Engineering and Management of Data Centers

Many corporations use central data centers to avoid building and maintaining a separate data center and are finding and adopting new ways to make the data centers more energy efficient. This chapter provides a relevant summary of key concepts for understanding the complexity of data centers and the types and value of data centers. Its purpose is to provide information and understanding needed to manage massive data generated every day in the era of internet technology.

The main international Information Technology Service Management (ITSM) process frameworks such as ITIL V3 and the ISO/IEC 20000-4, includes a service design process as part of their mandatory set of processes. Nevertheless, even with such availability of processes, their used nomenclature, their phase-activity structure, and their granularity level used for their descriptions, are non-standardized. Consequently ITSM academics are faced with a useful but disparate and diffused literature, and ITSM professionals lack suitable step-by-step service design methodologies. In this chapter, we present ITSDM (Information Technology Service Design Methodology), which is elaborated on best practices suggested in ISO/IEC 20000-4 and ITIL v3 design processes with the aim to provide a step-by-step guideline with phases, activities, tasks, roles, controls and input-output artifacts. We illustrate its utilization with a real IT service (Cloud Storage service) that is ready operational in a German Higher Education Institution. We found that ITSDM provided qualitative benefits such as: ease of use, usefulness, compatibility and value. However, given the novelty of this methodology, we encourage for further empirical studies for providing more definitive results.

A well designed and integrated database used to present risk management information by using a dashboard interface supported by real time risk management data makes it easy for risk managers to reach a full understanding of the surrounding threats and allows them to find the proper and right controls to mitigate them. The chapter presents a case study for a statistics data center that shows that the calculation of total risk at the organization level is possible by using the proposed risk database that supports decision makers when threats hit the organization. The chapter also shows that presenting the risk level on a dashboard viewer makes risk level clearer for a decision maker in a statistics data center and assists in the creation of a tool to follow-up risk management since the time a threat hits till the time of its mitigation.

In this chapter, we discuss the issues regarding risk related to data center planning. Organizations are dependent on computing services, which may become unavailable due to the manifestation of threats. These threats include natural disasters, man-made disasters and accidents. Prudent risk management is required to provide for continuation and recovery of operations in the event of a disruption. Risk assessment necessitates an in depth knowledge of the organization and a thorough analysis of the potential events which may have a negative impact on the data center and the associated computing services. Strategy development is discussed as well as preparedness, mitigation, exercises, response and recovery.

From interviews conducted in KIO Networks (Mexican Data Center) to individuals leading the Human Talent Management and Global Operations Management areas, this chapter presents a set of the best practices at KIO Networks, with a particular emphasis on the value of people and general recommendations on Data Center management. Regarding people, the recruitment, selection and training process is described, as well as the importance of the organization’s personnel in developing their competencies that allow them to improve their performance in the mission critical projects they take part in. Regarding processes, KIO considers that having the best human talent is as important as applying the best process certifications. KIO has also found it is important to learn how to recognize, implement and maintain those processes that generate value for the company, its customers, and the organization’s human capital.

Network-as-a-Service (NaaS) is one of new and promising cloud service models, through which the network infrastructure is offered to cloud customers as a service. Quality of Service (QoS) criteria play a significant role in the Service Level Agreement (SLA) for pricing and evaluation. This chapter describes NaaS service model and focuses on QoS criteria and issues in NaaS as well as for networking in general. A description about Software Defined Networking (SDN) has been presented, this promising network paradigm, which is considered one of programmable network designs, that depends on decoupling the control and data planes in order to achieve easier manageable networks. Finally, a deep research about applying a QoS policy in cloud and normal networks using SDN has been conducted, that shows the main categories for achieving the required QoS level. Their advantages and disadvantages have been mentioned, which in turn helps in choosing the best model according to multiple factors: the current network hardware, the traffic data type, the budget, etc.

Balancing costs and quality of offered IT service is a challenging task for data center providers. In the case of availability, fault tolerance can be applied by introducing redundancy mechanisms into the service design. Redundancy allocation problems can be defined as combinatorial optimization problems to identify cost-effective redundancy configurations in which availability objectives are met. However, these approaches should be flexible to trade-off effort and benefit in a specific scenario. Therefore, a redundancy allocation problem is proposed in this chapter that is capable of modeling the specific characteristics of the IT system to be analyzed. In order to identify suitable design configurations, a generic Petri net simulation model is combined with a genetic algorithm. By defining the solution algorithm adaptively to the complexity of the considered problem definition, users are able to reduce modeling as well as computational effort. The suitability of the approach is demonstrated in the use-case of an international application service provider.

