In an era where the Internet of Things (IoT) is becoming integral to daily life, the demand for efficient computing solutions is rapidly increasing, and the orchestration of containers in fog computing environments has emerged as a critical area of research. Single-Board Computers (SBCs) are particularly well-suited to fog computing environments due to their low cost, energy efficiency, and local processing capabilities. However, the efficient orchestration of containerized applications on these resource-constrained machines at the cost of performance is still an open issue. This paper presents Chronos (Customizable Heuristic Resource Orchestrator for Node Optimization Scheduling) , a customizable scheduling framework for Kubernetes designed to overcome existing architectural limitations. The framework features a programmable scheduling operator with advanced monitoring capabilities, enabling system administrators to adapt scheduling policies to application-specific needs. It operates by harnessing real-time telemetry and time-series performance data. Chronos was deployed on a realistic SBC testbed with representative services and synthetic workloads simulating user behavior. Experimental results show that Chronos improves performance and resource utilization over baseline scheduling algorithms. In particular, when compared to the default Kubernetes scheduler, Chronos customized scheduler was able to reduce network latency up to 23% for network-intensive workloads, disk write operations up to 42% for disk-intensive workloads, and response time up to 20% for CPU-intensive workloads, while maintaining low overhead.

With the proliferation of Internet of Things (IoT) ecosystems, traditional resource orchestration mechanisms, executed on fog devices, encounter significant scalability, reliability and security challenges. To tackle these challenges, recent decentralized algorithms in Fog-IoT use Distributed Ledger Technologies to orchestrate resources and payments between peers. However, while distributed ledgers provide many desirable properties, their consensus mechanism introduces a performance bottleneck. This paper introduces Light-HIDRA, a consensus-less and decentralized resource orchestration system for Fog-IoT environments. At its core, Light-HIDRA uses Byzantine Reliable Broadcast (BRB) to coordinate actions without centralized control, therefore drastically reducing communication overhead and latency compared to consensus-based solutions. Light-HIDRA coordinates the scheduling and execution of workloads, and securely manages the payments that peers receive for dedicating resources to workloads. Light-HIDRA further increases performance and reduces overhead by grouping peers into distinct domains. We conduct an in-depth analysis of the protocol’s security properties, investigating its efficiency and robustness in diverse situations. We evaluate the performance of Light-HIDRA, highlighting its performance over HIDRA, a state-of-the-art baseline that uses smart contracts. Our experiments demonstrate that Light-HIDRA reduces the bandwidth usage by up to 57x, the latency of workload offloading by up to 142x, and shows superior throughput compared to HIDRA.

Abstract Information and communication technologies backbone of a smart city is an Internet of Things (IoT) application that combines technologies such as low power IoT networks, device management, analytics or event stream processing. Hence, designing an efficient IoT architecture for real-time IoT applications brings technical challenges that include the integration of application network protocols and data processing. In this context, the system scalability of two architectures has been analysed: the first architecture, named as POST architecture, integrates the hyper text transfer protocol with an Extract-Transform-Load technique, and is used as baseline; the second architecture, named as MQTT-CEP, is based on a publish-subscribe protocol, i.e. message queue telemetry transport, and a complex event processor engine. In this analysis, SAVIA, a smart city citizen security application, has been deployed following both architectural approaches. Results show that the design of the network protocol and the data analytic layer impacts highly in the Quality of Service experimented by the final IoT users. The experiments show that the integrated MQTT-CEP architecture scales properly, keeps energy consumption limited and thereby, promotes the development of a distributed IoT architecture based on constraint resources. The drawback is an increase in latency, mainly caused by the loosely coupled communication pattern of MQTT, but within reasonable levels which stabilize with increasing workloads.

Fog computing arises as a complement to cloud computing where computing and storage are provided in a decentralized way rather than the centralized approach of the cloud paradigm. In addition, blockchain provides a decentralized and immutable ledger which can provide support for running arbitrary logic thanks to smart contracts. These facts can lead to harness smart contracts on blockchain as the basis for a decentralized, autonomous, and resilient orchestrator for the resources in the fog. However, the potentially vast amount of geographically distributed fog nodes may threaten the feasibility of the orchestration. On the other hand, fog nodes can exhibit highly dynamic workloads which may result in the orchestrator redistributing the services among them. Thus, there is also a need to dynamically support the network connections to those services independently of their location. Software Defined Networking (SDN) can be integrated within the orchestrator to carry out a seamless service management. To tackle both aforementioned issues, the S-HIDRA architecture is proposed. It integrates SDN support within a blockchain-based orchestrator of container-based services for fog environments, in order to provide low network latency and high service availability. Also, a domain-based architecture is outlined as potential scenario to address the geographic distributed nature of fog environments. Results obtained from a proof-of-concept implementation assess the required functionality for S-HIDRA.

