Sustainable Cloud and Energy Services

The Internet of Things (IoT) and Cloud Computing both are developing technologies. Cloud Computing blows up to provide support to IoT by working as a sort of front-end and it is based on the concept of permitting users to do computing tasks using services delivered with internet. The cloud computing empower an appropriate, on-demand, and scalable network access to a shared pool of configurable computing resources. The cloud-based IoT architecture includes features of cloud-based IoT platform and its interaction with three main cloud computing models: IaaS (infrastructure as a service), Paas (platform as a service), and SaaS (software as a service). The cloud and IoT integration empowers new scenarios, for smart services and applications, as Sensing as a Service (SaaS), DataBase as a Service (DBaaS), Video Surveillance as a Service (VSaaS), and many more. Various live company products, research projects, and projects with freely available source code in various areas of Cloud Computing and IoT integration are Nimbits, ThingSpeak, Paraimpu, Device Cloud, Sensor Cloud. REpresentational State Transfer (REST) architectural style web services and Constrained Application Protocol (COAP), Message Queue Telemetry Transport (MQTT), web transfer protocols are used for communication for the IoT resource-constrained things. Networking protocols like IPv6 over Low power Wireless Personal Area Network (6LoWPAN) and IPv6 over Bluetooth Low Energy are used for constrained networks in IoT and cloud integration. The data link layer protocols for IoT devices like IEEE 802.15.4, IEEE 802.11ah, Z-Wave, WirelessHART, Bluetooth, Zigbee are used for short range communication for IoT things. The applications of integrated cloud and IoT include agriculture, video surveillance, healthcare, smart city, smart home and smart metering, etc. IoT and cloud integration involves several challenges and issues as standardization of machine to machine (M2M) communication and interoperability, power and energy efficiency of devices for data transmission and processing, big data generated by several devices, security and privacy, integration methodology, pricing and billing, network communications, storage, etc. In this chapter, the introduction of cloud and IoT, their integration architecture, integration applications, and challenges and issues involved are discussed.

An extensible, decentralized network of self-governing objects connected to a shared but variable power supply is a realistic problem domain for certain demand-management problems arising in the context of futuristic systems connected to smart power grids. An algorithmic framework of such a network is presented and discussed in this paper. Each object of the network has a power demand and a priority, and it is interconnected and able to exchange information with all other objects in the system. As each object shares information (its power demand and priority) with the other objects, the system as a whole exhibits self-governance, and is able to be managed without a centralized controller. Our model, which we term as decentralized power distribution (DPD) for such a network, also allows us to formulate algorithms for distributing available power among objects, along with analyses of correctness and performance.

Please check the hierarchy of the section headings and confirm if correct.

The smart grid changes the way how energy and information are exchanged and offers opportunities for incentive-based load balancing. For instance, customers may shift the time of energy consumption of household appliances in exchange for a cheaper energy tariff. This paves the path towards a full range of modular tariffs and dynamic pricing that incorporate the overall grid capacity as well as individual customer demands. This also allows customers to frequently switch within a variety of tariffs from different utility providers based on individual energy consumption and provision forecasts. For automated tariff decisions it is desirable to have a tool that assists in choosing the optimum tariff based on a prediction of individual energy need and production. However, the revelation of individual load patterns for smart grid applications poses severe privacy threats for customers as analyzed in depth in literature. Similarly, accurate and fine-grained regional load forecasts are sensitive business information of utility providers that are not supposed to be released publicly. This paper extends previous work in the domain of privacy-preserving load profile matching where load profiles from utility providers and load profile forecasts from customers are transformed in a distance-preserving embedding in order to find a matching tariff. The embeddings neither reveal individual contributions of customers nor those of utility providers. Prior work requires a dedicated entity that needs to be trustworthy at least to some extent for determining the matches. In this paper we propose an adaption of this protocol, where we use blockchains and smart contracts for this matching process, instead. Blockchains are gaining widespread adaption in the smart grid domain as a powerful tool for public commitments and accountable calculations. The use of a blockchain for this protocol makes the calculations for tariff matching public, while still maintaining the privacy through embeddings. Further, such decentralized and trust-free blockchains improve the existing solution in terms of verifiability, reliability, and transparency.

