Search

To support the current trend of personalised medicine, a collaboration between different healthcare providers is increasingly vital. The main element is the ability to share data among all parties while abiding by a data sharing policy. The EPI (Enabling Personalised Intervention) project addresses the problem of personalised diagnosis by developing real-time monitoring services and digital health twins. The EPI services run over adaptive computing infrastructures which provide more flexibility to accommodate the different requests. This paper proposes the EPI framework to support these novel health services over programmable infrastructure. The framework works on aligning the parties’ ability to share data with the policy defined beforehand. We explain the approach by introducing the framework’s data sharing logic model. We define the formalism of the logic model to deduce feasible data movements between and possibly satisfy a data collaboration request. We reinforce the framework’s logic model by introducing the algorithms running on this federated system to simulate its workflow. We provide three healthcare use cases running on a typical EPI infrastructure. We evaluated our model according to three relevant parameters, performance, feasibility, and aggregation power, and we can conclude that our framework supports the required interoperability between the EPI partners.

TACTimat is a therapy tool developed for both hyper and hyposensitive children through a novel art and design focused approach. Our initial study explores how shape, texture, and density of elements on a tactile mat may influence the behaviour of young children with an Autistic Spectrum Disorder (ASD) or other cognitive disability during tactile exploration.

Wi-Fi Protected Access 3 (WPA3) became a mandatory part of the Wi-Fi certification on July 1st 2020. Therefore, the adoption rate of WPA3 is expected to grow soon. In this paper, we focus on WPA3 personal transition mode, in particular the security of this mode. We argue that transition mode is a requirement in home environments for the foreseeable future. We investigate whether it is possible to secure a WPA3 personal transition mode network in such a way that downgrade attacks are not feasible. We find that even with the security recommendations that the Wi-Fi Alliance recently issued for WPA3, common implementations running in transition mode can still be downgraded to WPA2. In our experiments, we can see that there are differences between WPA3 implementations in terms of security. The Wi-Fi Alliance has already announced upcoming additions to the WPA3 standard. These additions offer essential improvements to the security of WPA3 personal transition mode networks. We believe that the WPA3 certification should be extended to include the recently announced additions to WPA3. In addition to this, we make several recommendations to ensure the safe operation of WPA3. Together these changes will resolve most of the implementation differences we observed. Furthermore, we argue that mutual authentication is an essential stepping stone towards a more secure Wi-Fi ecosystem and discuss two mechanisms.

The push for data sharing and data processing across organisational boundaries creates challenges at many levels of the software stack. Data sharing and processing rely on the participating parties agreeing on the permitted operations and expressing them into actionable contracts and policies. Converting these contracts and policies into an operational infrastructure is still a matter of research and therefore begs the question of how should the software stack and infrastructure look like? What are the main building blocks of such an architecture? In this paper, we investigate the architecture of a multi-domain distributed architecture for policy driven application. The architecture spans components from auditing policies to managing network connections.

IEEE 802.11ax is the new standard introduced by the IEEE that focuses on improving efficiency of Wireless LANs. Among others, two of the newly introduced features are 1024 Quadrature Amplitude Modulation (QAM) and Orthogonal Frequency Division Multiple Access (OFDMA). In this paper, the expected throughput gain of 25% with the new 1024-QAM and expected latency reduction with downlink (DL) OFDMA are examined with state of the art reference boards of two different Wi-Fi chipset manufacturers. Using three Samsung S10 mobile phones as clients, experiments were performed to measure the expected increase in throughput and decrease in latency.The 25% increase in throughput with 1024-QAM was achieved, albeit with a maximum range of operation of less than 6 meters when using a maximal transmit power for MCS 11. We observed a very low percentage of packets transmitted using DL OFDMA in the traffic profiles used. We conclude that the introduction of DL OFDMA does not have a significant impact on the latency with the test scenarios covered. The performance of DL OFDMA is heavily dependent on the ability of the access point to properly schedule DL OFDMA transmissions. The number of clients, packet size, and buffer size play an essential role in the scheduling process. Based on our measurements the benefit of DL OFDMA is expected to be limited unless the number of clients is substantially higher than the three clients used, as would be the case in a stadium, train, or mall.

