Application-Arrival Rate Aware Distributed Run-Time Resource Management for Many-Core Computing Platforms

Abstract

Modern many-core computing platforms execute a diverse set of dynamic workloads in the presence of varying application arrival rates. This inflicts strict requirements on run-time management to efficiently allocate system resources. On the way towards kilocore processor architectures, centralized resource management approaches will most probably form a severe performance bottleneck, thus focus has been turned to the study of Distributed Run-Time Resource Management (DRTRM) schemes. In this article we examine the behaviour of a DRTRM of dynamic applications with malleable characteristics against stressing incoming application interval rate scenarios, using Intel SCC as the target many-core system. We show that resource allocation is highly affected by application input rate and propose an application-arrival aware DRTRM framework implementing an effective admission control strategy by carefully utilizing voltage and frequency scaling on parts of its resource allocation infrastructure. Through extensive experimental evaluation, we quantitatively analyse the behavior of the introduced DRTRM scheme and show that it achieves up to 44% performance gains while consuming 31% less energy, in comparison to a state-of-art DRTRM solution. In comparison to a centralized RTRM, the respective metric values rise up to 62% and 45% performance and energy gains, respectively.

The power impact of hardware and software actuators on self-adaptable many-core systems

“…Regarding software actuators, most of the works found in the literature consider Application Admission as a problem prior to the TM, which makes AA the least employed. For example, authors [16,[37][38][39][40][41] present frameworks that focus directly on deciding the best number of tasks for a given application before the TM. In Rahmani et al [30] , the application enters the system only if there are available processors, although applications can also be killed suddenly if the power overcomes the capping.…”