This year, the IEEE Symposium on Computers and Communications (ISCC) took place in Barcelona Spain for the period of 30 June and 3 July. The LARCC coordinator Dr. Dalvan Gribler was there to present a research paper entitled Minimizing Communication Overheads in Container-based Clouds for HPC Applications, which was authored in the following sequence Anderson Maliszewski (UFRGS, Brazil); Adriano Vogel (PUCRS, Brazil); Dalvan Griebler (PUCRS/SETREM, Brazil); Eduardo Roloff (UFRGS, Brazil); Luiz G. Fernandes (PUCRS, Brazil); Philippe O. A. Navaux (UFRGS, Brazil). This work is mainly developed in LARCC when Anderson was working for the HiPerfCloud project supported by the company Abase Sistemas. More information about the paper is described below and the full paper will be soon available at the IEEE library.

In this work, we contributed with a solution that uses a network aggregation technique on dedicated server machines in order to improve performance, regarding latency and throughput. The results attested significant performance improvement for network-intensive applications in a cloud environment deployed at LARCC. The paper abstract is presented below.

Abstract: Although the industry has embraced the cloud computing model, there are still significant challenges to be addressed concerning the quality of cloud services. Network-intensive applications may not scale in the cloud due to the sharing of the network infrastructure. In the literature, performance evaluation studies are showing that the network tends to limit the scalability and performance of HPC applications. Therefore, we proposed the aggregation of Network Interface Cards (NICs) in a ready-to-use integration with the OpenNebula cloud manager using Linux containers. We perform a set of experiments using a network microbenchmark to get specific network performance metrics and NAS parallel benchmarks to analyze the performance impact on HPC applications. Our results highlight that the implementation of NIC aggregation improves network performance in terms of throughput and latency. Moreover, HPC applications have different patterns of behavior when using our approach, which depends on communication and the amount of data transferring. While network-intensive applications increased the performance up to 38%, other applications with aggregated NICs maintained the same performance or presented slightly worse performance.