The current article presents a framework for the design and support of UOW Online, an entirely new unified university strategy for fully online learning. To aid teachers in the learning design process, we aim to create more awareness for teachers by determining the underlying learning design of their subject. To ensure the approach can be scaled up to cater for potentially hundreds of subjects, the manual labeling serves as input for an Artificial Intelligence (AI) algorithm that will train a model to label intended learning activities automatically. In addition to student demographics and behavior, the learning design and subject content will be used to augment an AI model that predicts future student outcomes. Future work focuses on the collection of necessary learning activities and manual encoding of these learning activities. Refereed/Peer-reviewed

The push for data sharing and data processing across organisational boundaries creates challenges at many levels of the software stack. Data sharing and processing rely on the participating parties agreeing on the permissible operations and expressing them into actionable contracts and policies. Converting these contracts and policies into a operational infrastructure is still a matter of research and therefore begs the question how should a digital data market place infrastructure look like? In this paper we investigate how communication fabric and applications can be tightly coupled into a multi-domain overlay network which enforces accountability. We prove our concepts with a prototype which shows how a simple workflow can run across organisational boundaries.

Applications and infrastructures are increasingly becoming more complex. Infrastructures have several layers of virtualisation, programmability and management while scientific applications are diverse in their computing model archetypes. Mapping these two opposing ends of the stack is a non-trivial task. Here we propose a new programming paradigm and architecture that takes into account the different layers of the stack in an effort to create isolated, portable and scalable application micro-infrastructures.

Traditional cloud Service Level Agreement (SLA) suffers from lacking a trustworthy platform for automatic enforcement. The emerging blockchain technique brings in an immutable solution for tracking transactions among business partners. However, it is still very challenging to prove the credibility of possible violations in the SLA before recording them onto the blockchain. To tackle this challenge, we propose a witness model using game theory and the smart contract techniques. The proposed model extends the existing service model with a new role called “witness” for detecting and reporting service violations. Witnesses gain revenue as an incentive for performing these duties, and the payoff function is carefully designed in a way that trustworthiness is guaranteed: in order to get the maximum profit, the witness has to always tell the truth. This is analyzed and proved through game theory using the Nash equilibrium principle. In addition, an unbiased sortition algorithm is proposed to ensure the randomness of the independent witnesses selection from the decentralized witness pool, to avoid possible unfairness or collusion. An auditing mechanism is also introduced in the paper to detect potential irrational or malicious witnesses. We have prototyped the system leveraging the smart contracts of Ethereum blockchain. Experimental results demonstrate the feasibility of the proposed model and indicate good performance in accordance with the design expectations.

Collaborative defences against, for example, Distributed Denial of Service attacks require trust between the collaborators. Even in a trusted alliance of collaborators, some members are more trusted than others with the execution of a specific task. This paper shows that trust can be used as a criterion in collaborative defences against attacks on computer infrastructures. We evaluate the implementation of a trust based defence in an overlay network and compare its' performance to other collaborative defences in the same environment. Results show that the trust-based defence has comparable results to the other collaborative defences after the trust values converge to reflect the behaviour of the nodes.

Workflow scheduling is one of the key issues in the management of workflow execution. Typically, a workflow application can be modeled as a Directed-Acyclic Graph (DAG). In this paper, we present GoDAG, an approach that can learn to well schedule workflows on multi-resource clusters. GoDAG directly learns the scheduling policy from experience through deep reinforcement learning. In order to adapt deep reinforcement learning methods, we propose a novel state representation, a practical action space and a corresponding reward definition for workflow scheduling problem. We implement a GoDAG prototype and a simulator to simulate task running on multiresource clusters. In the evaluation, we compare the GoDAG with three state-of-the-art heuristics. The results show that GoDAG outperforms the baseline heuristics, leading to less average makespan to different workflow structures.

This paper discusses the principles of organisation and infrastructure components of Open Data Markets (ODM) that would facilitate secure and trusted data exchange between data market participants, and other cooperating organisations. The paper provides a definition of the data properties as economic goods and identifies the generic characteristics of ODM as a Service. This is followed by a detailed description of the generic data market infrastructure that can be provisioned on demand for a group of cooperating parties. The proposed data market infrastructure and its operation are employing blockchain technologies for securing data provenance and providing a basis for data monetisation. Suggestions for trust management and data quality assurance are discussed.

