CloudNetSim++: A Toolkit for

Data Center Simulations in

OMNET++

ASAD W. MALIK

NUST SCHOOL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCE, PAKISTAN

Team Members

Kashif Bilal: North Dakota State University, Fargo, USA

Khurram Aziz: Comsats institute of information technology, PAK

Dzmitry Kliazovich: University of Luxembourg, Luxembourg

Nasir Ghani: University of south florida, Florida, USA

Samee U. Khan: North Dakota State University, Fargo, USA

Rajkumar Buyya: University of Melbourne, Australia

Outline

Introduction

Related Work

Motivation

CloudNetSim++ Features

CloudNetSim++ Architecture

Performance Evaluation

Conclusion

Introduction

Cloud computing services have become increasingly popular

“Market Tends” estimates that cloud-based SaaS will increase from US $ 13.4 billion in 2011 to $32.2 billion in 2016 *

Similarly, in IaaS and PaaS markets are estimed growth from $7.6 billion in 2011 to $ 35.5 billion in 2016 *

Require massive infrastructure to support this enormous growth

Large geographically distributed data centers requires considerable amount of energy

High power consumption generates heat and requires an accompanying cooling system that costs in a range of $2 to $5 million per year

* L. Columbus, “Cloud Computing and Enterprise Software Forecast Update, 2012,” Forbes, 8 Nov. 2012; www.forbes.com/sites/louiscolumbus/2012/11/08/cloud-computing-and-

enterprisesoftware-forecast-update-2012

Introduction

Failure to keep data center temperature within operational ranges drastically decreases hardware reliability

The techniques, Dynamic Voltage and Frequency Scaling (DVFS) and Dynamic Power Management (DPM) is widely adopted

Idle server may consume about 2/3 of the peak load*

Workload of data center fluctuates on the hourly basis

Average load account only 30% of data center resources**

This allow putting rest 70% into a sleep mode for most of the time

To achieve this, central coordination and energy-aware scheduling is required

*Chen G, He W, Liu J, Nath S, Rigas L, Xiao L, Zhao F (2008) Energy-aware server provisioning and load dispatching for connection-intensive internet services. In: The 5th USENIX

symposium on networked systems design and implementation, Berkeley, CA, USA

**Liu J, Zhao F, Liu X, He W (2009) Challenges Towards Elastic Power Management in Internet Data Centers. In: Proceedings of the 2nd international workshop on cyber-physical

systems (WCPS), in conjunction with ICDCS 2009, Montreal, Quebec, Canada, June

Related Work

Simulator Available Language GUI Comm.

Model

Energy

Model

Simulation

Time

CloudSim Open Source Java No Limited Yes Second

NetworkCloudSim Open Source Java No Full No Second

iCanCloud Open Source C++ Yes Full No Second

DCSim+ Open Source Java No No No Minutes

GreenCloud Open Source C++, oTcl Limited Full Yes Minutes

Motivation

To build a comprehensive Cloud simulator that facilitate

Students

Researchers

Industry

CloudNetSim++: Features

Support Service Level Agreement (SLA)

Support various scheduling algorithms

Distributed data centers

Configurable number of data centers

Configurable number of racks and servers

Configurable physical link properties

Energy Module

Support multiple users

CloudNetSim++: High Level Architecture

OMNeT++

Multiple

Client

Data center-I Data center-II

Centralize Scheduler

INET

CloudNetSim++: Node Level Architecture

App Module

Energy Module

Queue Module

Communication

Module

Compute Node

Energy Module

Communication

Module

Router/Switches

CloudNetSim++

Energy Computation

Flexible data center model, compute energy utilization of following components

Servers

Data center architecture, router and switches

Power management, Dynamic Voltage Frequency Scaling (DVFS) technique

V2 ∗ F

The average power consumption is stated as below

P = PC + CPUf ∗ f

PC : power consumed not scale to frequency

CPUf ∗ f : represent frequency depended power consumption

CloudNetSim++

Power consumption of switches stated as:

𝑃𝑠𝑤𝑖𝑡𝑐ℎ = 𝑃𝑐ℎ𝑎𝑠𝑠𝑖𝑠 + 𝑛𝑙𝑖𝑛𝑒𝑐𝑎𝑟𝑑 . 𝑃𝑙𝑖𝑛𝑒𝑐𝑎𝑟𝑑 + 𝑖=0𝑅 𝑛𝑝𝑜𝑟𝑡,𝑟 . 𝑃𝑟

𝑃𝑐ℎ𝑎𝑠𝑠𝑖𝑠 : Power consumed by switch hardware

𝑃𝑙𝑖𝑛𝑒𝑐𝑎𝑟𝑑 : Power consumed by a line card

𝑃𝑟 : Power consumed by a port operating at rate r

CloudNetSim++: Graphical User Interface

CloudNetSim++

CloudNetSim++: Performance Evaluation

Used two different traffic scenarios

Many-to-one model

Many-to-many model

S.No Simulation Parameters

Parameters Value

1 Inter-Data Center (DC) topology Star/Mesh2 Intra-DC topology three-tier3 Inter-DC link 100-Gbps4 Data center to data center link (Bit Error Rate) 10−12

