Luận án: Tiết kiệm năng lượng mạng SDN trong điện toán đám mây - Trần Mạnh Nam

MINISTRY OF EDUCATION AND TRAINING HANOI UNIVERSITY OF SCIENCE AND TECHNOLOGY TRAN MANH NAM CÁC PHƯƠNG PHÁP TIẾT KIỆM NĂNG LƯỢNG SỬ DỤNG CÔNG NGHỆ MẠNG ĐIỀU KHIỂN BẰNG PHẦN MỀM TRONG MÔI TRƯỜNG ĐIỆN TOÁN ĐÁM MÂY SDN-BASED ENERGY-EFFICIENT NETWORKING IN CLOUD COMPUTING ENVIRONMENTS DOCTORAL THESIS OF TELECOMMUNICATIONS ENGINEERING HANOI - 2018 MINISTRY OF EDUCATION AND TRAINING HANOI UNIVERSITY OF SCIENCE AND TECHNOLOGY TRAN MANH NAM CÁC PHƯƠNG PHÁP TIẾT KIỆM NĂNG LƯỢNG SỬ DỤNG CÔNG NGHỆ MẠNG ĐIỀU KHIỂN BẰNG PHẦN MỀM TRONG MÔI TRƯỜNG ĐIỆN TOÁN ĐÁM MÂY SDN-BASED ENERGY-EFFICIENT NETWORKING IN CLOUD COMPUTING ENVIRONMENTS Specialization: Telecommunications Engineering Code No: 62520208 DOCTORAL THESIS OF TELECOMMUNICATIONS ENGINEERING Supervisor: Assoc. Nguyen Huu Thanh HANOI - 2018 PREFACE I hereby assure that the results presented in this dissertation are my work under the guidance of my supervisor. The data and results presented in the dissertation are completely honest and have not been disclosed in any previous works. The references have been fully cited and in accordance with the regulations.

Tôi xin cam đoan các kết quả trình bày trong luận án là công trình nghiên cứu của tôi dưới sự hướng dẫn của giáo viên hướng dẫn. Các số liệu, kết quả trình bày trong luận án là hoàn toàn trung thực và chưa được công bố trong bất kỳ công trình nào trước đây. Các kết quả sử dụng tham khảo đều đã được trích dẫn đầy đủ theo đúng quy định. Hà Nội, Ngày 19 tháng 01 năm 2018 Tác giả Trần Mạnh Nam ii ACKNOWLEDGEMENTS First and foremost, I would like to thank my advisor, Associate Prof.

Nguyen Huu Thanh, for providing an excellent researching atmosphere, for his valuable comments, constant support and motivation. His guidance helped me in all the time and also in writing this dissertation. I could not have thought of having a better advisor and mentor for my PhD. Moreover, I would like to thank Associate Prof.

Pham Ngoc Nam, Dr. Truong Thu Huong for their advices and feedbacks, also for many educational and inspiring discussions. My sincere gratitude goes to the members (present and former) of the Future Internet Lab, School of `Electronics and Telecommunications, Hanoi University of Science and Technology. Without their support and friendship it would have been difficult for me to complete my PhD studies.

Finally, I would like to express my deepest gratitude to my family. They are always supporting me and encouraging me with their best wishes, standing by me throughout my life. Hanoi, 19th Jan 2018 iii CONTENTS LIST OF FIGURES. viii LIST OF TABLES.

AN OVERVIEW OF ENERGY-EFFICIENT APPROACHES IN CLOUD COMPUTING ENVIRONMENTS .1 Cloud Computing Services and Infrastructures .2 Energy consumption problems .2 An Overview of Energy-Efficient Approaches.1 Energy consumption characteristics .2 Energy-Efficient Approaches' Classification .3 Software-defined Networking (SDN) technology .2 SDN Southbound API - OpenFlow Protocol .4 Difficulties on Network Energy Efficiency and Motivations .1 Proposing an energy-aware and flexible data center network that is based on the SDN technology.2 Proposing energy-efficient approaches in a network virtualization for cloud environments.3 Proposing an energy-aware data center virtualization for cloud environments. SDN-BASED ENERGY-AWARE DATA CENTER NETWORK 16 2.1 DCN technique and architecture .2 Power-Control System of a DC Network .1 Energy modeling of a network .2 The Diagram of the Power-Control System .3 Energy-Aware Routing based on Power Profile of Devices in Data Center Networks using SDN.1 Energy-Aware Routing and Topology Optimization Algorithm.4 Green Data Center using centralized Power-control of the Network and servers .1 Extended Power-Control System .3 Topology-aware VM migration algorithm .4 VM Migration cost and Power modeling of a Server. ENERGY-EFFICIENT NETWORK VIRTUALIZATION FOR CLOUD ENVIRONMENTS.1 Network Virtualization and Virtual Network Embedding .2 Constructing Energy-Aware SDN-based Network Virtualization System 51 3.3 Modeling and Problem Formulation .2 Objective and Constraints .3 Time-based Embedding Strategies .4 Energy-efficient VNE algorithms .1 Energy-cost Coefficient of Capacity .2 Virtual Node Mapping algorithms .3 Virtual Link Mapping (VLiM) Algorithm. AN ENERGY-AWARE DATA CENTER VIRTUALIZATION FOR CLOUD ENVIRONMENTS.1 Virtual DC Technologies .1 Virtual data center embedding .2 Virtual machine migration and server consolidation .1 Data Center Modeling .2 Energy Modeling of DC Components .3 Energy-Efficient Problem Formulation .4 A New Concept for VDC Embedding .1 Energy-aware VDC architecture.2 Energy-aware VDC embedding algorithm .3 Joint VDC Embedding and VM Migration Algorithms.