Live TV production, due to its distributed nature, requires broadcasters to deploy equipments and human resources to several different places. This increases production costs. The traditional method through outside broadcasting vans is expensive. Migrating this type of application onto clouds is a promising method to reduce the cost. However, the Quality of Experience (QoE) can hardly be assured because of the cloud performance uncertainty. Auto-scaling the infrastructure based on dynamic workloads at runtime is also difficult. The feasibility of CloudsStorm framework, which is the core component of SWITCH (Software Workbench for Interactive, Time Critical and Highly self-adaptive Cloud applications) workbench, handling the lifecycle of the time critical application development and performance monitoring, infrastructure planning and provisioning, etc., is demonstrated with the aspect of live events broadcasting. It makes a significant attempt to fill the DevOps gap when migrating applications from legacy systems onto clouds. A live streaming application is demonstrated as a use case example.

It is universally accepted that the performance of graph algorithms is heavily dependent on the algorithm, the execution platform, and the structure of the input graph. This variability remains difficult to predict and hinders the choice of the right algorithm for a given problem. In this work, we focus on a case study: breadth-first search (BFS), a level-based graph traversal algorithm, running on GPUs. We first demonstrate the severity of this variability by presenting 32 variations of 5 implementation strategies for GPU-enabled BFS, and showing how selecting one single algorithm for the entire traversal can significantly limit performance. To alleviate these performance losses, we propose to mix-and-match, at runtime, different algorithms to compose the best performing BFS traversal. Our approach is based on two novel elements: a predictive model, based on a decision tree, which is able to dynamically select the best performing algorithm for each BFS level, and a quick context switch between algorithms, which limits the overhead of the combined BFS. We demonstrate empirically that our dynamic switching BFS achieves better performance, outperforming our non-switching implementations by 2x and existing state-of-the-art GPU BFS implementations by 3x. We conclude that mix-and-match BFS is a competitive approach for doing fast graph traversal, while being easily extended to include more BFS implementations and easily adaptable to other types of processors or specific types of graphs.

Memory bandwidth is critical for many modern applications and architectures. Parallel memories should alleviate this problem, but they are difficult to design and implement for non-trivial access patterns. Our work addresses these challenges through PolyMem, a Polymorphic Parallel Memory which acts as a software cache and enables parallel memory accesses for different combinations of access patterns. PolyMem's design is based on the Polymorphic Register File (PRF) design, which guarantees conflict-free parallel access for memory access patterns widely used in scientific and multimedia applications. We present in detail the design and implementation of MAXPolyMem, the first native implementation of PolyMem using Maxeler's toolchain. Our design supports multiple lanes, multiple read ports, and concurrent read and write operations. We further provide a detailed empirical analysis of the performance of MAX-PolyMem, including two sets of results. First, we conduct a thorough design space exploration to determine the best configurations and/or the performance bounds of MAX-PolyMem. For example, the design with the maximum read bandwidth is a 512KB memory, with 4 read ports, running at 137MHz, which can reach a peak read bandwidth of around 32GB/s. Second, to determine whether the MAXPolyMem can reach and sustain these peaks, we implement the STREAM-Copy benchmark. The benchmarking results demonstrate that, in practice, MAX-PolyMem reaches over 99% of the theoretical peak performance.

The IaaS (Infrastructure-as-a-Service) offered by Clouds provides applications with the capability of customizing VMs and configuring their network. Compared to traditional service-based IaaS applications such as persistent web services, most task-based applications have a relatively short duration but are triggered on demand. A typical way to support such kinds of application is to provision a shared and fixed virtual infrastructure based on pre-estimated size in advance, and then perform all the processing tasks. However, due to unpredictable workloads, this solution can lead to either cost inefficiency caused by over-provisioning, or failure to deliver the performance required by applications. CloudsStorm is a dynamic control framework proposed to provide applications with agile programmability and flexibility in controlling the virtual infrastructure. With its front end, applications can design their networked infrastructure and program that infrastructure with our interpreted infrastructure code language. With the back-end engine, the infrastructure code can be executed to provision the networked infrastructure, deploy and execute the application to obtain results, and release resources. Moreover, we adopt multi-threading to support parallel operation. Finally, we conduct experiments in an assumed scenario to demonstrate functionalities of CloudsStorm. The evaluation results prove CloudsStorm is efficient for task-based applications that need to exploit Clouds but reduce the monetary cost.

