Energy-Efficient Thermal-Aware Autonomic Management of Virtualized HPC Cloud Infrastructure

Virtualized datacenters and clouds are being increasingly considered for traditional High-Performance Computing (HPC) workloads that have typically targeted Grids and conventional HPC platforms. However, maximizing energy efficiency and utilization of datacenter resources, and minimizing undesired thermal behavior while ensuring application performance and other Quality of Service (QoS) guarantees for HPC applications requires careful consideration of important and extremely challenging tradeoffs. Virtual Machine (VM) migration is one of the most common techniques used to alleviate thermal anomalies (i.e., hotspots) in cloud datacenter servers as it reduces load and, hence, the server utilization. In this article, the benefits of using other techniques such as voltage scaling and pinning (traditionally used for reducing energy consumption) for thermal management over VM migrations are studied in detail. As no single technique is the most efficient to meet temperature/performance optimization goals in all situations, an autonomic approach that performs energy-efficient thermal management while ensuring the QoS delivered to the users is proposed. To address the problem of VM allocation that arises during VM migrations, an innovative application-centric energy-aware strategy for Virtual Machine (VM) allocation is proposed. The proposed strategy ensures high resource utilization and energy efficiency through VM consolidation while satisfying application QoS by exploiting knowledge obtained through application profiling along multiple dimensions (CPU, memory, and network bandwidth utilization). To support our arguments, we present the results obtained from an experimental evaluation on real hardware using HPC workloads under different scenarios.     

A platform for the development of patient applications in the domain of personalized health

Personalized health (p-health) systems can contribute significantly to the sustainability of healthcare systems, though their feasibility is yet to be proven. One of the problems related to their development is the lack of well-established development tools for this domain. As the p-health paradigm is focused on patient self-management, big challenges arise around the design and implementation of patient systems. This paper presents a reference platform created for the development of these applications, and shows the advantages of its adoption in a complex project dealing with cardio-vascular diseases.     

The multi-modal interface of Robot-Era multi-robot services tailored for the elderly

Socially assistive robotic platforms are now a realistic option for the long-term care of ageing populations. Elderly users may benefit from many services provided by robots operating in different environments, such as providing assistance inside apartments, serving in shared facilities of buildings or guiding people outdoors. In this paper, we present the experience gained within the EU FP7 ROBOT-ERA project towards the objective of implementing easy-to-use and acceptable service robotic system for the elderly. In particular, we detail the user-centred design and the experimental evaluation in realistic environments of a web-based multi-modal user interface tailored for elderly users of near future multi-robot services. Experimental results demonstrate positive evaluation of usability and willingness to use by elderly users, especially those less experienced with technological devices who could benefit more from the adoption of robotic services. Further analyses showed how multi-modal modes of interaction support more flexible and natural elderly–robot interaction, make clear the benefits for the users and, therefore, increase its acceptability. Finally, we provide insights and lessons learned from the extensive experimentation, which, to the best of our knowledge, is one of the largest experimentation of a multi-robot multi-service system so far.     

Polychaete-Like Undulatory Robotic Locomotion in Unstructured Substrates

A biological paradigm of versatile locomotion and effective motion control is provided by the polychaete annelid worms, whose motion adapts to a large variety of unstructured environmental conditions (sand, mud, sediment, water, etc.), and could thus be of interest to replicate by robotic analogs. Their locomotion is characterized by the combination of a unique form of tail-to-head body undulations (opposite to snakes and eels), with the rowing-like action of numerous lateral appendages distributed along their long segmented body. Focusing on the former aspect of polychaete locomotion, computational models of crawling and swimming by such tail-to-head body undulations have been developed in this paper. These are based on the Lagrangian dynamics of the system and on resistive models of its interaction with the environment, and are used for simulation studies demonstrating the generation of undulatory gaits. Several biomimetic robotic prototypes have been developed, whose undulatory actuation achieves propulsion on sand and other granular unstructured environments. Extensive experimental studies demonstrate the feasibility of robot propulsion by tail-to-head body undulations in such environments, as well as the agreement of its qualitative and quantitative characteristics to the predictions of the corresponding computational models.     

Lane-change detection using a computational driver model

OBJECTIVE: This paper introduces a robust, real-time system for detecting driver lane changes. Background: As intelligent transportation systems evolve to assist drivers in their intended behaviors, the systems have demonstrated a need for methods of inferring driver intentions and detecting intended maneuvers. METHOD: Using a "model tracing" methodology, our system simulates a set of possible driver intentions and their resulting behaviors using a simplification of a previously validated computational model of driver behavior. The system compares the model's simulated behavior with a driver's actual observed behavior and thus continually infers the driver's unobservable intentions from her or his observable actions. RESULTS: For data collected in a driving simulator, the system detects 82% of lane changes within 0.5 s of maneuver onset (assuming a 5% false alarm rate), 93% within 1 s, and 95% before the vehicle moves one fourth of the lane width laterally. For data collected from an instrumented vehicle, the system detects 61% within 0.5 s, 77% within 1 s, and 84% before the vehicle moves one-fourth of the lane width laterally. CONCLUSION: The model-tracing system is the first system to demonstrate high sample-by-sample accuracy at low false alarm rates as well as high accuracy over the course of a lane change with respect to time and lateral movement. APPLICATION: By providing robust real-time detection of driver lane changes, the system shows good promise for incorporation into the next generation of intelligent transportation systems.     

