This work addresses the problem of ensuring service availability, trust, and profitability in sensor-cloud architecture designed toSensors-as-a-Service(Se-aaS) using IoT generated data. Due to the requirement of geographically distributed wireless sensor networks for Se-aaS, it is not always possible for a single Sensor-cloud Service Provider (SCSP) to meet the end-users requirements. To address this problem, we propose a federated sensor-cloud architecture involving multiple SCSPs for provisioning high-quality Se-aaS. Moreover, for ensuring trust in such a distributed architecture, we propose the use ofconsortium blockchainto keep track of the activities of each SCSP and to automate several functionalities throughSmart Contracts. Additionally, to ensure profitability and end-user satisfaction, we propose a composite scheme, named BRAIN, comprising of two parts. First, we defineminer's scoreto select an optimal subset of SCSPs asminersperiodically. Second, we propose a modifiedmultiple-leaders-multiple-followers Stackelberg game-theoretic approach to decide the association of an optimal subset of SCSPs to each service. Thereafter, we evaluate the performance of BRAIN by comparing with three existing benchmark schemes through simulations. Simulation results depict that BRAIN outperforms existing schemes in terms of profits and resource consumption of SCSPs, and price charged from end-users.

With the growth of multi-cloud computing across a heterogeneous substrate of public cloud, edge, and on-premise sites, observability has been gaining importance in compre-hending the state of availability and performance of large-scale geo-distributed systems. Collecting, processing and analyzing observability data from multiple geo-distributed clouds can be naturally modelled as dataflows comprising chained functions. These observability flows pose a unique set of challenges in-cluding ($a$) keeping cost budgets, resource overheads, network bandwidth consumed and latency low, (b) scaling to a large number of clusters, (c) adapting the volume of observability data to satisfy resource constraints and service level objectives (d) supporting diverse engines per dataflow depending on each processing function of the flow and (e) automating and optimizing placement of observability processing functions including closed-loop orchestration. Towards this end, we propose Octopus, a multi-cloud multi-engine observability processing framework. In Octopus, declarative observability dataflows (DODs) serve as an intent-driven abstraction for site reliability engineers (SREs) to specify self-driven observability dataflows. A dataflow engine in Octopus, then orchestrates these DODs to automatically deploy and self-manage observability dataflows over large fabrics spanning multiple clouds and clusters. Octopus supports a mix of streaming and batch functions and supports pluggable run-time engines, thereby enabling flexible composition of multi-engine observability flows. Our early deployment experience with Octopus is promising. We have successfully deployed production-grade metrics analysis and log processing data flows in an objective-optimized fashion across 1 cloud and 10 edge clusters spanning continents. Our results indicate data volume and WAN bandwidth savings of$2.3\mathrm{x}$and 56 %, respectively, the ability to support auto-scaling of DODs as input load varies, and the ability to flexibly relocate functions across clusters without hurting latency targets.

In the evolving computing landscape, extending from centralized clouds to multi-cloud and edge, the need for adaptable observability is becoming increasingly critical. Traditional static monitoring approaches grapple with inefficient data transfer, limited scalability, heterogeneous environments, and rigid metric processing pipelines. This paper introduces a novel metric processing system, Programmable Metric Flows (PMF), which is rooted in the principle of dynamism. PMF is a first- of-its-kind, light-weight, SQL-based metric processor. It empowers optimization-driven transformations of metrics, tailored to evolving resource availability and application requirements. This paper demonstrates how PMF enables various transformations for dynamic and fine-grained metric collection. We also showcase the capability of PMF for dynamically tuning the frequency of metrics to reduce the WAN cost in edge environments. Our experiments show that PMF performs at par with state-of-the- art techniques in terms of metric processing capability, with 10X lesser resource utilization. We envision PMF to usher in an era of lightweight programmability for observability platforms. PMF is open-source and available at https://github.com/observ-vol-mgtIPMF

