact with their customers and conduct their core activities. These trends have raised the bar on the expectations of IT to innovate as quickly as possible.
IT has always had to balance rate of change with dependability and stability. We all know that most outages are a result of changes. Increase the rate of change and you increase the likelihood of breakage. Slow the rate of change in the modern era and impatient line of business (LOB) users may use third-party options to create shadow IT with the myriad of SaaS applications and other services outside the control of IT. While this can quickly meet the business need, it creates compliance, security, continuity, and integration problems. The best outcome is for IT to move faster with purpose and discipline to increase the rate of change without affecting dependability.
IT is reacting to the need for speed with new processes such as DevOps, new application deployment technologies like containers, and further expansion of cloud infrastructures. It is not enough to build and deploy apps and infrastructures faster—the command and control must advance as well. The back room of IT contains thousands of steps and details where infrastructures must be built up and configured, applications must be created and installed, and workloads must be defined, structured, and monitored. Automation is required to move faster with purpose and maintain consistency, auditability, and productivity. This is where the very old IT function of job scheduling, having evolved into modern workload automation, has become key to the success of a faster moving, digital transforming IT department.
Workload automation (WLA) allows an organization to visualize and execute end-to-end business processes. New technologies must be integrated into this process ecosystem. WLA continues to evolve to allow IT to move faster and integrate new technologies. Recent changes include the following.
- Support for All of the Major Public Clouds, Whether SaaS, PaaS, or IaaS – While most organizations use one or two public clouds (e.g., AWS, Google, vCloud, Softlayer, Azure, etc.) along with their on-premises operations, most WLA solutions support all the popular cloud providers. This is more than just being able to place and monitor workloads running in a public cloud. It includes the support for workload mobility to allow workloads to move freely in a hybrid cloud environment between on-premises and various public cloud environments as needed.
- Automate and Monitor Workloads from a Single Point Across Hybrid Environments – It wasn’t too long ago that running workloads in different environments meant running different WLA instances. Support for multiple operating systems brought a single view, but then the potential locations and environments blew up with the advent of public clouds. Today, you should expect to see all your workloads, wherever they may be at the moment, through a single view.
- WLA as a SaaS Offering – If you can manage workloads across hybrid cloud environments, maybe you do not want to host your WLA solution on-premises. Many solutions are now available as SaaS.
- More Complex Scheduling – Time or calendar scheduling fits some workflows, however more and more the trigger for a workflow may be the arrival or availability of a file or an interaction on a web page. Event-based scheduling allows for triggers such as file transfers, email or messaging, API, database, system startups, and other events to execute processes. Some WLA solutions can interrogate data and certain data can then be used as a trigger.
- Critical Path Monitoring – Over the past five years, predictive analytics has gone from a new idea supported by only a few products to being a part of every major WLA solution. It has evolved into the ability to not just monitor, but look downstream and highlight potential problems before they happen, providing the opportunity to head them off and predicting the impacts on other workloads.
- Modeling New Workloads – Carrying the analytics capabilities beyond current workloads, many products now have the ability to model new workloads, making run-time predictions and anticipating the impact.
- Integration with Service Management Systems and Awareness of SLAs – In addition to monitoring the critical path and modeling new workloads, integration with service management means that problems can be communicated, impacts can be understood, and priorities can be driven by SLAs.
- Open Integration Options – RESTful APIs and other open programmatic interactions are now supported in many WLA products, opening up full UI functions and expanding the integration of WLA with other systems.
- Integrate Applications and Share Customization with Others – Tools and support for building your own application integrations to the WLA system mean no longer waiting on vendors to offer certain applications or relying on minimal scripting support. As more users go down this path, there are opportunities to share these customizations with others in the community of users on your chosen WLA solution.
- Involving LOB Users in Workload Status – Moving beyond dashboards for operations staff, many WLA solutions are supporting dashboards for LOB users to get self-service status updates. Delivered through browsers and mobile apps, business users can be included in both the happy and not-so-happy outcomes of workloads.
While the list above exists in various forms today and will continue to evolve, there are some new features on the horizon that will bring more automation and control to IT operations. These include the following.
- Community Awareness – “Salesforce Communities” like features to allow more interactions and sharing of customizations and integrations within the community of users of various WLA products.
- Agent Change Management – WLA solutions tend to be agent-based. With thousands of servers in many on-premises and public cloud environments, updates to agents can be overwhelming. Change management features to update agents will be welcome improvements.
- Data Awareness, File Transfer Control, and Manipulation – Big data is big business for WLA solutions. While Hadoop and the many commercial tools have some built-in scheduling capabilities, they do not come close to the features and controls of the WLA solutions. Many of the workloads managed and monitored relate to data movement. More awareness and control, and even minor manipulations, of data and the file transfer processes are likely as big data gets bigger and more automation is applied to these workloads.
- Monitoring/Control of Release Process – As DevOps and continuous delivery become more common, there will be more need to orchestrate the application release process. Look for WLA to step up and offer solutions or at least monitoring of the release process.
Workload automation may have its roots in scheduling batch jobs on mainframes, but in its modern and still evolving form, it is a foundation of successful hybrid cloud computing and key to integrating new technologies and processes in a rapid fire IT world.
For information on how HelpSystems is addressing fast paced IT with workload automation, go to: http://www.helpsystems.com/skybot/skybot-scheduler
