Designing Reconfigurable Interconnection Network of Heterogeneous Chiplets Using Kalman Filter

TL;DR

This paper presents a reconfigurable interconnection network for heterogeneous chiplets that uses Kalman Filter-based predictions to dynamically allocate bandwidth, improving performance and fairness in resource sharing.

Contribution

It introduces a novel Kalman Filter-based approach for real-time, cost-efficient resource management in reconfigurable chiplet interconnection networks.

Findings

Improved system performance through dynamic bandwidth allocation.

Fair resource distribution among heterogeneous applications.

Low-cost and low-complexity network reconfiguration.

Abstract

Heterogeneous chiplets have been proposed for accelerating high-performance computing tasks. Integrated inside one package, CPU and GPU chiplets can share a common interconnection network that can be implemented through the interposer. However, CPU and GPU applications have very different traffic patterns in general. Without effective management of the network resource, some chiplets can suffer significant performance degradation because the network bandwidth is taken away by communication-intensive applications. Therefore, techniques need to be developed to effectively manage the shared network resources. In a chiplet-based system, resource management needs to not only react in real-time but also be cost-efficient. In this work, we propose a reconfigurable network architecture, leveraging Kalman Filter to make accurate predictions on network resources needed by the applications and then adaptively change the resource allocation. Using our design, the network bandwidth can be fairly allocated to avoid starvation or performance degradation. Our evaluation results show that the proposed reconfigurable interconnection network can dynamically react to the changes in traffic demand of the chiplets and improve the system performance with low cost and design complexity.