Energy consumption in buildings is responsible for a significant portion of the total energy use and carbon emissions in large cities. One of the main approaches to reduce energy consumption and its environmental impact is to convert buildings into smart buildings using computer, software, sensor, and network technologies. Using smart building energy management systems provides intelligent procedures to control buildings’ equipment such as HVAC (heating, ventilating, and air-conditioning) systems, home and office appliances, and lighting systems to reduce energy consumption while maintaining the required quality of living in all of the building’s spaces. This chapter discusses and reviews utilizing cloud computing to provide energy-related services to enhance the operations of smart buildings’ energy management systems. Cloud computing can provide many advantages for smart buildings’ energy management systems such as providing the required software models that implement different control and monitoring algorithms and providing optimization methods for more efficient energy consumption in smart buildings. This chapter will also discuss the benefits and issues of utilizing cloud computing services for enhancing energy consumption in smart buildings.

Virtual machine (VM) consolidation is one of the key mechanisms of designing an energy-efficient dynamic Cloud resource management system. It is based on the premise that migrating VMs into fewer number of Physical Machines (PMs) can achieve both optimization objectives, increasing the utilization of Cloud servers while concomitantly reducing the energy consumption of the Cloud data center. However, packing more VMs into a single server may lead to poor Quality of Service (QoS), since VMs share the underlying physical resources of the PM. To address this, VM Consolidation (VMC) algorithms are designed to dynamically select VMs for migration by considering the impact on QoS in addition to the above-mentioned optimization objectives. VMC is a NP-hard problem and hence, a wide range of heuristic and meta-heuristic VMC algorithms have been proposed that aim to achieve near-optimality. Since, VMC is highly popular research topic and plethora of researchers are presently working in this area, the related literature is extremely broad. Hence, it is a non-trivial research work to cover such extensive literature and find strong distinguishing aspects based on which VMC algorithms can be classified and critically compared, as it is missing in existing surveys. In this chapter, we have classified and critically reviewed VMC algorithms from multitude of viewpoints so that the readers can be truly benefitted. Finally, we have concluded with valuable future directions so that it would pave the way of fellow researchers to further contribute in this area.

The dynamic demand for resources from different users of Cloud makes resource management extremely important in design and decision-making processes in cloud computing environments. Providers of resources on Cloud offer heterogeneous resources such as compute units, memory and storage in Virtual Machine instances (VM). Large-scale data centers are essential to service the huge rise in demand for reliable high performance computational and storage services over cloud.

From real traces, resource requests of the users on cloud show that they are mostly overestimated and sometimes underestimated. Over multiple requests in each task, multiple jobs, cumulative unutilized resource amounts to considerable wasted resources for provider and unnecessary expenditure for users. Possible approaches and techniques that can ensure that resources are not overbooked and wasted are an urgent necessity.

With increase in computational and storage needs of business users who are moving their applications and data to cloud, scaling of cloud resources at the providers end, results in increased energy consumption and carbon dioxide emissions. This necessitates looking for possible energy saving approaches be adopted in data centers.

Mega data centers that house thousands of servers consume huge energy per hour at peak times and lead to increased operational costs. More important consideration is that, the huge power consumption hastens climate change due to carbon dioxide emissions and use of nonrenewable energy sources. Therefore, for environmental and financial reasons it is imperative to try to reduce unnecessary booking of resources when they are not actually used.

This problem can be overcome by finding suitable resource requirement prediction approach, such that, based on predicted required amount of resource, resource reservation will be done. Efficient resource requirement prediction approach will ensure that resources are efficiently utilized. Resource usage studied from existing workloads on cloud, more specifically the Google Compute Cloud usage data, shows that these do not have any specific pattern, trend, seasonality in the use of resources. Various researchers who have used resource prediction techniques such as time Series, exponential smoothing, neural networks, Bayes method etc., have shown results for only Web based distributed data, which is much different from actual cloud usage data. Cloud Usage data shows bursty non cyclic and non-seasonal behavior. This makes resource requirement prediction for Cloud Challenging.

