Empirical modeling and simulation of an heterogeneous Cloud computing environment

Abstract Cloud computing offers users a convenient scalable and flexible scenario for developing, running, and testing different applications in a pay-as-you-go model. However, the increasing demand of computational resources have resulted in more power consumption. In order to address this issue, the use of energy efficient resources such as FPGAs within a Cloud computing environment is nowadays a hot research topic, driven both from industry and academy. The efficient use of this type of resources can lead to increase the profit obtained by Cloud providers by serving more client requests while fulfilling their requirements with the minimum set of resources. In this paper, we survey an architectural framework and algorithms for managing heterogeneous resources, focusing on FPGAs. Based on this architecture, a simulation tool for studying the impact on performance and energy consumption of scaling the number of FPGAs in the system is presented. This tool is based on statistical models of processing time and energy consumption. The scalability study demonstrates that increasing the number of FPGAs can improve energy consumption up to 40%, while also admitting up to 35% more requests into the system.

[1]  Rubén S. Montero,et al.  IaaS Cloud Architecture: From Virtualized Datacenters to Federated Cloud Infrastructures , 2012, Computer.

[2]  Paul Chow,et al.  FPGAs in the Cloud: Booting Virtualized Hardware Accelerators with OpenStack , 2014, FCCM 2014.

[3]  María Blanca Caminero,et al.  Towards a Green, QoS-Enabled Heterogeneous Cloud Infrastructure , 2016, 2016 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW).

[4]  Rainer G. Spallek,et al.  RC3E: Provision and Management of Reconfigurable Hardware Accelerators in a Cloud Environment , 2015, ArXiv.

[5]  Ryan Kastner,et al.  RIFFA 2.0: A reusable integration framework for FPGA accelerators , 2013, 2013 23rd International Conference on Field programmable Logic and Applications.

[6]  Dimitrios Tzovaras,et al.  On the power consumption modeling for the simulation of Heterogeneous HPC clouds , 2017, CloudNG@EuroSys.

[7]  Karin Strauss,et al.  Accelerating Deep Convolutional Neural Networks Using Specialized Hardware , 2015 .

[8]  Rainer G. Spallek,et al.  Ready PCIe data streaming solutions for FPGAs , 2014, 2014 24th International Conference on Field Programmable Logic and Applications (FPL).

[9]  Andreas Herkersdorf,et al.  Enabling FPGAs in Hyperscale Data Centers , 2015, 2015 IEEE 12th Intl Conf on Ubiquitous Intelligence and Computing and 2015 IEEE 12th Intl Conf on Autonomic and Trusted Computing and 2015 IEEE 15th Intl Conf on Scalable Computing and Communications and Its Associated Workshops (UIC-ATC-ScalCom).

[10]  Rajkumar Buyya,et al.  CloudSim: a toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithms , 2011, Softw. Pract. Exp..

[11]  Wayne Luk,et al.  Elastic Management of Reconfigurable Accelerators , 2014, 2014 IEEE International Symposium on Parallel and Distributed Processing with Applications.

[12]  Yu Zhang,et al.  Enabling FPGAs in the cloud , 2014, Conf. Computing Frontiers.

[13]  N.H. Hamid,et al.  Performance analysis of FPGA based Sobel edge detection operator , 2008, 2008 International Conference on Electronic Design.

[14]  James R. Larus,et al.  A reconfigurable fabric for accelerating large-scale datacenter services , 2014, 2014 ACM/IEEE 41st International Symposium on Computer Architecture (ISCA).

[15]  Dimitrios Tzovaras,et al.  A review of cloud computing simulation platforms and related environments , 2017, CLOSER 2017.

[16]  Albert Y. Zomaya,et al.  Energy efficient utilization of resources in cloud computing systems , 2010, The Journal of Supercomputing.

[17]  María Blanca Caminero,et al.  An Open-Source Framework for Integrating Heterogeneous Resources in Private Clouds , 2014, CLOSER.

[18]  Jason Cong,et al.  Programming and Runtime Support to Blaze FPGA Accelerator Deployment at Datacenter Scale , 2016, SoCC.