William Tärneberg

Detection of Runtime Conflicts among Services in Smart Cities

The populations of large cities around the world are growing rapidly. Cities are beginning to address this problem by implementing significant sensing and actuation infrastructure and building services on this infrastructure. However, as the density of sensing and actuation increases and as the complexities of services grow there is an increasing potential for conflicts across Smart City services. These conflicts can cause unsafe situations and disrupt the benefits that the services were originally intended to provide. Although some of the conflicts can be detected and avoided during designing the services, many can still occur unpredictably during runtime. This paper carefully defines and enumerates the main issues regarding the detection and resolution of runtime conflicts in smart cities. In particular, it focuses on conflicts that arise across services. This issue is becoming more and more important as Smart City designs attempt to integrate services from different domains (transportation, energy, public safety, emergency, medical, and many others). Research challenges are identified and then addressed that deal with uncertainty, dynamism, real-time, mobility and spatio-temporal availability, duration and scale of effect, efficiency, and ownership. A watchdog architecture is also described that oversees the services operating in a Smart City. This watchdog solution detects and resolves conflicts, it learns and adapts, and it provides additional inputs to decision making aspects of services. Using data from a Smart City dataset, an emulated set of services and activities using those services are created to perform a conflict analysis. A second analysis hypothesizes 41 future services across 5 domains. Both of these evaluations demonstrate the high probability of conflicts in smart cities of the future.

How Beneficial Are Intermediate Layer Data Centers in Mobile Edge Networks

To reduce the congestion due to the future bandwidth-hungry applications in domains such as Health care, Internet of Things (IoT), etc., we study the benefit of introducing additional Data Centers (DCs) closer to the network edge for the optimal application placement. Our study shows that the edge layer DCs in a Mobile Edge Network (MEN) infrastructure is cost beneficial for the bandwidth-hungry applications having their strong demand locality and in the scenarios where large capacity is deployed at the edge layer DCs. The cost savings for such applications can go up to 67%. Additional intermediate layer DCs close to the root DC can be marginally cost beneficial for the compute intensive applications with medium or low demand locality. Hence, a Telecom Network Operator should start building an edge DC first having capacity up to hundreds of servers at the network edge to cater the emerging bandwidth-hungry applications and to minimize its operational cost.

Experiences creating a framework for smart traffic control using AWS IOT

Public clouds such as Amazon Web Services (AWS) and Microsofts Azure provide excellent capabilities for scalable Web applications and Hadoop-based processing. Recent additions to public clouds to support connected devices and IoT have the potential to similarly disrupt emerging home-grown and/or proprietary approaches. While early public cloud IoT success stories have focused on smaller-scale scenarios such as connected houses, it is unclear to what extent these new public cloud mechanisms and abstractions are suitable and effective for larger-scale and/or scientific scenarios, which often have a different set of constraints or requirements. In this paper, the design and implementation of a representative cloud-based IoT infrastructure in a specific public cloud – AWS – is presented. The system created is for dynamic vehicle traffic control based on vehicle volumes/patterns and public transport punctuality. We find that constructing server-less, stateful, and data driven IoT applications in AWS that can operate in real-time is non-trivial. The primary challenges span application manageability and design, latency performance, asynchronicity, and scalability.

Utilizing Massive MIMO for the Tactile Internet: Advantages and Trade-Offs

Controlling robots in real-time over a wireless interface present fundamental challenges for forthcoming fifth generation wireless networks. Mission critical real-time applications such as telesurgery over the tactile Internet require a communication link that is both ultra-reliable and low-latency, and that simultaneously serving multiple devices and applications. Wireless performance requirements for these applications surpass the capabilities of current wireless cellular standards. The prevailing ambitions for the fifth generation wireless specifications go beyond higher throughput and embrace the wireless performance demands of mission critical real-time applications in robotics and the Internet of Things. To accommodate these demands, changes have to be made across all layers of the wireless infrastructure. The fifth generation wireless standards are far from finalized but massive Multiple-Input Multiple- Output has surfaced as a strong radio access technology candidate and has great potential to cope with all these stringent requirements. In this paper, we investigate how Ultra-Reliable and Low-Latency Communication with massive MIMO can be achieved for bilateral teleoperation, an integral part of the tactile Internet. We conclude through simulation what the performance bounds are for massive MIMO and thus how to configure such a system for near deterministic latency and what the inherit trade-offs are.

