Just-in-Time Compilation-Inspired Methodology for Parallelization of Compute Intensive Java Code

Abstract

Compute intensive programs generally consume significant fraction of execution time in a small amount of repetitive code. Such repetitive code is commonly known as hotspot code. We observed that compute intensive hotspots often possess exploitable loop level parallelism. A JIT (Just-in-Time) compiler profiles a running program to identify its hotspots. Hotspots are then translated into native code, for efficient execution. Using similar approach, we propose a methodology to identify hotspots and exploit their parallelization potential on multicore systems. Proposed methodology selects and parallelizes each DOALL loop that is either contained in a hotspot method or calls a hotspot method. The methodology could be integrated in front-end of a JIT compiler to parallelize sequential code, just before native translation. However, compilation to native code is out of scope of this work. As a case study, we analyze eighteen JGF (Java Grande Forum) benchmarks to determine parallelization potential of hotspots. Eight benchmarks demonstrate a speedup of up to 7.6x on an 8-core system.

Authors and Affiliations

Ghulam Mustafa, Waqar Mahmood, Muhammad Usman Ghani

Keywords

Related Articles

Sensor-Fusion Based Navigation for Mobile Robot in Outdoor Environment

Autonomous navigation of the vehicles or robots is very challenging and useful task used by many scientists and researchers these days. By keeping this fact in mind, an algorithm for autonomous navigation of mobile robot...

Comparison of Gain Measurement Techniques for Characterization of Quantum Dot Lasers

This paper presents a comparative analysis of three gain measurement methods which are H&P (Hakki & Paoli), SC (Segmented-Contact) and IA (Integrated-Amplifier) for the gain characterization of 1300nm (O-band) InAs/GaAs...

Effect of Steel Fibers on Heat of Hydration and Mechanical Properties of Concrete Containing Fly Ash

This study investigated the effects of steel fibers on the fresh and hardened properties, and heat of hydration of concrete containing FA (Fly Ash). A total of 192 samples were cast comprising cubes, cylinders, and prism...

Kinetic Study of Cotton Stalk and Rice Husk Samples under an Inert and Oxy Combustion Atmospheres

Biomass materials such as CS (Cotton Stalks) and RH (Rice Husk) are a renewable source of energy. As biomass resources have potential to offer a uninterpted supply of biofuels through thermal decomposition processes. Nev...

An Improved Data Model for Uncertain Data

Uncertain data can be categorized as imprecise data and probabilistic data. In each of these categories, the uncertainty can be found at different granularity levels. Conventional data models are developed for the purpos...

Download PDF file
  • EP ID EP195535
  • DOI 10.22581/muet1982.1701.08
  • Views 138
  • Downloads 0

How To Cite

Ghulam Mustafa, Waqar Mahmood, Muhammad Usman Ghani (2017). Just-in-Time Compilation-Inspired Methodology for Parallelization of Compute Intensive Java Code. Mehran University Research Journal of Engineering and Technology, 36(1), 67-86. https://europub.co.uk/articles/-A-195535