Designing a modern data center is not only an engineer’s challenge, but also always a try to predict the future. A data center must provide resources to operate the current state of applications, but it must also provide additional potential for upcoming tasks and requirements. Applications, their demands and the technology of servers change rapidly. However, data center infrastructure, which is costly to acquire, has a much slower cycle of renewal. Hence, these devices, mainly power supply and cooling, must be able to fulfil future requirements. But overestimating demands and provisioning these devices by unrealistic loads may lead to serious energy wasting. During the last years several power saving techniques have evolved that allow data centers to save energy and costs during operation. However, these techniques may be significantly enhanced by considering them during the data center design process and also when hardware device updates are planned. To combine the power saving techniques with an intelligent design method, the following questions need to be answered first: Which energetic interdependencies exist in the data center in terms of electrical and thermal energy? How can these be controlled, regulated and systematically exploited to save energy and still preserve future potential? Therefore, this chapter will provide some basic information about data center energy flow and interdependencies. Based on this, an improved data center design method is presented that does not deviate too much from existing approaches, but enables significant power savings by following only a few new design choices. It is analyzed in what data center usage concepts the proposed energetic improvements may be used and how they can be exploited during operation. A special focus is placed on the combination of server consolidation strategies with improvements of the cooling infrastructure. But these can only be fully exploited if the described design methods and improvements are implemented.

Lately power grids have been subject to one of their major evolutions since their design and conception. The traditional structure of electricity being generated by a small number of huge and centralized power plants is being defied by the increasing penetration of renewable energy sources. The major drawback of such sources is their intermittent behavior rendering power generation planning even more cumbersome. This problem can be alleviated through the implementation of intelligent energy management systems (EMS) whose main objective is to exploit demand-side flexibilities for the purpose of better supply-side management and planning. Data centers, on one hand due to their significant power (in the order of up to 200 MW) as well as energy demand, and on the other hand thanks to their highly automated ICT infrastructure providing flexibilities without human interventions, have been shown to be excellent candidates for participation to such EMS. To this end, in this chapter we study such energy management systems by considering the use case of data centers both from local as well as coordinated management perspectives. For each considered perspective we describe thoroughly the concept as well as give its corresponding architectural building blocks. Furthermore, we specify the mechanisms and strategies that can be used for the case of data centers in exploiting demand-side flexibilities.

Nowdays the importance of the term DevOps (Developer Operations) has been increasing around the world, and Mexico is not the exception. This chapter describes the DevOps approach established in a Data Center of Mexico of a large Mexican governmental organization. The DevOps approach proposed aims to produce a seamless bridge and path between the software development teams and the release and deployment teams at Data Centers for developing and releasing software products. The approach was developed taking into account the process, the people and the technology. Therefore, the foundations, phases, activities, roles and artifacts are described. Besides, the chapter also reports benefits and challenges found in two illustrative real cases implemented in a large Mexican governmental organization data centers. Finally, the main recommendations and cautions in the implementation of a DevOps approach are reported.

In this chapter we firstly review the state of the art of the data center networks (DCNs) topological structures and explain the challenges in this field, which motivates our work. Then we propose a method for evaluating the topological metrics related to network robustness and node centrality so as to identify the most critical nodes and links in DCNs, as well as measuring the overall DCN performance such as throughput, latency, packet drop ratio according to the various faults occurred in the network. Moreover, we have identified the energy consumption behaviours according to the change of DCN’s internal structure. Our simulation studies showed that the DCN topology and traffic load have significant impact on its overall energy consumption and also on other network-related performance aspects.

The digital transformation of the economy, as well as society in general, is in large part based on data centers’ capabilities to process, store, and transport data. As the range of application requirements, the volume of data and system complexities, as well as security requirements increase, appropriate application system landscapes to support the tasks performed in a data center become crucial to success. Within this chapter we provide an overview on application software used to manage IT service production in data centers. With the advent of cloud computing, data centers increasingly face outside competition on all layers of the IT stack. Cloud service providers, such as Software as a Service companies, may even operate their own data centers. Furthermore, the people who use the investigated application software often themselves undergo an evolution from pure administrative to engineering responsibilities. Lastly, progress in the domain of application software, in general, has influenced application software discussed here. These phenomena are examined in order to compare common practice against competitive advantage of using the software from the presented taxonomy. The presented research has implications for practitioners because it provides an overview on the essential application fields of cloud-ready data centers. Additionally, the chapter contributes to the knowledge base by providing a taxonomy that is derived from a previously conducted case study. The taxonomy may be used in further research, for instance, in application system landscape maturity models.