Radar: Reliable Resource Scheduling for Composable/Disaggregated Data Centers

Hardware disaggregation decouples resources (e.g., processors and memory) from monolithic servers, potentially improving service reliability. However, from another perspective, directly exposing resource modules to a shared network may adversely affect service reliability. In this article, we study a reliable resource allocation problem in disaggregated data centers, considering network impact and different disaggregation scales. We provide a mixed-integer linear programming formulation and a resource allocation framework named Radar for this problem. The numerical results demonstrate that the benefits of hardware disaggregation may be adversely affected by an imperfect network. It also shows that both the hardware backup and a proposed migration-based restoration can be applied to overcome this potential adverse effect.

Automated summary

Hardware disaggregation decouples resources from monolithic servers to potentially improve service reliability. However, directly exposing resource modules to a shared network may negatively impact service reliability. This article studies reliable resource allocation in disaggregated data centers, considering network impact and different disaggregation scales. The results demonstrate that an imperfect network can adversely affect the benefits of hardware disaggregation, but hardware backup and migration-based restoration can mitigate this effect.