CONCLUSION AND FUTURE WORK .2 Future research directions. 93 LIST OF PUBLICATIONS. 96 v ABBREVIATIONS APCI Advanced Configuration & Power Interface APEX Capital expenditure ASIC Application specific integrated circuits BAU Business-as-usual BFS Breadth-first Search CAPEX Capital Expenditure DC Data center DCN Data center network D-ITG Distributed internet traffic generator EA-NV Energy-aware network virtualization EA-VDC Energy-aware Virtual Data Center ECO Eco sustainable FM Full migration FPGA Field programmable gate arrays GH GreenHead HEA-E Heuristic Energy-aware VDC Embedding HEE Heuristic energy-efficient IaaS Infrastructure-as-a-service ICT Information and communication technologies ISP Internet service provider MoA Migrate on arrival MST Minimum spanning tree NaaS Network-as-a-service NFV Network function virtualization NV Network virtualization OLD OpenDayLight OPEX Operating expenses PaaS Platform-as-a-service PCS Power-Control System PM Partial migration POD Optimized data centers PSnEP Power scaling and energy-profile-aware RMD-EE Reducing middle node energy efficiency SaaS Software-as-a-service SDSN Software-Defined Substrate Network SN SecondNet vi SNMP Simple network management protocol TCAM Ternary content-addressable memory VDC Virtual data center VDCE Virtual data center embedding VLiM Virtual link mapping VM Virtual Machine VmM Virtual machine mapping VNE Virtual network embedding VNoM Virtual node mapping VNR Virtual network requests vii LIST OF FIGURES Figure 1.1: Estimate of the global carbon footprint of ICT (including PCs, telcos’ networks and devices, printers and datacenters) [15].2: Energy consumption estimation for the European telcos’ network infrastructures in the”Business-As-Usual” (BAU) and in the Eco-sustainable (ECO) scenarios, and cumulative energy savings between the two scenarios [16].5: OpenFlow controller and switches.2: Three-tier DCN Architecture [45] .3: Fat-tree DCN Topology .4: Dcell DCN Architecture [53] .5: BCube DCN Architecture [54] .6: Fat-tree architecture with k = 4 .7: Diagram of the ElasticTree system [57] .8: Energy – Utilization relation of a network [58] .9: Power-control System of a Network .10: Fat-tree topology with Minimum Spanning Tree .11: Power Scaling Algorithm .12: Power Scaling and Energy-Profile-Aware - PSnEP algorithm (Proposed Algorithm 1). The flowchart describes the process between Edge and Aggregation switches .13: use-case with PSnEP algorithm in a DCN .14: PSnEP vs Power scaling (PS) with k=6 Fat-tree, mix scenario .15: Energy-saving level ratio of the PSnEP algorithm to the PS algorithm in different sizes .16: Extended Power-Control system (Ext-PCS) .18: First-fit Migration [67] Algorithm .19: Topology-Aware Placement Algorithm .20: K=8, comparison with full mesh scenario .21: K=16, comparison with full mesh scenario .22: K=8, comparison with Honeyguide .23: K=16, comparison with Honeyguide .1: FlowVisor – Hypervisor-like Network Layer [71] .2: Example of a virtual network on top of a physical network .3: Energy-Aware Network Virtualization system’s Diagram .4: Online VNE mapping method .5: Online using Time Window method.6: The GUI of an Energy-aware network virtualization platform.8: AR – Online using Time Windows .9: Percentage of Power Consumption to Full State in Online Strategy .10 Percentage of Power Consumption to Full State in OuTW Strategy .11: Comparison of comsumed energy between Online and OuTW strategies .12: Comparison of acceptance ratio between Online and OuTW strategies .1: Traditional cloud service provider vs NaaS .2: Embedding virtual data center requests on a physical data center .3: Virtual data center embedding - Static mapping; .4: Virtual data center embedding - Dynamic mapping .5: Energy proportional property of energy-aware data centers .6: Energy-Aware VDC Architecture .7: VDC Embedding Flowchart .8: Flowchart of Partial Migration (PM) .9: Migration on Arrival .10: Fluctuation of system utilization (SecondNet).11: DC Utilization per Load .12: Acceptance Ratio per VM .13: Acceptance Ratio per VDC.14: Total power consumption of the physical DC .15: Average consumed power per serving VDC.16: Number of migrations for different strategies .17: Comparison of embedding - migration strategies .18: Different embedding-magrition strategies: (a) GreenHead, (b) SecondNet, (c) Partial Migration, (d) Migration on Arrival, (e) Full Migration.