5 Core to aggregate link 10 Gbps6 Aggregate to access link 1 Gbps7 Access to servers link 1 Gbps8 Core to aggregate link (BER) 10−12

9 Aggregate to access link (BER) 10−12

10 Access link to computing servers (BER) 10−5

11 Packet size 1500 bytes

12 Core nodes 813 Aggregate nodes 1614 Access nodes 25615 Computing server 2200 - 9000

CloudNetSim++: Performance Evaluation

98

45

200

156DC-East(kWh)

DC-West(kWh)

DC-South(kWh)

DC-North(kWh)

4.086

9.21

16.218

70.633

Core Switch(kWh)

Aggregate Switch(kWh)

Access Switch(kWh)

Server(kWh)

CloudNetSim++: Performance Evaluation

CloudNetSim++: Performance Evaluation

CloudNetSim++: Performance Evaluation

CloudNetSim++: Available download

Available for download at

http://cloudnetsim.seecs.edu.pk/

Conclusion

Designed to facilitate students, researchers and industry requirement

Provide rich Graphical User Interface – GUI

Modular approach, new modules can easily be incorporated

Configurable architecture

Open source, available to download

CloudNetSim++

cloudnetsim.seecs.edu.pk

Thank You!

References

A. Vahdat, M. Fares, N, Farrington, R. N. Mysore, G. Porter, and S. Radhakrishnan, “Scale-Out Networking in the Data Center,” IEEE Micro. Vol. 30, no. 4, p. 29-41, 2010.

P. Mahadevan, P. Sharma, S. Banerjee, and P. Ranganathan, “Energy Aware Network Operations,” in INFOCOM Workshops, pp. 25-30, 2009.

J. Shuja, S. A. Madani, K. Bilal, K. Hayat, S. U. Khan, and S. Sarwar, “Energy-efficient data centers,” Computing, vol. 94 no. 12 pp. 973-994, 2012.

R. Beik, “Green Cloud Computing: An Energy-Aware Layer in Software Architecture,” in Spring Congress on Engineering and Technology (S-CET), pp. 1-4, 2012.

J. Moore, J.Chase, P. Ranganathan, and R.Sharma, “Making scheduling cool: temperature-aware workload placement in data centers,” in USENIX Annual Technical Conference pp. 61-75, 2005.

N. Rasmussen, “Calculating Total Cooling Requirements for Data Centers,” produced by Schneider Electrics Data Center Science Center, 2011, http://www.apcmedia.com/salestools/NRAN5TE6HE/NRAN-5TE6HE R3 EN.pdf. Accessed 31 August 2014.

References

K. Bilal, S. U. R. Malik, S. U. Khan, and A. Y. Zomaya, "Trends and Challenges in Cloud Data Centers," IEEE Cloud Computing Magazine, vol. 1, no. 1, pp. 10-20, 2014.

K. Bilal, S. U. Khan, and A. Y. Zomaya, "Green Data Center Networks: Challenges and Opportunities," in 11th IEEE International Conference on Frontiers of Information Technology (FIT), pp. 229-234. 2013.

R. Buyya, R. Ranjan, and R.N. Calheiros, “Modeling and simulation of scalable Cloud computing environments and the CloudSim toolkit: Challenges and opportunities,” in International Conference of High Performance Computing & Simulation, pp. 1-11, 2009.

X. Li, X. Jiang, K. Ye, and P. Huang, “DartCSim+: Enhanced CloudSim with the Power and Network Models Integrated,” in IEEE Sixth International Conference on Cloud Computing, pp.644- 651,2013.

S. K. Garg, and R. Buyya, “NetworkCloudSim: modelling parallel applications in cloud simulations.” in Fourth IEEE International Conference on Utility and Cloud Computing (UCC), pp. 105-113, 2011.

D. Kliazovich, P. Bouvry, Y. Audzevich, and S. U. Khan, “GreenCloud: A Packet-Level Simulator of Energy-Aware Cloud Computing Data Centers,” in IEEE Global Telecommunications Conference (GLOBECOM), pp. 1-5, 2010.