Similar articles

Abstract: Traditional high performance computing (HPC) systems provide a standard preset environment to support scientific computation. However, HPC development needs to provide support for more and more diverse applications, such as artificial intelligence and big data. The standard preset environment can no longer meet these diverse requirements. If users still run these emerging applications on HPC systems, they need to manually maintain the specific dependencies (libraries, environment variables, and so on) of their applications. This increases the development and deployment burden for users. Moreover, the multi-user mode brings about privacy problems among users. Containers like Docker and Singularity can encapsulate the job's execution environment, but in a highly customized HPC system, cross-environment application deployment of Docker and Singularity is limited. The introduction of container images also imposes a maintenance burden on system administrators. Facing the above-mentioned problems, in this paper we propose a self-deployed execution environment (SDEE) for HPC. SDEE combines the advantages of traditional virtualization and modern containers. SDEE provides an isolated and customizable environment (similar to a virtual machine) to the user. The user is the root user in this environment. The user develops and debugs the application and deploys its special dependencies in this environment. Then the user can load the job to compute nodes directly through the traditional HPC job management system. The job and its dependencies are analyzed, packaged, deployed, and executed automatically. This process enables transparent and rapid job deployment, which not only reduces the burden on users, but also protects user privacy. Experiments show that the overhead introduced by SDEE is negligible and lower than those of both Docker and Singularity.

面向高性能计算的自部署运行环境

[1]Azginoglu N, Atasever MU, Aydin Z, et al., 2017. Open source slurm computer cluster system design and a sample application. Proc Int Conf on Computer Science and Engineering, p.403-406.

[2]Bailey DH, Harris T, Saphir W, et al., 1995. The NAS Parallel Benchmarks 2.0. Technical Report.

[3]Belkin M, Haas R, Arnold GW, et al., 2018. Container solutions for HPC systems: a case study of using shifter on blue waters. Proc Practice and Experience on Advanced Research Computing, p.1-8.

[4]Bernstein D, 2014. Containers and cloud: from LXC to Docker to Kubernetes. IEEE Cloud Comput, 1(3):81-84.

[5]Beserra D, Moreno ED, Endo PT, et al., 2015. Performance analysis of LXC for HPC environments. Proc 9^th Int Conf on Complex, Intelligent, and Software Intensive Systems, p.358-363.

[6]Biederman EW, Networx L, 2006. Multiple instances of the global linux namespaces. Proc Linux Symp, p.101-112.

[7]Boettiger C, 2015. An introduction to Docker for reproducible research. ACM SIGOPS Oper Syst Rev, 49(1):71-79.

[8]Casalicchio E, Perciballi V, 2017. Measuring Docker performance: what a mess!!! Proc 8^th ACM/SPEC Int Conf on Performance Engineering Companion, p.11-16.

[9]Che JH, Shi CC, Yu Y, et al., 2010. A synthetical performance evaluation of OpenVZ, Xen and KVM. Proc IEEE Asia-Pacific Services Computing Conf, p.587-594.

[10]Christer E, 2012. Simple Linux Utility for Resource Management. Platform LSF. Technical Report.

[11]Feng HH, Misra V, Rubenstein D, 2007. PBS: a unified priority-based scheduler. Proc ACM SIGMETRICS Int Conf on Measurement and Modeling of Computer Systems, p.203-214.

[12]Gantikow H, Klingberg S, Reich C, 2015. Container-based virtualization for HPC. Int Conf on Cloud Computing and Services Science, p.543-550.

[13]Georgiou Y, Hautreux M, 2013. Evaluating scalability and efficiency of the resource and job management system on large HPC clusters. Workshop on Job Scheduling Strategies for Parallel Processing, p.134-156.

[14]Gerhardt L, Bhimji W, Canon S, et al., 2017. Shifter: containers for HPC. J Phys Conf Ser, 898:082021.

[15]Godlove D, 2019. Singularity: simple, secure containers for compute-driven workloads. Proc Practice and Experience in Advanced Research Computing on Rise of the Machines (Learning), p.1-4.

[16]Hale JS, Li LZ, Richardson CN, et al., 2017. Containers for portable, productive, and performant scientific computing. Comput Sci Eng, 19(6):40-50.

[17]Herbein S, Dusia A, Landwehr A, et al., 2016. Resource management for running HPC applications in container clouds. Int Conf on High Performance Computing, p.261-278.

[18]Huang Z, Wu S, Jiang S, et al., 2019. FastBuild: Accelerating Docker image building for efficient development and deployment of container. Proc 35^th Symp on Mass Storage Systems and Technologies, p.28-37.

[19]Kopytov A, 2012. SysBench Manual. MySQL AB.

[20]Kovari A, Dukan P, 2012. KVM & OpenVZ virtualization based IaaS open source cloud virtualization platforms: OpenNode, Proxmox VE. Proc IEEE 10^th Jubilee Int Symp on Intelligent Systems and Informatics, p.335-339.

[21]Kurtzer GM, Sochat V, Bauer MW, 2017. Singularity: scientific containers for mobility of compute. PLOS ONE, 12(5):e0177459.

[22]Kwon S, Lee JH, 2020. DIVDS: Docker image vulnerability diagnostic system. IEEE Access, 8:42666-42673.

[23]Lingayat A, Badre RR, Kumar Gupta A, 2018. Performance evaluation for deploying Docker containers on baremetal and virtual machine. Proc 3^rd Int Conf on Communication and Electronics Systems, p.1019-1023.

[24]Manco F, Lupu C, Schmidt F, et al., 2017. My VM is lighter (and safer) than your container. Proc 26^th Symp on Operating Systems Principles, p.218-233.

[25]Merkel D, 2014. Docker: lightweight linux containers for consistent development and deployment. Linux J, 2014(239):2.

[26]Mizusawa N, Nakazima K, Yamaguchi S, 2017. Performance evaluation of file operations on OverlayFS. Proc 5^th Int Symp on Computing and Networking, p.597-599.

[27]Rosen R, 2013. Resource Management: Linux Kernel Namespaces and Cgroups. Huafix. Technical Report.

[28]Saha P, Beltre A, Uminski P, et al., 2018. Evaluation of Docker containers for scientific workloads in the cloud. Proc Practice and Experience on Advanced Research Computing, p.1-8.

[29]Wang B, Chen ZG, Xiao N, 2020. A survey of system scheduling for HPC and big data. Proc 4^th Int Conf on High Performance Compilation, Computing and Communications, p.178-183.

[30]Wang K, Zhou XB, Chen H, et al., 2014. Next generation job management systems for extreme-scale ensemble computing. Proc 23^rd Int Symp on High-Performance Parallel and Distributed Computing, p.111-114.

[31]Wright CP, Dave J, Gupta P, et al., 2006. Versatility and Unix semantics in namespace unification. ACM Trans Stor, 2(1):74-105.

[32]Xavier MG, Neves MV, Rossi FD, et al., 2013. Performance evaluation of container-based virtualization for high performance computing environments. Proc 21^st Euromicro Int Conf on Parallel, Distributed, and Network-Based Processing, p.233-240.