This type of computing allows firms to off-load peak requests to remote processing centers.

The concept referred to in the question aligns well with on-demand computing. This type of computing enables businesses to scale their computational resources quickly by utilizing remote processing centers, effectively handling peak demand without the need for significant upfront investment in infrastructure. On-demand computing is often associated with cloud services, where firms can access additional resources such as computing power, storage, or applications as needed, allowing them to manage fluctuating workloads more efficiently.

In contrast, other options such as grid computing involve a more distributed approach, where various computing resources across different locations are harnessed to work on a single task. Edge computing, however, focuses on processing data closer to the source of generation to reduce latency, which does not directly address the issue of managing peak demands through remote processing. Autonomic computing pertains to systems that manage themselves based on high-level objectives, which is relevant to automation but does not specifically involve scaling resources to accommodate peak requests.

Thus, on-demand computing is the most suitable answer, as it directly addresses the need for flexibility and resource management in response to peak usage scenarios.