Load-balancing for mesh-based applications on heterogeneous cluster computers

This paper discusses load-balancing issues when using heterogeneous cluster computers. There is a growing trend towards the use of commodity microprocessor clusters. Although today's microprocessors have reached a theoretical peak performance in the range of one GFLOPS/s, heterogeneous clusters of commodity processors are amongst the most challenging parallel systems to programme efficiently. We will outline an approach for optimising the performance of parallel mesh-based applications for heterogeneous cluster computers and present case studies with the GeoFEM code. The focus is on application cost monitoring and load balancing using the DRAMA library.

Modelling the Performance of Flexible Substrates for Lead-Free Applications

In this paper, the performance of flexible substrates for lead-free applications was studied using finite element method (FEM). Firstly, the thermal induced stress in the flex substrate during the lead free solder reflow process was predicted. The shear stress at the interface between the copper track and flex was plotted. This shear stress increases with the thickness of the copper track. Secondly, an ACF flip chip was taken as a typical lead-free application of the flex substrate. The reflow effect on the reliability of ACF interconnections was analyzed. Higher stress was identified along the interface between the conductive particle and the metallization, and the interfacial stress increases with the reflow peak temperature and the coefficient of thermal expansion (CTE) of the adhesive. The moisture effect on the reliability of ACF joints were studied using a macro-micro modeling technique, the predominantly tensile stress found at the interface between the conductive particle and metallization could reduce the contact area and even cause the electrical failure. Modeling results are consistent with the findings in the experimental work