Across the board, enterprise software development organizations are grappling with server sprawl, setup and provisioning overhead, and costly system failures.

• Server Sprawl. Organizations face an explosion in the number of machines required to develop and test enterprise applications, with some application development (AD) organizations reaching server-to-staff ratios over 7:1, even though average server utilization rates are often below 10%.
• Setup and Provisioning Overhead. An enormous amount of time is wasted on repetitive system setup, provisioning and configuration tasks – sometimes more than 50% of the total time expended in an application development and test cycle.
• Costly System Failures. Difficulties reproducing and correcting software defects discovered in remote development facilities, or by outsourcing partners, are leading to serious system failures in production, when the cost to repair can be over 470x higher than if resolved earlier in the AD process.

Virtual Lab Automation (VLA) substantially mitigates these problems – reducing server-to-staff ratios by over 75%; slashing the percentage of cycle time spent configuring systems from 50% or greater, to less than 5%; and ensuring software defects can be rapidly reproduced and resolved early in the cycle. The result is higher quality software, built faster and with lower server- and data center-related capital and operating costs.

A Virtual Lab Automation System automates the setup and teardown of complex, multi-machine software configurations on a centralized pool of servers shared by the application development and QA teams in an enterprise. These operations are performed in a self-service manner by developers and QA engineers, relieving the tedious provisioning burden often shouldered by the IT organization.

Virtual Lab Automation System

Enabled by the emergence of reliable virtual machine technology from vendors such as VMware and Microsoft, a Virtual Lab Automation System allows application developers and QA professionals to suspend, then capture to a shared storage library, the complete state of a “complex configuration”– a collection of running, interdependent software systems that usually span multiple servers. Over time, an organization builds up its library, including test scenarios, configurations exhibiting software defects, historical build archives, replicated production environments and customer configurations.

When a configuration in the library is later needed for development or test purposes, a VLA System can instantly deploy the entire configuration to the best available resources in a pool of shared servers, exactly as captured – running and ready for use.

VLA-managed capture and restore operations literally take seconds. What would normally be a painstaking, multi-hour or multi-day IT provisioning exercise (gathering machines, installing operating systems, installing and configuring applications, establishing inter-machine connections, booting) is now a simple self-service provisioning task, accomplished with a click of the mouse.

The following diagram highlights the components of a Virtual Lab Automation System.

Virtualized Server Pool. The most visible component of a VLA System is the central collection of virtualized servers and other supporting systems on which configurations are deployed.

VLA Server and Image Storage Library. The VLA server directs the capture, storage, movement, management and restoration of multi-machine configurations. Each machine image in a configuration contains the complete suspended state of machine CPUs, memory and disks.

VLA System Users. Application developers and QA engineers request the deployment and capture of multi-machine configurations through either user or programmatic (API) interfaces. API interfaces make it possible to interact with a VLA server directly from test scripts and test management systems.

Virtual Lab Automation System Benefits

A Virtual Lab Automation System delivers measurable value to the application development teams within an enterprise and to the IT organizations that support them.