With the increasing complexity of modern applications, particularly those relying on microservices architectures, the volume of observability data, encompassing logs, metrics, traces, etc., has surged significantly. This is further exacerbated by extensive cloud deployments, where observability is crucial for comprehending the health and performance of these systems, leading operations teams to collect as much data as possible for the “fear of missing out”. However, the collection, storage, and analysis of observability data entail significant costs, both in terms of resources and finances. Specifically, logs comprise the most substantial portion of observability data volume, thus exerting the greatest impact on observability cost. Moreover, logs also exhibit unstructured and noisy characteristics, where the efficacy of downstream AI for IT operations (AIOps tasks or Day-2 operations), such as fault classification, fault diagnosis, log anomaly detection etc., can be negatively impacted by log data volume. Hence, striking a balance between the verbosity of log observability and its impact on day-2 operations and debuggability is essential. In this paper, we introduce an autonomous system named SALO, which stands for Self-Adjusting Log Observability. SALO selectively collects logs based on real-time necessity, location, and granularity, as opposed to the conventional practice of collecting indiscriminately from all the components continuously. Our experiments show that SALO drastically decreases the log volume, by as much as 95 %, while still maintaining data quality for downstream AIOps usage, especially for post-hoc diagnosis tasks. Operating on a reduced volume of log data not only decreases storage, transfer, and retention costs but also streamlines observability pipelines, making them leaner, more efficient, and less resource-hungry.

This work presents an efficient data-centric client selection approach, named DICE, to enable federated learning (FL) over distributed edge networks. Prior research focused on assessing the computation and communication ability of the client devices for selection in FL. On-device data quality, in terms of data volume and heterogeneity, across these distributed devices is largely overlooked. The obvious outcome is the selection of an improper subset of clients with poor-quallity data, which inevitably results in an inefficient trained model. With an aim to address this problem, in this work, we design DICE which prioritizes the data quality of the client devices in the selection phase, in addition to their computation and communication abilities, to improve the accuracy of FL. Additionally, in DICE, we introduce the assistance of vicinal edge devices to account for the lack of computation or communication abilities in certain devices without violating the privacy-preserving guarantees of FL. Towards this aim, we propose a scheme to decide the optimal edge device, in terms of latency and workload, to be selected as the helper device. The experimental results show that DICE improves convergence speed for a given level of model accuracy. Further, the simulation results show that DICE reduces delay by at least 16%, energy consumption by at least 17%, and packet loss by at least 55% compared to the existing benchmarks while prioritizing the on-device data quality across clients.

In this work, the problem of high-quality energy service provisioning in the presence of competitive prosumers and micro-grids in cloud-enabled smart grid is studied. Oligopolistic prosumers behave non-cooperatively and store the excess generated energy for future use, which increases the load on the main grid and degrades the performance of the smart grid. To address this issue, we propose a dynamic cloud-based pricing scheme, named SmartPrice, to enforce cooperation among the prosumers for ensuring high quality of service provided by the micro-grids. In SmartPrice, using cloud infrastructure, each micro-grid calculates a reward factor for each prosumer based on his/her behavior to enforce cooperation among them. We model the interaction between each micro-grid and the prosumers using a single-leader-multiple-followers Stackelberg game, where the micro-grids and the prosumers act as the leaders and the followers, respectively. Each micro-grid determines the unit energy price to be charged/paid and each prosumer determines the quantity of excess energy to be supplied for ensuring high revenue. Thus, SmartPrice enforces cooperation among the micro-grids and prosumers. Additionally, using SmartPrice, the price for unit energy charged from the prosumers reduces by 23.37-$35.63\%$, thereby ensuring high revenue and the number of prosumers served by the micro-grids increases by 38.19-$53.14\%$.

