This paper presents our design and experience with a microkernel-inspired approach to host networking called Snap. Snap is a userspace networking system that supports Google's rapidly evolving needs with flexible modules that implement a range of network functions, including edge packet switching, virtualization for our cloud platform, traffic shaping policy enforcement, and a high-performance reliable messaging and RDMA-like service. Snap has been running in production for over three years, supporting the extensible communication needs of several large and critical systems.

The set-top and portable device market continues to grow, as does the demand for more performance under increasing cost, power, and thermal constraints. The integration of Graphics Processing Units (GPUs) into these devices and the emergence of general-purpose computations on graphics hardware enable a new set of highly parallel applications. In this paper, we propose and make the case for a GPU multitasking technique called spatial multitasking. Traditional GPU multitasking techniques, such as cooperative and preemptive multitasking, partition GPU time among applications, while spatial multitasking allows GPU resources to be partitioned among multiple applications simultaneously. We demonstrate the potential benefits of spatial multitasking with an analysis and characterization of General-Purpose GPU (GPGPU) applications. We find that many GPGPU applications fail to utilize available GPU resources fully, which suggests the potential for significant performance benefits using spatial multitasking instead of, or in combination with, preemptive or cooperative multitasking. We then implement spatial multitasking and compare it to cooperative multitasking using simulation. We evaluate several heuristics for partitioning GPU stream multiprocessors (SMs) among applications and find spatial multitasking shows an average speedup of up to 1.19 over cooperative multitasking when two applications are sharing the GPU. Speedups are even higher when more than two applications are sharing the GPU.

Evaluating the performance of reconfigurable computing applications in multi-tasking systems using simulation (as can be needed in early design-space exploration) faces several challenges. The complexity of full-system, cycle-accurate simulation prevents executing applications of any appreciable size to completion. One must sample only a portion of execution; yet unless care is taken, the measured performance for the sampled interval will not be indicative of the complete execution. Although this is generally a problem for simulation-based evaluation, the problem is exacerbated for multi-tasking systems. This paper therefore presents work to develop a performance evaluation methodology that accurately measures hybrid (both hardware and software) application performance, accounts for additional overhead introduced by hybrid resource management (such as run-time allocation of reconfigurable hardware), and correctly compensates for momentary imbalances in processor time allocation that are only artifacts of the (necessarily) short simulated execution timespan and would balance out over time.

Performance evaluation of hybrid (heterogeneous ISA) computing systems faces three major challenges: hybrid execution, multi-tasking, and system-level simulation variation. To evaluate system-level design decisions, a metric must encompass all forms of execution in a system, and incorporate any overheads introduced by hybrid execution. Differences in relative application speedups in a multi-tasking system complicate overall system performance evaluation. In full-system simulation, the relatively limited time-span for feasible tests compounds the evaluation problem. This paper discusses these challenges and presents metrics that address them.

Sensor coverage is a fundamental sensor networking design and use issue that in general tries to answer the questions about the quality of sensing (surveillance) that a particular sensor network provides. Although isotropic sensor models and coverage formulations have been studied and analyzed in great depth recently, the obtained results do not easily extend to, and address the coverage of directional and field-of-view sensors such as imagers and video cameras. In this paper, we present an optimal polynomial time algorithm for computing the worst-case breach coverage in sensor networks that are comprised of directional "field-of-view" (FOV) sensors. Given a region covered by video cameras, a direct application of the presented algorithm is to compute "breach", which is defined as the maximal distance that any hostile target can maintain from the sensors while traversing through the region. Breach translates to "worst-case coverage" by assuming that in general, targets are more likely to be detected and observed when they are closer to the sensors (while in the field of view). The approach is amenable to the inclusion of any sensor detection model that is either independent of, or inversely proportional to distance from the targets. Although for the sake of discussion we mainly focus on square fields and model the sensor FOV as an isosceles triangle, we also discuss how the algorithm can trivially be extended to deal with arbitrary polygonal field boundaries and sensor FOVs, even in the presence of rigid obstacles. We also present several simulation-based studies of the scaling issues in such coverage problems and analyze the statistical properties of breach and its sensitivity to node density, locations, and orientations. A simple grid-based approximation approach is also analyzed for comparison and validation of the implementation