Filling the gap between education and industry: evidence-based methods for introducing undergraduate students to HPC

Abstract

Educational institutions provide in most cases basic theoretical background covering several computational science topics. However, High-Performance Computing (HPC) and Parallel and Distributed Computing (PDC) markets require specialized technical profiles. Even the most skilled students are often not prepared to face production HPC applications of thousands of lines nor complex computational frameworks from other disciplines nor heterogeneous multinode machines accessed by hundreds of users. In this paper, we offer an educational package for filling this gap. Leveraging the 4-years experience of the Student Cluster Competition, we present our educational journey together with the lessons learned and the outcomes of our methodology. We show how, in a time span of a semester and an affordable budget, a university can implement an educational package preparing pupils for starting competitive professional careers. Our findings also highlight that 78% of the students exposed to our methods remain within the HPC high-education, research or industry.

Improvement of the model by preprocessing big data of tapping temperature prediction industry

“…The second method is to use the mean value filling method, but in the mean value filling, the variance of data will decrease which also means the information will be dropped. The third method is also a common scheme for steel data, filling algorithm based on K-Means method [7]. The traditional K-mean filling method can be divided into the following steps.…”

“…For the problem of uneven distribution of steel samples, many studies have not dealt with this aspect or just carried out simple down-sampling. [7][8] These processing methods do not consider the internal logic among the data samples and do not meet the business requirements in actual production. If the depth of the inquiry is carried out, it is easily found that some valuable data has been deleted, and the information has been dropped.…”

Leveraging teaching on demand: Approaching HPC to undergrads

“…There are several previous works in which this fact is also highlighted and addressed [33,34]. The CS degree of UJI is four years long and includes, basic training (BT), compulsory (C), and optional (OP) subjects, as well as a final degree project (FDP), but the presence of HPC is marginal in the curriculum.…”

Performance study of HPC applications on an Arm-based cluster using a generic efficiency model

“…The Dibona cluster has been integrated by Bull/ATOS within the framework of the European project Mont-Blanc 3 [8]. It integrates 40 Arm-based compute nodes powered by two Marvell ThunderX2 CN9980 processors 1 Compute nodes are interconnected with a fat-tree network, implemented with Mellanox IB EDR-100 switches. A separated 1 GbE network is employed for the management of the cluster and a network file system (NFS).…”

“…In this paper we leverage a performance analysis and modeling method for an in depth study of four parallel applications on an HPC cluster powered by the same CPUs as the Astra supercomputer. For validating our tests we selected a set of complex HPC applications proposed as challenge within the Student Cluster Competition held at the ISC conference [1].…”