Q&A: IT Automation, Virtualization, and the Cloud

Virtualization and cloud computing present a blind spot: traiditional IT management and automation tools don’t offer the necessary visibility and control. What can IT do?

As businesses use virtualization and now cloud computing to store data and help run applications associated with core internal programs and customer offerings, they are faced with some very real challenges. One is a big blind spot -- a loss of visibility and control because traditional IT management and automation tools aren't making the jump to virtualization and cloud. For answers, we turned to UC4's chief marketing officer, Fred Kohout.

Enterprise Systems: What types of new IT infrastructure management requirements are important when considering a cloud or hybrid data center?

Fred Kohout: Cloud-class automation platform needs to meet four key requirements. First, it needs to be scalable to fully take advantage of the elasticity that virtualization and cloud offer. It needs to work in heterogeneous environments, spanning not just physical, virtual, and cloud, but also multiple applications, platforms, OSes, and hypervisors. A cloud class automation platform also needs to offer always-on processing, in effect acting as an automation cluster to provide high availability. Finally, it requires an intelligence layer capable of sensing and responding in real time to complex, interdependent events taking place across the IT infrastructure in order to predicatively model and manage processes.

Most IT shops have a heterogeneous environment, which makes cross-platform visibility key. Isn't IT automation considerably more difficult in a cloud or hybrid environment?

Any automation platform in the future has to be heterogeneous by design if you believe you’re going to access resources no matter where they reside. It’s not that automation is more difficult in a cloud or hybrid environment; the issue is that so few tools are a capable of addressing the new requirements I just mentioned. Today, IT organizations do a lot of cobbling together of management and automation solutions or they simply live with silos because the different vendors comprising a data center only offer “tools” that manage and automate their own applications. Those approaches won’t cut it long-term in a virtual or cloud environment.

How do traditional IT management and automation solutions address cross-platform visibility? Where do they fall short?

With most scheduling solutions, users usually aren’t getting a complete picture. They’re managing multiple processes and relying heavily on institutional knowledge instead of defined procedures. It’s a combination of manual labor, custom procedures, and policies and scripting, all of which doesn’t scale and can’t extend. You can make it work, but it’s not replicable and certainly not manageable. The traditional approach makes it difficult for IT to understand workload dependencies across platforms. When problems occur, mapping out a solution is very difficult. With a cloud-class solution, you eliminate all of that.

What are the consequences when IT management and automation tools can’t make the jump to a cloud or hybrid environment? How does IT cope? For example, do they not use automation, use it along with other tools, or perform tasks manually?

Typically, the way IT copes is to throw bodies at the challenge at hand. They’re putting highly skilled labor in front of a monitor when they could be a strategic resource for the organization. This approach will always be reactive, unable to anticipate changes in business conditions or optimize processing to meet SLAs. When you look at the costs of operating manually or in silos, both from a time and resource perspective, it becomes quickly apparent that an enterprise automation solution will pay for itself extremely quickly.

What are the real-world business implications this “blind spot” can cause when not provisioned for at the outset?

The question really becomes what part of the business function won’t be adversely affected? It goes back to the fact that platforms, applications and data are all intertwined and inter-dependent, regardless of where they reside within the IT infrastructure. From longer cash to collection to inadequate supply, the issues are very real to the business. The processes associated with the application that you automated on premise to keep the lifeblood of the business up and running smoothly will need to be done manually. From making sound financial and resource-based decisions to running core business systems, a blind spot can be costly. It becomes a question of do I want to manage it through headcount or through technology?

Is visibility enough? Aren't there other considerations, such as the level of control IT automation provides across platforms?

In a hybrid or cloud environment, automation takes center stage. If you don’t have an automation platform, you can’t manage the complexity you’re going to encounter. Lots of tools provide just visibility or monitoring, but the reality of virtual and cloud environments is automation solutions have to evolve from simply being batch schedulers to dynamic, intelligent service automation platforms.

Beyond cloud migration, where is IT automation headed? What new challenges (from new technologies, for example) lie ahead?

The automation architecture has to expand to include technology such as CEP that delivers the real-time intelligence to enable IT organizations to proactively monitor and dynamically interpret information, processes, and events across the enterprise and trigger the right downstream action at the right time. The new challenges that lie ahead are in intelligent remediation and dynamic orchestration of resources to changing business conditions.

What is UC4 working on in the automation market, especially to target the "blind spot" you mentioned?

We’re working on building a next-generation automation platform for service governance that works seamlessly across heterogeneous platforms and computing environments. Our objective is to enable real-time infrastructure and help customers gain maximum utilization and performance. The offering stands to extend typical virtualization cost savings by an additional 20-30 percent, also helping our customers lower data center capex via greater efficiency and automation. Bottom line: we are helping customers improve service level delivery, increase the quality of service from IT to the business, and drive corporate financial performance through the data center.

Must Read Articles