The big data revolution has led to the emergence of many technologies for processing the enormous volume, high velocity and variety of data sources. The capabilities provided by high-performance computing (HPC) have had a substantial impact in supporting this revolution through the integration of heterogeneous hardware and crafting software and algorithms that exploit the parallelism of HPC. However, big data platforms, written in high-level programming languages, and their architectures differ from HPC languages and architectures. This misalignment has resulted in the underutilization of HPC resources and capabilities for big data problems. In this paper, we highlight and evaluate many of the techniques used in the HPC environment based on the most critical factors influencing performance and resource utilization: load balancing and data locality, job scheduling strategies, topology awareness, decomposition techniques and programming models. We recommend establishing a research roadmap that considers these influential factors in the field of resource and job management within HPC environments for big data computing.