A Container Scheduler Accelerated by GPU and Based on Multicriterio Method

• Leonardo R. Rodrigues Universidade do Estado de Santa Catarina
• Marcelo Pasin Université de Neuchâtel
• Omir C. Alves Jr. Universidade do Estado de Santa Catarina
• Mauricio A. Pillon Universidade do Estado de Santa Catarina
• Charles C. Miers Universidade do Estado de Santa Catarina
• Guilherme P. Koslovski Universidade do Estado de Santa Catarina

DOI: https://doi.org/10.5753/sbrc.2019.7383

Abstract

Containers have been recently adopted as support for fast provisioning of distributed systems. They can be used to implement microservices, dataflow processing, edge computing, and other complex systems. However, due to the settings heterogeneity of requests and dimensionality of the hosting DCs, container scheduling is an NP-Hard problem. An efficient path to ease the scheduler complexity is to use the of high-performance parallel processing. In this context, we present present EMULAG: a GPU-accelerated multi-criteria scheduler. The schedulers objective function represents the providers perspective, aiming at data center (DC) consolidation. We present an experimental analysis revealing our solution is scalable and presents higher results than those foundin the literature, but with lower processing time.

Keywords: Container Scheduler, GPU, Parallel Computing, Multicriteria Methods

References

Al-Fares, M., Loukissas, A., and Vahdat, A. (2008). A scalable, commodity data center network architecture. SIGCOMM Comput. Commun. Rev., 38(4):63–74.

Assuncao, M. D. d., da Silva Veith, A., and Buyya, R. (2018). Distributed data stream processing and edge computing: A survey on resource elasticity and future directions. Journal of Network and Computer Applications, 103:1–17.

Cavalcanti, G. A. d. S., Obelheiro, R. R., and Koslovski, G. (2014). Optimal resource allocation for survivable virtual infrastructures. In 2014 10th Int. Conf. on the Design of Reliable Communication Networks (DRCN), pages 1–8.

Guerrero, C., Lera, I., and Juiz, C. (2018). Genetic algorithm for multi-objective optimization of container allocation in cloud architecture. Journal of Grid Computing.

Guo, Y. and Yao, W. (2018). A container scheduling strategy based on neighborhood division in micro service. In NOMS 2018-2018 IEEE/IFIP Network Operations and Management Symp., pages 1–6. IEEE.

Havet, A., Schiavoni, V., Felber, P., Colmant, M., Rouvoy, R., and Fetzer, C. (2017). GENPACK: A generational scheduler for cloud data centers. In 2017 IEEE Int. Conf. on Cloud Engineering (IC2E), pages 95–104.

Hindman, B., Konwinski, A., Zaharia, M., Ghodsi, A., Joseph, A. D., Katz, R. H., Shenker, S., and Stoica, I. (2011). Mesos: A platform for fine-grained resource sharing in the data center. In NSDI, volume 11, pages 22–22.

Kaewkasi, C. and Chuenmuneewong, K. (2017). Improvement of container scheduling for docker using ant colony optimization. In Knowledge and Smart Technology (KST), 2017 9th Int. Conf. on, pages 254–259. IEEE.

Nesi, L. L., Pillon, M. A., de Assunc¸ ão, M. D., and Koslovski, G. P. (2018a). GPUaccelerated algorithms for allocating virtual infrastructure in cloud data centers. In 18th IEEE/ACM Int. Symp. on Cluster, Cloud and Grid Computing (CCGrid 2018).

Nesi, L. L., Pillon, M. A., de Assunc¸ ão, M. D., Miers, C. C., and Koslovski, G. P. (2018b). Tackling virtual infrastructure allocation in cloud data centers: a gpu-accelerated framework. In 14th Int. Conf. on Network and Service Management (CNSM 2018).

Reiss, C., Wilkes, J., and Hellerstein, J. L. (2011). Google cluster-usage traces: format + schema. Technical report, Google Inc., Mountain View, CA, USA. Revised 2012.03.20. Posted at http://code.google.com/p/ googleclusterdata/wiki/TraceVersion2.

Rodriguez, M. A. and Buyya, R. (2018). Container-based cluster orchestration systems: A taxonomy and future directions. Software: Practice and Experience.

Saaty, T. L. (2005). Making and validating complex decisions with the AHP/ANP. Journal of Systems Science and Systems Engineering, 14(1):1–36.

Schwarzkopf, M., Konwinski, A., Abd-El-Malek, M., and Wilkes, J. (2013). Omega: flexible, scalable schedulers for large compute clusters. In Proc. of the 8th ACM European Conf. on Computer Systems, pages 351–364. ACM.

Singh, A., Ong, J., Agarwal, A., Anderson, G., Armistead, A., Bannon, R., Boving, S., Desai, G., Felderman, B., Germano, P., et al. (2015). Jupiter rising: A decade of clos topologies and centralized control in google’s datacenter network. In ACM SIGCOMM Computer Communication Review, volume 45, pages 183–197. ACM.

Trihinas, D., Tryfonos, A., Dikaiakos, M. D., and Pallis, G. (2018). Devops as a service: Pushing the boundaries of microservice adoption. IEEE Internet Computing, 22(3).

Van Dongen, S. M. (2001). Graph clustering by flow simulation. PhD thesis, University of Utrecht, Utrecht, Holanda.

Vaucher, S., Pires, R., Felber, P., Pasin, M., Schiavoni, V., and Fetzer, C. (2018). SGXaware container orchestration for heterogeneous clusters. In 2018 IEEE 38th Int. Conf. on Distributed Computing Systems (ICDCS), pages 730–741.