Dealing with midair collisions in dense collective aerial systems

The idea of creating collective aerial systems is appealing because several rather simple flying vehicles could join forces to cover a large area in little time in applications such as monitoring, mapping, search and rescue, or airborne communication relays. In most of these scenarios, a fleet of cooperating vehicles is dispatched to a confined airspace area and requested to fly close to a nominal altitude. Moreover, depending on the task each vehicle is assigned to, individual flight trajectories in this essentially two‐dimensional space may interfere, resulting in disastrous collisions. This paper begins by introducing a probabilistic model to predict the rate of midair collisions that would occur if nothing is done to prevent them. In a second step, a control strategy for midair collision avoidance is proposed, which is interesting because it requires only local communication and information about flight altitudes. The proposed strategy is systematically analyzed in theory and simulation as well as in experiments with five physical aerial vehicles. A significant reduction in collision rates can be achieved. Statistically, values close to zero are possible when the swarm's density is below an application‐dependent threshold. Such low collision rates warrant an acceptable level of confidence in collision‐free operation of a physical swarm. © 2011 Wiley Periodicals, Inc.     

Privacy in geo-social networks: proximity notification with untrusted service providers and curious buddies

A major feature of the emerging geo-social networks is the ability to notify a user when any of his friends (also called buddies) happens to be geographically in proximity. This proximity service is usually offered by the network itself or by a third party service provider (SP) using location data acquired from the users. This paper provides a rigorous theoretical and experimental analysis of the existing solutions for the location privacy problem in proximity services. This is a serious problem for users who do not trust the SP to handle their location data and would only like to release their location information in a generalized form to participating buddies. The paper presents two new protocols providing complete privacy with respect to the SP and controllable privacy with respect to the buddies. The analytical and experimental analysis of the protocols takes into account privacy, service precision, and computation and communication costs, showing the superiority of the new protocols compared to those appeared in the literature to date. The proposed protocols have also been tested in a full system implementation of the proximity service.     

Enhancing a dependable multiserver operating system with temporal protection via resource reservations

Nowadays, microkernel-based systems are getting studied and adopted with a renewed interest in a wide number of IT scenarios. Their advantages over classical monolithic solutions mainly concern the dependability domain. By being capable of dynamically detect and solve non-expected behaviours within its core components, a microkernel-based OS would eventually run forever with no need to be restarted. Dependability in this context mainly aims at isolating components from a spatial point of view: a microkernel-based system may definitely not be adopted in the context of real-time environments, simply basing on this kind of protection only. One of the most active real-time research areas concerns adding temporal protection mechanisms to general purpose operating systems. By making use of such mechanisms, these systems become suitable for being adopted in the context of time-sensitive domains. Microkernel-based systems have always been thought of as a kind of platform not suited to real-time contexts, due to the high latencies introduced by the message passing technique as the only inter-process communication (IPC) facility within the system. With computer performances growing at a fairly high rate, this overhead becomes negligible with respect to the typical real-time processing times. In the last years, many algorithms belonging to the class of the so-called Resource Reservations (RRES) have been devised in order to provide the systems with the needed temporal isolation. By introducing a RRES-aware scheduler in the context of a microkernel-based system, we may enrich it with the temporal benefits it needs in order to be deployed within domains with real-time requirements. In this paper we propose a generic way to implement these mechanisms, dependent for a very small part on the underlying OS mechanisms. In order to show the generality of our RRES framework we implemented it in the context of Minix 3, a highly dependable microkernel-based OS with an impressive users base.     

An ensemble evolutionary constraint-based approach to understand the emergence of metabolic phenotypes

Constraint-based modeling is largely used in computational studies of metabolism. We propose here a novel approach that aims to identify ensembles of flux distributions that comply with one or more target phenotype(s). The methodology has been tested on a small-scale model of yeast energy metabolism. The target phenotypes describe the differential pattern of ethanol production and O₂ consumption observed in “Crabtree-positive” and “Crabtree-negative” yeasts in changing environment (i.e., when the upper limit of glucose uptake is varied). The ensembles were obtained either by selection among sampled flux distributions or by means of a search heuristic (genetic algorithm). The former approach provided indication about the probability to observe a given phenotype, but the resulting ensembles could not be unambiguously partitioned into “Crabtree-positive” and “Crabtree-negative” clusters. On the contrary well-separated clusters were obtained with the latter method. The cluster analysis further allowed identification of distinct groups within each target phenotype. The method may thus prove useful in characterizing the design principles underlying metabolic plasticity arising from evolving physio-pathological or developmental constraints.