Summary Fog computing has become a complementary technology to cloud computing and addresses some of the cloud computing threats such as the response time and network bandwidth demand. Fog computing successes processing data and storing data near to the edge, and usually is combined with container virtualization to provide hardware isolation. Empowered by these capabilities, numerous Internet of Things (IoT) applications are developed as virtualized instances on resource‐constrained fog nodes such as single‐board computers (SBC). In addition, blockchain has emerged as a key technology that is transforming the way we share information. Blockchain technology represents a decentralised, distributed, and immutable database ledger and is a potential solution for the distributed ecosystem of IoT applications. The distributed structure of blockchain is naturally suitable for IoT applications. However, it introduces new challenges related to CPU overhead or response time. This paper proposes a layered architecture that integrates blockchain technology and OS‐level virtualization technology to develop fog‐based IoT applications. It also provides insights for future deployments through a proof‐of‐concept use case harnessing SBCs, in this case Raspberry Pi, as blockchain‐enabled fog nodes to drive virtualized IoT applications. The study shows that the maximum CPU overhead added by a permissioned blockchain based on Ethereum on the Raspberry Pi is around a 25% under stress situations while the overhead introduced by the sealer process is negligible. These results support the feasibility of using blockchain on resource‐constrained fog nodes for supporting IoT applications.

Summary Energy management has become increasingly necessary in data centers to address all energy‐related costs, including capital costs, operating expenses, and environmental impacts. Heterogeneous systems with mixed hardware architectures provide both throughput and processing efficiency for different specialized application types and thus have a potential for significant energy savings. However, the presence of multiple and different processing elements increases the complexity of resource assignment. In this paper, we propose a system for efficient resource management in heterogeneous clouds. The proposed approach maps applications' requirement to different resources reducing power usage with minimum impact on performance. A technique that combines the scheduling of custom hardware accelerators, in our case, Field‐Programmable Gate Arrays (FPGAs) and optimized resource allocation technique for commodity servers, is proposed. We consider an energy‐aware scheduling technique that uses both the applications' performance and their deadlines to control the assignment of FPGAs to applications that would consume the most energy. Once the scheduler has performed the mapping between a VM and an FPGA, an optimizer handles the remaining VMs in the server, using vertical scaling and CPU frequency adaptation to reduce energy consumption while maintaining the required performance. Our evaluation using interactive and data‐intensive applications compare the effectiveness of the proposed solution in energy savings as well as maintaining applications performance, obtaining up to a 32% improvement in the performance‐energy ratio on a mix of multimedia and e‐commerce applications.

Heterogeneous hardware acceleration architectures are becoming an important tool for achieving high performance in cloud computing environments. Both FPGAs and GPUs provide huge computing capabilities overlarge amounts of data. At the same time, energy efficiency is a big concern in nowadays computing systems. In this paper we propose a novel architecture aimed at integrating hardware accelerators into a Cloud Computing infrastructure, and making them available as a service. The proposal harnesses advances in FPGA dynamic reconfiguration and efficient virtualization technology to accelerate the execution of certain types of tasks. In particular, applications that can be described as Big Data would greatly benefit from the proposed architecture.

Providing Quality of Service (QoS) in Grid environments is still a challenging task because advance reservation of resources is not always possible. By contrast, scheduling the use of resources in advance is a way of enhancing the provision of QoS, without physically reserving them. However, in a Grid scenario fragmentation may appear as a logic result of the allocation process, leading to poor resource utilization. To try to avoid that problem a rescheduling technique has been developed and applied periodically depending on the existing fragmentation. Nevertheless, reallocating jobs in an optimal way is an NP-Hard problem. Consequently, simple and scalable ways of distributing and allocating those jobs into the resources are presented and evaluated in a real Grid environment involving heterogeneous computing resources distributed across different national organizations, highlighting the benefits of using them to increase the resource usage.

In highly heterogeneous and distributed systems, like Grids, it is rather difficult to provide QoS to the users. As reservations of resources may not always be possible, another possible way of enhancing the perceived QoS is by performing meta-scheduling of jobs in advance, where jobs are scheduled some time before they are actually executed. Thank to this, it is more likely that the appropriate resources are available to execute the job when needed. When using this type of scheduling, fragmentation appears and may become the cause of poor resource utilization. Because of that, some techniques are needed to perform rescheduling of tasks that may reduce the existing fragmentation. To this end, knowing the status of the system is a must. However, how to measure and quantify the existing fragmentation in a Grid system is a challenging task. This paper proposes different metrics aiming at measuring that fragmentation not only at resource level but also taking into account all the resources of the Grid environment as a whole. Finally, a performance evaluation of the proposed metrics over a real test bed is presented.

