Tuning garbage collection for reducing memory system energy in an embedded java environment

Abstract

Java has been widely adopted as one of the software platforms for the seamless integration of diverse computing devices. Over the last year, there has been great momentum in adopting Java technology in devices such as cellphones, PDAs, and pagers where optimizing energy consumption is critical. Since, traditionally, the Java virtual machine (JVM), the cornerstone of Java technology, is tuned for performance, taking into account energy consumption requires reevaluation, and possibly redesign of the virtual machine. This motivates us to tune specific components of the virtual machine for a battery-operated architecture. As embedded JVMs are designed to run for long periods of time on limited-memory embedded systems, creating and managing Java objects is of critical importance. The garbage collector (GC) is an important part of the JVM responsible for the automatic reclamation of unused memory. This article shows that the GC is not only important for limited-memory systems but also for energy-constrained architectures.This article focuses on tuning the GC to reduce energy consumption in a multibanked memory architecture. Tuning the GC is important not because it consumes a sizeable portion of overall energy during execution, but because it influences the energy consumed in the memory during application execution. In particular, we present a GC-controlled leakage energy optimization technique that shuts off memory banks that do not hold live data. Using two different commercial GCs and a suite of thirteen mobile applications, we evaluate the effectiveness of the GC-controlled energy optimization technique and study its sensitivity to different parameters such as bank size, the garbage collection frequency, object allocation style, compaction style, and compaction frequency. We observe that the energy consumption of an embedded Java application can be significantly more if the GC parameters are not tuned appropriately. Further, we notice that the object allocation pattern and the number of memory banks available in the underlying architecture are limiting factors on how effectively GC parameters can be used to optimize the memory energy consumption.

Memory power optimization of Java-based embedded systems exploiting garbage collection information

“…The second approach performs an additional run-time code reallocation to improve the static assignment of Java methods to scratch-pad memories by the JVM, adjusting at run-time the assigned methods based on a continuous history monitoring of recent GC behavior. Our results show up to 45% leakage reduction with negligible performance impact, as well as 50% overall energy and 30% performance improvements for the GC with respect to classical cache-based JVM memory architectures [10,3,6].…”

“…Thus, they theoretically put less pressure in the processing power of the final embedded system and achieving good overall results (e.g., performance of memory hierarchy, L1 cache behavior, etc.). We demonstrate in this work that generational GCs achieve significantly better global results than more classical GCs used in the JVMs proposed for embedded systems [19,6,15].…”

“…Chen et al [6] focus on reducing the static energy consumption in a multi-banked main memory by tuning the collection frequency of a Mark&Sweep-based collector that shuts off memory banks that do not hold live data. Thus, the static leakage is decreased by turning off the unnecessary banks.…”

“…As a result, state-of-the-art GCs (e.g., generational GCs, incremental mark-and-sweep algorithms) have reduced their latency of response and the amount of time that the system needs to be stopped to compact the whole list of unused objects in Java-based designs. However, the increasing need for low-power systems limits very significantly the use of Java for new embedded devices since GCs are usually efficient enough in performance, but very costly in energy and power [6]. Thus, optimized (from the energy viewpoint) automatic dynamic memory reclamation mechanisms and methodologies to tune them have to be proposed for a complete integration of Java in the design of latest high-performance embedded systems, which include tight low-power constraints for portability purposes.…”

Energy Characterization of Garbage Collectors for Dynamic Applications on Embedded Systems

“…Chen et al [7] focus in reducing the static energy consumption in a multibanked main memory by tuning the collection frequency of a Mark&Sweep-based collector that shuts off memory banks that do not hold live data. The reduction of leakage approach is parallel to ours and can be used complementary.…”