Trends in On-chip Dynamic Resource Management

Abstract

The Complexity of emerging multi/many-core architectures and diversity of modern workloads demands coordinated dynamic resource management methods. We introduce a classification for these methods capturing the utilized resources and metrics. In this work, we use this classification to survey the key efforts in dynamic resource management.We first cover heuristic and optimization methods used to manage resources such as power, energy, temperature, Qualityof-Service (QoS) and reliability of the system. We then identify some of the machine learning based methods used in tuning architectural parameters in computer systems. In many cases, resource managers need to enforce design constraints during runtime with a certain level of guarantee. Hence, we also study the trend in deploying formal control theoretic approaches in order to achieve efficient and robust dynamic resource management.

Hierarchical adaptive Multi-objective resource management for many-core systems

“…According to Castilhos et al [27] , a cluster size with 16 (4 × 4) PEs represents a good trade-off between execution time optimization and resources reserved for management. Moazzemi et al [28] survey RMs for many-core systems and conclude that the Supervisory Control Theory (an approach similar to ours) is the most suitable approach regarding scalability.…”

“…At the beginning of the system execution, all clusters operate in performance mode . The algorithm may update the operation mode of the clusters when an application requests admission into the system (lines 2-29), or it finishes the execution (lines [28][29][30][31][32][33][34].…”