In this article, we study the problem of quality of service (QoS)-aware cost management of sensor-cloud comprising multiple sensor-cloud service providers (SCSPs) and sensor-owners. The rapid adaptation of the wireless sensor network (WSN) and Internet-of-Things (IoT) technology led to the conceptualization of the sensor-cloud infrastructure which primarily aims to reduce the complexities associated with operating WSN-based applications by rendering Sensors-as-a-Service (Se-aaS). However, the oligopolistic market scenario of sensor-cloud involving multiple SCSPs and sensor-owners significantly impacts its profitability and QoS. Thus, there is a need to explore the dynamics of this market competition elaborately in order to maintain the usability of sensor-cloud. The existing works fail to address the aforementioned issues in sensor-cloud. Hence, in this work, we analyze the interactions among the sensor-owners and the SCSPs using a game-theoretic approach. We propose a QoS-aware dynamic cost management scheme, named QUEST, to determine the optimal strategies of the various actors in sensor-cloud market. Through simulations, we observe that, using QUEST, the price paid by the end-users decreases by 10.31-20.43 percent and the revenue of sensor-owners improves by 66.83-89.94 percent. Moreover, QUEST ensures the service satisfaction of the end-users while optimally distributing the services among the SCSPs and the sensor-owners.

In this paper, the problem of ensuring profitability for multiple sensor-owners in sensor-cloud, while satisfying the service requirements of end-users, is studied. In traditional sensor-cloud, Sensor-Cloud Service Provider (SCSP) solely dictates the service provisioning process. However, the SCSP cannot always ensure high profits for sensor-owners, who incur significant maintenance costs for their sensor-nodes. Contrarily, it is highly essential to meet the Quality-of-Information (QoI) requirements of end-users to ensure their service satisfaction. Existing works proposed a few node allocation schemes which neither consider the cost incurred by sensor-owners nor the QoI of sensed-data in sensor-cloud. To address this problem, a strategic resource allocation scheme, named RACE, is proposed, which introduces the participation of sensor-owners in the node allocation process. First, utility theory is used to calculate the optimum number of nodes to be allocated for a service. Thereafter, single leader multiple followers Stackelberg game is formulated to decide the number of nodes to be contributed by each sensor-owner and the price to be charged. Simulation-based experimental results reveal that, using RACE, the profits of the sensor-owners and those of the SCSP increase by 86.11–89.26 percent and 41.95–80.82 percent, respectively, as compared to the existing benchmark schemes, while considering that each sensor-node is capable of serving multiple applications simultaneously. Moreover, service availability in sensor-cloud increases by 31.70–96.96 percent using RACE.

Sensor-cloud architecture is a wireless sensor network (WSN)-based Service-Oriented Architecture (SOA), in which a Sensor-Cloud Service Provider (SCSP) obtains WSNs on rental basis from multiple sensor-owners and provides these resources to the users in the form of chargeable units of services, termed as Sensors-as-a-Service (Se-aaS). A fraction of the revenue earned by the SCSP from the users is distributed among the oligopolistic sensor-owners for the usage of their nodes. Due to the inter-dependency among the sensor-owners for Se-aaS provisioning, selfish sensor-owners behave dishonestly to gain higher profits, thereby degrading the overall QoS. Existing works on sensor-cloud fail to address this issue. Hence, in this work, we propose DETER, a dynamic trust enforcing pricing scheme, which enforces trust among the selfish sensor-owners while ensuring profits for the SCSP.

In this paper, the problem of provisioning high quality of Sensors-as-a-Service (Se-aaS) in the presence of competitive sensor-owners, i.e., oligopolistic market, and heterogeneous sensor nodes in service-oriented sensor-cloud is studied. Oligopolistic sensor-owners adopt unfair means to degrade the quality of service provided by other sensor-owners in the sensor-cloud market. In order to address this problem, a dynamic pricing scheme, named DETER, is proposed in this work to enforce trust among the sensor-owners for maintaining the quality of Se-aaS provided by the Sensor-Cloud Service Provider (SCSP). Each sensor node calculates distributed trust opinion for other nodes, while the SCSP calculates centralized trust opinion for each sensor-owner. A Single-Leader-Multiple-Follower Stackelberg Game is formulated in which the SCSP acts as the leader and decides price to be paid to each sensor-owner, while ensuring maximum profit. On the other hand, the sensor-owners act as the followers and decide their strategies for earning maximum profit. Thereby, using DETER, SCSP enforces high trust among the sensor-owners. Additionally, using DETER, energy consumption of sensor nodes in sensor-cloud decreases by 4.69-11.56 percent, and network overhead decreases by 52.6-56.53 percent. The trade-off between price earned by the sensor-owners and profit of the SCSP in service-oriented sensor-cloud is also maintained using DETER.