The need of mechanisms to guarantee the Quality of Service (QoS) in Grid environments has empowered the interest on Service Level Agreements (SLAs), which represent the contract established by negotiation between users and service providers that must be accomplished. Consequently, the SLAs contain numerous terms related to the QoS expected. The aim of this paper is to address different QoS levels in Grid environments by using specific metrics on SLAs. These metrics represent the expected QoS by the user for the given service level terms. Hence, a characterization and classification of these terms, making a differentiation of QoS in levels, is needed in order to improve the number of fulfilled agreements. Our proposal is based on a WS-Agreement compliant architecture developed to provide support for SLAs in Grid environments, also providing a framework prepared for new terms and policy definitions. More precisely, the QoS expected by users is clearly defined in three levels. These levels are used to classify the compromise adquired by each SLA (depending on the expectations of the user that submitted it) and deal with the confidence that Grid resources provide. The evaluation on a real Grid testbed shows the efficiency of the proposal.

Grids are made of heterogeneous computing resources geographically dispersed where providing Quality of Service (QoS) is a challenging task. One way of enhancing the QoS perceived by users is by performing scheduling of jobs in advance, since reservations of resources are not always possible. This way, it becomes more likely that the appropriate resources are available to run the job when needed. One drawback of this scenario is that fragmentation appears as a well known effect in job allocations into resources and becomes the cause for poor resource utilization. So, a new technique has been developed to tackle fragmentation problems, which consists of rescheduling already scheduled tasks. To this end, some heuristics are implemented to calculate the intervals to be replanned and to select the jobs involved in the process. Moreover, another heuristic is implemented to put rescheduled jobs as close together as possible to minimize the fragmentation. This technique has been tested using a real test bed.

In current distributed systems, such as Grids, Clouds, or P2P systems, the amount of information to handle influences the way the system is managed. In P2P systems containing large quantities of data, or in Grid systems containing a large number of (often heterogeneous) resources, information about data or resources must be spread through the system in an efficient way in order to allow them to be found. An information discovery technique based on data summarization, via clustering, is presented. These summaries can be used to classify information to provide users with greater insight about documents or computing resources compared to raw data. Also, meta-schedulers or brokers would benefit from the proposed technique due to the fact that they would have to deal with less data from resources, thus aiding to the scalability of the system. An evaluation of the approach is subsequently provided to identify the impact of choosing particular parameters to be used as part of the summary.

The aggregation of heterogeneous and geographically distributed resources for new science and engineering applications has been made possible thanks to the deployment of Grid technologies. These systems have communication network requirements which should be taken into account when performing usual tasks such as scheduling, migrating or monitoring of jobs. Recall the network should be used in an efficient and fault-tolerant way since the users, services, and data use it to communicate with each other. However, most of the existing efforts to provide QoS in Grids do not take network issues into account, and focus instead on processor workload and disk access. Authors have previously proposed a proof-of-concept implementation of a network-aware metascheduler, developed as an extension to GridWay. This extension harnesses network information to perform scheduling tasks. In this work, a tuning of the network monitoring tool is presented and evaluated. Results show that the overhead caused by this monitoring tool varies depending on the monitoring parameters, and these ones, in turn, depend on the actual Grid environment.

Abstract Grid technologies have enabled the aggregation of geographically distributed resources in the context of a particular application. The network remains an important requirement for any Grid application, as entities involved in a Grid system (such as users, services, and data) need to communicate with each other over a network. The performance of the network must therefore be considered when carrying out tasks such as scheduling, migration or monitoring of jobs. Surprisingly, many existing quality of service efforts ignore the network and focus instead on processor workload and disk access time. Making use of the network in an efficient and fault‐tolerant manner is challenging. In a previous contribution, we proposed an autonomic network‐aware scheduling architecture that is capable of adapting its behavior to the current status of the environment. Now, we present a performance evaluation in which our proposal is compared with a conventional scheduling strategy. We present simulation results that show the benefits of our approach. Copyright © 2009 John Wiley & Sons, Ltd.

Grid technologies are emerging as the next generation of distributed computing, allowing the aggregation of resources that are geographically distributed across different locations. However, these resources are independent and managed separately by various organizations with different policies. This will have a major impact to users who submit their jobs to the Grid, as they have to deal with issues such as policy heterogeneity, security and fault tolerance. Moreover, the changes of Grid conditions, such as resources that may become unavailable for a period of time due to maintenance and/or suffer failures, would significantly affect the quality of service (QoS) requirements of users. Therefore, it is essential for users to take into account the effects of resource failures during jobs execution. In this paper, we present our work on introducing resource failures and failure detection into the GridSim simulation toolkit. As we need to conduct repeatable and controlled experiments, it is easier to use simulation as a means of studying complex scenarios. We also give a detailed description of the overall design and a use case scenario demonstrating the conditions of resources varied over time.

