Software Defined Systems

The newly emerged software defined system (SDS) promises a new information system resource allocation and management way. The main concept of SDS is to separate the control plane from the data plane in various subsystems, e.g., sensing, communication, networking, and computation. By such means, various resources of the information system are virtualized and therefore can be managed in a more friendly and flexible manner. It is widely believed that SDS is able to lower the barrier for system and application innovation, and will become an inevitable trend towards the future generation of the information system. In this chapter, we first identify the emergence and then give an overview as well as the main concepts of SDS. Regarding that many enabling technologies are already available to realize the vision of SDS, we also present a brief summarization of the main enabling technologies for SDS and identify their key characteristics.

After a decade of extensive research on application-specific WSNs, the recent development of information and communication technologies makes it practical to realize SDSNs, which are able to adapt to various application requirements and to fully explore the resources of WSNs. A sensor node in SDSN is able to conduct multiple tasks with different sensing targets simultaneously. A given sensing task usually involves multiple sensors to achieve a certain quality-of-sensing, e.g., coverage ratio. It is significant to design an energy-efficient sensor scheduling and management strategy with guaranteed quality-of-sensing for all tasks. To this end, three issues shall be considered: (1) the subset of sensor nodes that shall be activated, i.e., sensor activation, (2) the task that each sensor node shall be assigned, i.e., task mapping, and (3) the sampling rate on a sensor for a target, i.e., sensing scheduling. In this chapter, they are jointly considered and formulated as a mixed-integer with quadratic constraints programming (MIQP) problem, which is then reformulated into a mixed-integer linear programming (MILP) formulation with low computation complexity via linearization. To deal with dynamic events such as sensor node participation and departure, during SDSN operations, an efficient online algorithm using local optimization is developed. Simulation results show that the proposed online algorithm approaches the globally optimized network energy efficiency with much lower rescheduling time and control overhead.

The fast development of mobile computing has raised ever-increasing diverse communication needs in wireless networks. To catch up with such needs, cloud-radio access networks (CRAN) is proposed to enable efficient radio resource sharing and management. By CRAN, it is possible to realize software defined access networks. At the same time, the massive deployment of radio access networks has caused huge energy consumption. Incorporating renewable green energy to lower the brown energy consumption also has become a widely concerned topic. In this chapter, we are motivated to investigate a green energy aware remote radio head (RRH) activation problem for coordinated multi-point (CoMP) communications in green energy powered CRAN, aiming at minimizing the network brown energy power consumption. The problem is first formulated into a non-convex optimization form. By analyzing the characteristics of the formulation, we further propose a heuristic algorithm based on an ordered selection method. Extensive simulation based experiment results show that the proposed green energy aware algorithm provides an effective way to reduce brown energy power consumption, well fitting the goal of developing green communications.

Software defined network (SDN) is a newly emerging network architecture with the core concept of separating the control plane and data plane. Centralized controller is introduced to manage and configure network equipments to realize flexible control of network traffic and provide a good platform for application-oriented network innovation. It thus be able to improve network resource utilization, simplify network management, reduce operating cost, and promote innovation and evolution. Routing is always a major concern in network management. When SDN devices (e.g., OpenFlow switches) are introduced, routing becomes different. The flow table is usually implemented in expensive and power-hungry Ternary Content Addressable Memory (TCAM), which is thus capacity-limited. How to optimize the network performance in the consideration of limited TCAM capacity is therefore significant. For example, multi-path routing (MPR) has been widely regarded as a promising method to promote the network performance. But MPR is at the expense of additional forwarding rule, imposing burden on the limited flow table. On the other hand, a logical centralized programmable controller manages the whole SDN by installing rules onto switches. It is widely regarded that one controller is restricted on both performance and scalability. To address these limitations, pioneering researchers advocate deploying multiple controllers in SDNs where each controller is in charge of a set of switches. This raises the switch-controller association problem on a switch shall be managed by which controller. In this chapter, we first investigate how to schedule MPR with joint consideration of forwarding rule placement. Then, we study a minimum cost switch-controller association (MC-SCA) problem on how to minimize the number of controllers needed in an SDN while guaranteeing the flow setup time.

Network function virtualization (NFV) emerges as a promising technology to increase the network flexibility, customizability, and efficiency by softwarizing traditional dedicated hardware based functions to virtualized network functions. The prosperous potential of edge cloud makes it an ideal platform to host the network functions. From the perspective of network service providers, an inevitable concern is how to reduce the overall cost for renting various resources from infrastructure providers. This first relates to the virtualized network function (VNF) placement, which shall not be discussed independently without the consideration of flow scheduling. In this chapter, we first discuss a static VNF placement problem with preknown service request rate. Then, we consider a more practical scenario and we alternatively investigate how to dynamically minimize the overall operational cost with joint consideration of packet scheduling, network function management, and resource allocation, without any prior knowledge.

In this book, we introduce the concept and architecture of software defined systems (SDS). The core enabling technologies, including software defined front-end devices (sensors, IoT devices), software defined access networks (e.g., cognitive radio, CRAN), software defined core networks (e.g., SDN, NFV), software defined storage and computing (e.g., microservice). These technologies jointly enable the programmers or system administrators escape from the heavy reliance on hardware. With SDS, all sensing, communication, computation, and storage resources can be managed in a software-defined way, much promoting the system flexibility and lowering the barrier for information system innovation. SDS brings not only new opportunities, but also new challenges, in resource management and optimization. With the advent of new information technologies, corresponding new resource management and optimization algorithms shall be designed to cater for the indistinguishable characteristics of these new technologies.