In this paper, we study the problem of dynamic mapping of virtual sensors in sensor-cloud for provisioning high quality of Sensors-as-a-Service (Se-aaS) in the presence of multiple sensor-owners and heterogeneous sensor nodes. We divide this problem into two subproblems—optimal dispersed node selection and optimal data-rate distribution, and analyze that these problems are NP-complete. Hence, we propose a game theory-based online scheme, named QADMAP, to solve these two problems in polynomial time. For the optimal node selection problem, we design a dynamic coalition-formation game-based online scheme, while maximizing thedispersion indexof the selected nodes. On the other hand, we propose an evolutionary game theory-based scheme for distributing the data-rate requirements of the services among the selected nodes, optimally. As per our knowledge, none of the existing works on dynamic mapping of virtual sensors considers the stochastic behavior of sensor-cloud for provisioning Se-aaS. From simulations, we observe that, using QADMAP, the energy consumption of the network reduces by 29.88-31.73 percent, thereby improving the QoS in terms of service availability by 11 percent and increasing the profit of the SCSP by 3.63-9.82 percent, compared to the existing benchmark schemes.

In this work, we address the problem of efficient utilization of resource-constrained wireless sensor nodes for provisioning Sensors-as-a-Service (Se-aaS) with high quality. In sensor-cloud, the sensor-owners provide their respective sensor nodes to the sensor-cloud service provider (SCSP) on rent. The SCSP utilizes these nodes to create virtual sensors and provisions them as Se-aaS for serving their WSN-dependent applications of the end-users and earns revenue in exchange. To ascertain high quality-of-service (QoS) of Se-aaS while simultaneously ensuring profits for itself and the sensor-owners, the SCSP needs to optimally allocate physical sensor nodes to serve the virtual sensors, while considering their limited capacity and the fair distribution of service load among different sensor-owners. Although a few existing works focused on resource allocation problem in sensor-cloud, none of them considered the possibility of sharing the same physical sensor node among multiple virtual sensors. Hence, in this work, we propose a resource orchestration scheme for sensor-cloud, named SensOrch, which is based on coalition formation game with transferable utility. Using SensOrch, the SCSP ensures the optimal allocation of sensor nodes to form virtual sensors while maintaining high QoS and profitability of Se-aaS. Through simulations, we yield that, using SensOrch, the network lifetime increases by 25.31 - 59.6% along with a simultaneous increase in the profit of the SCSP by 23.64 - 29.49%, compared to the existing schemes. Additionally, SensOrch ensures fair revenue distribution among the sensor-owners.

In this work, we propose a dynamic cache-based pricing scheme, named CASH, for service-oriented sensor-cloud. In sensor-cloud, the Sensor-Cloud Service Provider (SCSP) provisions Sensors-as-a-Service (Se-aaS) to multiple end-users based on a pay-per-use model. The service-requests of the end-users have heterogeneous data-rate requirements. In the cache-enabled architecture of sensor-cloud, these service-requests are served by the SCSP using either the Internal or the External cache, which incurs different costs to the SCSP. Thereby, the SCSP tries to maximize its own profit by distributing these service-requests, optimally, among the caches. Additionally, the SCSP ensures that the end-users are minimally charged. Existing literature fails to propose any pricing scheme for service-oriented sensor-cloud, while considering the cost incurred for data caching. In CASH, we propose a dynamic pricing model for sensor-cloud using dynamic coalition formation game with transferable utility. Using CASH, based on the preference relation of the partitions, we determine the optimal internal cache refresh rate, while maximizing the coalition value. Through simulation, we observe that the cost incurred by the SCSP reduces by 34.32–51.15% and the price paid by the end-users decreases by 9.60–17.47% as compared to the existing schemes. Additionally, CASH ensures 9.60–21.85% increase in the profit of the SCSP.