During last years, new applications needing more requirements have been developed, and their requirements have boosted the search for what we know as Quality of Service, QoS. To provide QoS, diverse formulas have been studied. Among them, there are the call admission control algorithms (CAC) and the usage parameters control algorithms (UPC). In networks that support QoS, a call admission control (CAC) algorithm determines whether or not a new traffic flow can be admitted to the network such that all users will receive their performance. Such an algorithm is a key component of multi-service networks as it determines the extent to which network resources are utilized and whether the promised QoS parameters are actually delivered. In this paper, a new lightweight algorithm, the Two Tests algorithm, is presented. This new proposal requires neither intense use of CPU to make decisions nor a complex traffic characterization, as opposed to other well-known algorithms, such as the Equivalent Capacity algorithm. Evaluation has been performed through simulation by using the OPNET (http://www.opnet.com/) environment, and considering CBR (constant bit rate,) traffic and VBR (variable bit rate,) traffic based on MPEG-2 video traces. It is shown that, when compared to the well-known Equivalent Capacity algorithm, the Two Tests algorithm is able to make a more effective admission control, at a fraction of its computational cost.

Quality of service (QoS) support in local and cluster area environments has become an issue of great interest in recent years. Most current high-performance interconnection solutions for these environments have been designed to enhance conventional best-effort traffic performance, but are not well-suited to the special requirements of the new multimedia applications. The multimedia router (MMR) aims at offering hardware-based QoS support within a compact interconnection component. One of the key elements in the MMR architecture is the algorithms used in traffic scheduling. These algorithms are responsible for the order in which information is forwarded through the internal switch. Thus, they are closely related to the QoS-provisioning mechanisms. In this paper, several traffic scheduling algorithms developed for the MMR architecture are described. Their general organization is motivated by chances for parallelization and pipelining, while providing the necessary support both to multimedia flows and to best-effort traffic. Performance evaluation results show that the QoS requirements of different connections are met, in spite of the presence of best-effort traffic, while achieving high link utilizations.

The explosive growth in using scalable and cost-effective clusters and local area environments involve the design of high performance networks aimed at providing QoS to multimedia flows. Thus, the main goal pursued by the Multi-Media (MMR) project is to design a single-chip router able to efficiently handle multimedia flows and best-effort traffic. In this paper we focus on the performance evaluation of the MMR architecture using a mix of CBR, VBR and best effort workload. Preliminary simulation results show that, by using simple link and switch scheduling algorithms, the router is able to achieve a link bandwidth utilization of 80%, while still providing QoS guarantees to both CBR and VBR traffic in the presence of best-effort traffic.

The performance of many parallel applications takes great benefits from the improvements made on multicomputer interconnection networks. The routers on which these networks are based should be carefully designed in order to offer good performance, while keeping their implementations as simple and cost-effective as possible. In this paper we present a new starvation and deadlock free routing algorithm: the ghost-packet algorithm. Detailed evaluation of both deterministic and adaptive versions of the algorithm show that their performance is better than the one of other state-of-the-art routing algorithms, mainly because of the hardware cost reduction achieved in router design.

The primary objective of the Multimedia Router (MMR) project is the design and implementation of a compact router optimized for multimedia applications. The router is targeted for use in cluster and LAN interconnection networks, which offer different constraints and therefore differing router solutions than WANs. One of the key design parameters is the amount of buffer space, which is closely related to the silicon area required to implement the router. In this paper, the MMR performance obtained when varying the size of the input buffers is explored. Preliminary results show that buffers as small as one flit large suffice to guarantee QoS to both CBR and VBR traffic, thanks to the use of flow control and short links.

We propose a flow control algorithm for k-ary n-cube networks which avoids the deadlock problems without using virtual channels. Some basic definitions and theorems are proposed in order to establish the necessary and sufficient conditions to verify that an algorithm is deadlock-free. Our proposal is based on a restriction of the virtual cut-through flow control rather than of the routing algorithm and it can be applied both over central buffers or edge buffers. A minimum free buffer space of two packets is required. The implementation complexity of the router according to Chien's (1993) model, is much easier and faster than using virtual channels. Network simulations considering the router complexity show the performance achieved by this new algorithm. The results display a latency improvement of 20% to 35% compared with the use of virtual channels depending on the load of the network.

The analysls, design and evaluation of the interconnection subsystem for massively parallel arch i~ectures is norm ally carried out using computer simulation tools, requiring