GPU compute devices have become very popular for general purpose computations. However, the SIMD-like hardware of graphics processors is currently not well suited for irregular workloads, like searching unbalanced trees. In order to mitigate this drawback, NVIDIA introduced an extension to GPU programming models called Dynamic Parallelism. This extension enables GPU programs to spawn new units of work directly on the GPU, allowing the refinement of subsequent work items based on intermediate results without any involvement of the main CPU.This work investigates methods for employing Dynamic Parallelism with the goal of improved workload distribution for tree search algorithms on modern GPU hardware. For the evaluation of the proposed approaches, a case study is conducted on the N-Queens problem. Extensive benchmarks indicate that the benefits of improved resource utilization fail to outweigh high management overhead and runtime limitations due to the very fine level of granularity of the investigated problem. However, novel memory management concepts for passing parameters to child grids are presented. These general concepts are applicable to other, more coarse-grained problems that benefit from the use of Dynamic Parallelism.

Figure 1: With our proposed gesture output, the device outputs messages to users using the same gesture language used for input. (a) Here, the user draws an to check the house number of the upcoming meeting. (b) The device replies by translating the user's finger along the path of an . (c) The pocketOuija is one of the two force feedback touchscreen devices we built that support gesture output. It translates the user's finger by means of a transparent plastic foil overlaid onto the screen actuated using motors located on the back of the device. ABSTRACTWe propose using spatial gestures not only for input but also for output. Analogous to gesture input, the proposed gesture output moves the user's finger in a gesture, which the user then recognizes. We use our concept in a mobile scenario where a motion path forming a "5" informs users about new emails, or a heart-shaped path serves as a message from a friend. We built two prototypes: (1) The long-RangeOuija is a stationary prototype that offers a motion range of up to 4cm; (2) The pocketOuija is self-contained mobile device based on an iPhone with up to 1cm motion range. Both devices actuate the user's fingers by means of an actuated transparent foil overlaid onto a touchscreen. We conducted 3 studies on the longRangeOuija. Participants recognized 2cm marks with 97% accuracy, Graffiti digits with 98.8%, pairs of Graffiti digits with 90.5%, and Graffiti letters with 93.4%. Participants previously unfamiliar with Graffiti identified 96.2% of digits and 76.4% of letters, suggesting that properly designed gesture output is guessable. After the experiment, the same participants were able to enter 100% of Graffiti digits by heart and 92.2% of letters. This suggests that participants learned gesture input as a side effect of using gesture output on our prototypes.

The increasing prevalence of the microservice paradigm creates a new demand for low-overhead virtualization techniques. Complementing containerization, unikernels are emerging as alternative approaches. With both techniques undergoing rapid improvements, the current landscape of lightweight virtualization approaches presents a confusing scenery, complicating the task of choosing a suited technology for an intended purpose. This work provides a comprehensive performance comparison covering containers, unikernels, whole-system virtualization, native hardware, and combinations thereof. Representing common workloads in microservice-based applications, we assess application performance using HTTP servers and a key-value store. With the microservice deployment paradigm in mind, we evaluate further characteristics such as startup time, image size, network latency, and memory footprint.

No abstract

Cloud federation is receiving increasing attention due to the benefits of resilience and locality it brings to cloud providers and users. Our analysis of three diverse use cases shows that existing solutions are not addressing the federation needs of such use case applications. In this paper, we present an alternative approach to network federation, providing a model based on cloud-to-cloud agreements. In our scenarios, companies hosting their own OpenStack clouds need to run machines transparently in another cloud, provided by a company they have an agreement with. Our solution provides multiple benefits to cloud providers and users detailed in this paper. Our implementation outperforms the VPNaaS solution in OpenStack in terms of throughput.