91 LIST OF TABLES Table 1.1: The Internet’s users in the world [1] .2: Estimated power consumption sources in a generic platform of IP router .3: Classification of energy-efficient approaches of the future Internet [4] .1: Power Summary For A 48-Port Pronto 3240 .2: Energy consumption of NetFPGA-Based OpenFlow Switch .3: Energy-saving ratio of PSnEP to Power scaling algorithm in different topology’s sizes.5: Power profile of server Dell PowerEdge R710.1: Virtual Network Embedding Terminology .2: Acceptance ratio and power consumption of the system under different window size in OuTW.1: Standard deviation of system utilization. Overview of Network Energy Efficiency in Cloud Computing Environments The advances in Cloud Computing services as well as Information and Communication Technologies (ICT) in the last decades have massively influenced economy and societies around the world. The Internet infrastructure and services are growing day by day and play a considerable role in all aspects including business, education as well as entertainment. In the last four years, the percentage of people using Internet witnesses an annual growth of 3.5%, from 39% world population’s percentage in Dec-2013 to 51.

To support the demand of cloud network infrastructure and Internet services in the rapid growth of users, it is necessary for the Internet providers to have a large number of devices, complex design and architecture that have the capacity to perform increasingly number of operations for a scalability. Consequently, many huge cloud infrastructures have been employed by Telcos, Internet Service Providers (ISPs) and enterprises for the exploded demand of various applications and data cloud-services such as YouTube, Dropbox, e-learning, cloud office etc. To meet the requirements of these booming services all around the world, cloud network infrastructures have been built up in a very large scale, even geographically distributed data centers with a huge number of network devices and servers. In addition, the maintenance of the systems with high availability and reliability level requires a notable redundancy of devices such as routers, switches, links etc.

As a result, having such a large infrastructure consumes a huge volume of energy, which leads to consequent environmental and economic issues: - Environmentally, the amount of energy consumption and carbon footprint of the ITC-sector is remarkable. The manufacture of ICT equipment is estimated its use and disposal account for 2% of global CO2 emissions, which is equivalent to the contributions from the aviation industry [2]. The networking devices and components estimate around 37% of the total ICT carbon emission [3]; - Economically, the huge consumed power leads to the costs sustained by the providers/operators to keep the network up and running at the desired service level and their need to counterbalance ever-increasing cost of energy. Although network energy efficiency has recently attracted much attention from communities [4], there are still many issues in realization of the energy-efficient network including inflexibility and the lack of an energy-aware network.

The main difficulties of the network energy efficiency as well as its research motivations are shortly described as follows: - Inflexible network: first, one important point the network in cloud data centers (DC) nowadays is the inflexibility issue. For changing the processing algorithm and the control plane of a network, its administrators should carefully re-design, 1 re-configure and migrate the network for a long time. In many cases, there is a technical challenge for an administrator to apply new approaches and evaluate their efficiency. Consequently, the flexible and programmable network is strictly necessary.

Secondly, there are difficulties in evaluating the energy-saving levels of new energy-efficient approaches in a network due to the lack of the centralized power-control system. This system allows administrators and developers to monitor, control and managing the working states as well as power consumption of all network devices in real-time. - Energy-aware networking for virtualization technologies in cloud environments: cloud computing has emerged in the last few years as a promising paradigm that facilitates such new service models as Infrastructure-as-a-Service (IaaS), Storage- as-a-Service (SaaS), Platform-as-a-Service (PaaS), Network-as-a-Service (NaaS).