The proposed prediction approach applies Enhanced Instance Based Learning (EIBL) approach. Two variations of EIBL—Distance weighted averaging (DWA) and Locally weighted Regression (LWR) have been used in experiments with Google Cluster Data. Google Cluster Data is a trace of the data used by various users of Google Cloud over a period of 29 day. The proposed prediction approaches are tested using Google cluster trace data.

The increasing demand for services offered by cloud providers results in a large amount of electricity usage by their data center sites and a high impact on the environment. This has motivated many cloud providers to move towards using on-site renewable energy sources to partially power their data centers using sustainable sources. This way, they can reduce their reliance on brown electricity delivered by off-site providers, which is typically drawn from polluting sources. However, most sources of renewable energy are intermittent and their availability changes over time. Therefore, having short-term prediction helps the cloud provider to make informed decisions and migrate the virtual machines (VMs) between data center sites in the absence of the renewable energy. In this chapter, we propose a short-term prediction model using Gaussian mixture model (GMM). The model uses the previously observed energy levels to train itself and predict the energy level for many-steps ahead into the future. We analyzed the accuracy of the proposed prediction model using real meteorological data. The experiment results show that the GMM model can predict up to 15 min ahead into the future with nearly 98% accuracy around ± 10% of the actual values. This helps the cloud provider to perform online VM migration with performance close to the optimal offline algorithm, which has the full knowledge of renewable energy level in the system. Moreover, the accuracy of the model has been verified using the workload data from Amazon biggest region in US East (N. Virginia). However, due to the confidentiality of that data set, we only rely on the results of the carried experiments using real meteorological renewable energy traces.

Due to high investment costs of renewable energy systems, optimal sizing of hybrid renewable energy system is critical to adequately meet the load demand at low cost. Studies based on optimal sizing of the hydrokinetic hybrid systems used the hybrid optimization model for electric renewable (HOMER) software as a simulation tool. The HOMER Legacy Version that was used in these studies does not have a built-in hydrokinetic module in its library. As a result, the authors used a wind turbine module to model a hydrokinetic turbine. In this study, the optimal size of a river-based micro-hydrokinetic pumped-hydro-storage (MHK-PHS) hybrid system is determined using HOMER Pro Version 3.6.1 since it has a built-in hydrokinetic module in its library. The results obtained using HOMER Pro Version 3.6.1 will be compared to the results obtained using the methodology adopted by previous researchers (using HOMER Legacy Version). The objective is to validate the best economical approach for sizing a hydrokinetic hybrid system. The second objective of the study is to investigate the effect brought by different demand sectors such as residential, commercial, and industrial load on sizing and operation of the proposed MHK-PHS hybrid system. The results have shown that the methodology applied in HOMER Legacy Version leads to oversizing of the storage system. This resulted into higher initial capital cost, net present cost (NPC), levelized cost of energy (COE), and operating cost. The results also proved that for the same daily energy consumption, a type of a demand sector does not affect the size of a hydrokinetic turbine and annual excess energy. Instead, it affects the size of the storage capacity as well as the size of the hydro-turbine. The commercial load profile proved to lead to the lowest state of charge (SOC) of the storage reservoir.

For the past few years, the South African government has been promoting the use of solar energy system which can be of benefit to the development of small power producers dispersed throughout the local grid. This support from the government usually comes with an obligation for the main electricity supplier to purchase the excessive power produced by small independent generators. In this paper the impact brought by different demand sector profiles on the daily operational cost and optimal scheduling of grid connected photovoltaic systems with bidirectional power flow is analysed for the specific case of Bloemfontein in South Africa. For this purpose, residential, commercial and industrial daily load curves are used to estimate daily load demands. For comparison purposes, three load profiles representing the demands from the residential, commercial and industrial sectors have been used and normalized to display the same daily energy consumption level with different demand patterns. The results of the simulations, obtained using Matlab 2016, have revealed that for the same energy consumption and renewable resources, the running expenses of any proposed scheme are mainly dependent on the demand sector. Consequently, it can be recommended that in Bloemfontein and South Africa in general more focus should be on implementing grid-connected renewable technologies with storage systems on the commercial and industrial sector instead of on the residential sector.