On scheduling garbage collector in dynamic real-time systems with statistical timing assurances

Abstract

We consider garbage collection (GC) in dynamic real-time systems. We consider the time-based GC approach of running the collector as a separate, concurrent thread, and focus on real-time scheduling to obtain assurances on mutator timing behavior, while ensuring that memory is never exhausted. We present a scheduling algorithm called GCUA. The algorithm considers mutator activities that are subject to time/utility function time constraints, variable execution time demands, the unimodal arbitrary arrival model that allows a strong adversary, and resource overloads. We establish several properties of GCUA including probabilistically-satisfied utility lower bounds for each mutator activity, a lower bound on the system-wide total accrued utility, bounded sensitivity for the assurances to variations in mutator execution time demand estimates, and no memory exhaustion at all times. Our simulation experiments validate our analytical results and confirm the algorithm's effectiveness and superiority.

Load Balancing of Java Applications by Forecasting Garbage Collections

“…To assess the performance benefits that can be achieved by adapting the load balancing based on the MaGC forecast information, we modified the well-known round robin load balancing algorithm 3 . Our proposed algorithm is presented in Algorithm 2.…”

“…Multiple research works have proposed new GC algorithms [3], [4], [10], [12] that have smaller performance impacts on the applications. Even though all these works have helped to reduce the impact of the MaGC, GC remains a concern due to the different factors that can affect its performance.…”

Overview of Real-Time Java Computing

“…Schoeberl and Vitek [54] presented the algorithm to compute the GC quota and interval in their study. Cho et al [55] used Yu Sun and Wei Zhang statistical tools to guarantee the algorithm's effectiveness and superiority.…”

On Best-Effort Utility Accrual Real-Time Scheduling on Multiprocessors

“…These cover a wide range of problem spaces: from processor scheduling [55,31,22,50,51,52,27]; to memory management and garbage collection [28,23,34]; to non-blocking synchronization [22,24,26]; to energy management [72,73,7,74]; to packet scheduling [69,70]; and to network routing [19,20,62]; They also cover step TUFs [31]; to non-step TUFs [22,24,26,69,72,19,62]; to abitrarilyshaped TUFs [50].…”

“…This implies that we have detected a circular dependency and hence a deadlock. To resolve the deadlock, we need to find the least local value density phase in σ J , abort the phase and remove it from the graph (lines [23][24][25][26].…”