Distributed Approach to the Holistic Resource Management of a Mobile Cloud Network

The Mobile Cloud Network is an emerging cost and capacity heterogeneous distributed cloud topological paradigm that aims to remedy the application performance constraints imposed by centralised cloud infrastructures. A centralised cloud infrastructure and the adjoining Telecom network will struggle to accommodate the exploding amount of traffic generated by forthcoming highly interactive applications. Cost effectively managing a Mobile Cloud Network computing infrastructure while meeting individual applications performance goals is non-trivial and is at the core of our contribution. Due to the scale of a Mobile Cloud Network, a centralised approach is infeasible. Therefore, in this paper a distributed algorithm that addresses these challenges is presented. The presented approach works towards meeting individual applications performance objectives, constricting system-wide operational cost, and mitigating resource usage skewness. The presented distributed algorithm does so by iteratively and independently acting on the objectives of each component with a common heuristic objective function. Systematic evaluations reveal that the presented algorithm quickly converges and performs near optimal in terms of system-wide operational cost and application performance, and significantly outperforms similar naïve and random methods.

Telco Clouds: Modelling and Simulation

In this paper, we propose a telco cloud meta-model that can be used to simulate different infrastructure con- figurations and explore their consequences on the system performance and costs. To achieve this, we analyse current telecommunication and data centre infrastructure paradigms, describe the architecture of the telco cloud and detail the benefits of merging both infrastructures in a unified system. Next, we detail the dynamics of the telco cloud and identify the components that are the most relevant from the perspective of modelling performance and cost. A number of well established simulation technologies exist for most of the telco cloud components, we thus proceed with surveying existing models in an attempt to construct a suitable composite meta-model. Finally, we present a showcase scenario to demonstrate the scope of our telco cloud simulator. (Less)

Cross-layer control for bounded shared state inconsistency in wireless IoT devices

The devices that constitute the Internet of Things are evolving to include more than just enabling sensing and actuation over a wireless interface. In a contemporary scenario, these devices perform tasks and contribute to an aggregate information flow, in a distributed manner. In the wake of this evolution, new distributed Internet of Things frameworks have emerged. These frameworks maintain a distributed shared state in a distributed hash table. An Internet of Things systems ability to make decisions autonomously and distributively depend on the level of consistency of the shared state. As wireless resources are scarce, the amount of deferred state information in each device is unknown when the system is highly utilised. In this paper, we have developed a controller with the objective to achieve a bounded time-average of deferred state information in each device while maintaining system stability. The controller is derived using Lyapunov drift optimisation with penalty. The resulting controller can successfully bound the shared state consistency level within a narrow margin, maintain system stability, and balance the traffic flow trade-off more successfully than comparable and conventionally used methods.

A control theoretical approach to non-intrusive geo-replication for cloud services

Complete data center failures may occur due to disastrous events such as earthquakes or fires. To attain robustness against such failures and reduce the probability of data loss, data must be replicated in another data center sufficiently geographically separated from the original data center. Implementing geo-replication is expensive as every data update operation in the original data center must be replicated in the backup. Running the application and the replication service in parallel is cost effective but creates a trade-off between potential replication consistency and data loss and reduced application performance due to network resource contention. We model this trade-off and provide a control-theoretical solution based on Model Predictive Control to dynamically allocate network bandwidth to accommodate the objectives of both replication and application data streams. We evaluate our control solution through simulations emulating the individual services, their traffic flows, and the shared network resource. The MPC solution is able to maintain a consistent performance over periods of persistent overload, and is quickly able to indiscriminately recover once the system return to a stable state. Additionally, the MPC balances the two objectives of consistency and performance according to the proportions specified in the objective function.

Workload displacement and mobility in an omnipresent cloud topology

Latency and throughput demands on cloud hosted services are growing more complex as cloud services are at an increasing rate being consumed on mobile devices. On mobile devices, cloud services are accessed through a WAN and a mobile access network, through which latency is added and throughput restricted, resulting in an inconsistent user experience. The proposed omnipresent cloud topology paradigm attempts to remedy this latency decay by placing generic cloud data centres, of arbitrary size, in closer geographic proximity to the end user, thus reducing the geographic discrepancy that contribute to congestion and latency. A key performance challenge in the omnipresent cloud paradigm is the incurred cost of service migration as a result of user mobility. In this paper we examine fundamental resource costs and dynamics of user mobility in an omnipresent cloud topology. Furthermore, this paper also propose and evaluates a simulation model capturing the fundamental dynamics of an omnipresent cloud architecture in an extreme operating scenario. Our simulations reveals that mobility significantly affects the proportion of sessions that are migrated between consecutive nodes and that migration can consume up to 20% of the systems resources.