This paper presents results of the ongoing development of the Intercloud Security Framework (ICSF), that is a part of the Intercloud Architecture Framework (ICAF), and provides an architectural basis for building security infrastructure services for multi-cloud applications. The paper refers to general use case of the data intensive applications that indicate need for multi-cloud applications platforms that will require corresponding multi-cloud security services. The paper presents analysis of the general multi-cloud use case that helps eliciting the general requirement to ICSF and identifying the security infrastructure functional components that would allow using distributed cloud based resources and data sets. The paper defines the main ICSF services and functional components, and explains importance of consistent implementation of the Security Services Lifecycle Management in cloud based applications. The paper provides overview of the cloud compliance standards and their role in cloud security. The paper refers to the security infrastructure development in the CYCLONE project that implements federated identify management, secure logging service, and multi-domain Attribute Based Access Control, security services lifecycle management. The paper discusses implementation of the Trust Bootstrapping Protocol as an important mechanism to ensure consistent security in the virtualised inter-cloud environment

Software Defined Networking (SDN) technologies provide applications opportunities to manipulate underlying network flows and topologies via network controllers during runtime. In cloud environments, networked virtual machines can be enhanced by SDN by providing applications with controllable infrastructures to meet system-level quality requirements; however, customizing a suitable network topology with optimally placed controller(s) for given quality requirements and workload characteristics is often not an easy task. We call such problem virtual SDN network planning problem. In this paper, a Topology-Controller planner (TCPlanner) is proposed for customizing the network topology and placing the controllers. Experiments with different scales of network show that our approach can effectively plan virtual SDN networks to meet the various QoS requirements and reduce costs.

This paper describes the general architecture and functional components of the cloud based Big Data Infrastructure (BDI). The proposed BDI architecture is based on the analysis of the emerging Big Data and data intensive technologies and supported by the definition of the Big Data Architecture Framework (BDAF) that defines the following components of the Big Data technologies: Big Data definition, Data Management including data lifecycle and data structures, Big Data Infrastructure (generically cloud based), Data Analytics technologies and platforms, and Big Data security, compliance and privacy. The paper provides example of requirements analysis and implementation of two bioinformatics use cases on cloud and using SlipStream based cloud applications deployment and management automation platform being developed in the CYCLONE project. The paper also refers to importance of standardisation of all components of BDAF and BDI and provides short overview of the NIST Big Data Interoperability Framework (BDIF). The paper discusses importance of automation of all stages of the Big Data applications developments, deployment and management and refers to existing cloud automation tools and new developments in the SlipStream cloud automation platform that allows multi-cloud applications deployment and management

High-level tools for analyzing and predicting the performance GPU-accelerated applications are scarce, at best. Although performance modeling approaches for GPUs exist, their complexity makes them virtually impossible to use to quickly analyze the performance of real life applications and obtain easy-to-use, readable feedback. This is why, although GPUs are significant performance boosters in many HPC domains, performance prediction is still based on extensive benchmarking, and performance bottleneck analysis remains a nonsystematic, experience-driven process. In this context, we propose a tool for bottleneck analysis and performance prediction for GPU-accelerated applications. Based on random forest modeling, and using hardware performance counters data, our method can be used to quickly and accurately evaluate application performance on GPU-based systems for different problem characteristics and different hardware generations. We illustrate the benefits of our approach with three detailed use cases: a simple step-by-step example on a parallel reduction kernel, and two classical benchmarks (matrix multiplication and sequence alignment). Our results so far indicate that our statistical modeling is a quick, easy-to-use method to grasp the performance characteristics of applications running on GPUs. Our current work focuses on tackling some of its applicability limitations (more applications, more platforms) and improving its usability (full automation from input to user feedback).

usc