The advent of petascale computing has introduced new challenges (e.g. heterogeneity, system failure) for programming scalable parallel applications. Increased complexity and dynamism in science and engineering applications of today have further exacerbated the situation. Addressing these challenges requires more emphasis on concepts that were previously of secondary importance, including migratability, adaptivity, and runtime system introspection. In this paper, we leverage our experience with these concepts to demonstrate their applicability and efficacy for real world applications. Using the CHARM++ parallel programming framework, we present details on how these concepts can lead to development of applications that scale irrespective of the rough landscape of supercomputing technology. Empirical evaluation presented in this paper spans many miniapplications and real applications executed on modern supercomputers including Blue Gene/Q, Cray XE6, and Stampede.

Task mapping on torus networks has traditionally focused on either reducing the maximum dilation or average number of hops per byte for messages in an application. These metrics make simplified assumptions about the cause of network congestion, and do not provide accurate correlation with execution time. Hence, these metrics cannot be used to reasonably predict or compare application performance for different mappings. In this paper, we attempt to model the performance of an application using communication data, such as the communication graph and network hardware counters. We use supervised learning algorithms, such as randomized decision trees, to correlate performance with prior and new metrics. We propose new hybrid metrics that provide high correlation with application performance, and may be useful for accurate performance prediction. For three different communication patterns and a production application, we demonstrate a very strong correlation between the proposed metrics and the execution time of these codes.

The genome of a red fox (Vulpes vulpes) was recently sequenced and assembled using next-generation sequencing (NGS). The assembly is of high quality, with 94X coverage and a scaffold N50 of 11.8 Mbp, but is split into 676,878 scaffolds, some of which are likely to contain assembly errors. Fragmentation and misassembly hinder accurate gene prediction and downstream analysis such as the identification of loci under selection. Therefore, assembly of the genome into chromosome-scale fragments was an important step towards developing this genomic model. Scaffolds from the assembly were aligned to the dog reference genome and compared to the alignment of an outgroup genome (cat) against the dog to identify syntenic sequences among species. The program Reference-Assisted Chromosome Assembly (RACA) then integrated the comparative alignment with the mapping of the raw sequencing reads generated during assembly against the fox scaffolds. The 128 sequence fragments RACA assembled were compared to the fox meiotic linkage map to guide the construction of 40 chromosomal fragments. This computational approach to assembly was facilitated by prior research in comparative mammalian genomics, and the continued improvement of the red fox genome can in turn offer insight into canid and carnivore chromosome evolution. This assembly is also necessary for advancing genetic research in foxes and other canids.

The development, production, and distribution of vaccines is imperative to saving lives, preventing illness, and reducing the economic and social burdens caused by the COVID-19 pandemic. Vaccines that use cutting-edge biotechnology have played an important role in mitigating the effects of SARS-CoV-2.

No abstract