Unmanned Aerial Vehicles (UAVs) or drones are increasingly used for urban applications like traffic monitoring and construction surveys. Autonomous navigation allows drones to visitwaypointsand accomplishactivitiesas part of theirmission. A common activity is to hover and observe a location using on-board cameras. Advances in Deep Neural Networks (DNNs) allow such videos to be analyzed for automated decision making. UAVs also host edge computing capability for on-board inferencing by such DNNs. To this end, for a fleet of drones, we propose a novelMission Scheduling Problem ()that co-schedules the flight routes to visit and record video at waypoints, and their subsequent on-board edge analytics. The proposed schedule maximizes the data capture and computing utilities from the activities while meeting the activity deadlines, and the energy and computing constraints. We first prove that is NP-hard and then optimally solve it by formulating a mixed integer linear programming (MILP) problem. Next, we design five time-efficient heuristic algorithms that provide sub-optimal but fast solutions that are empirically competitive with the optimal solution. Evaluation of these five schedulers using real drone traces demonstrate utility–runtime trade-offs under diverse workloads.

Functions as a Service (FaaS) have gained popularity for programming public clouds due to their simple abstraction, ease of deployment, effortless scaling and granular billing. Cloud providers also offer basic capabilities to compose these functions into workflows. FaaS and FaaS workflow models, however, are proprietary to each cloud provider. This prevents their portability across cloud providers, and requires effort to design workflows that run on different cloud providers or data centers. Such requirements are increasingly important to meet regulatory requirements, leverage cost arbitrage and avoid vendor lock-in. Further, the FaaS execution models are also different, and the overheads of FaaS workflows due to message indirection and cold-starts need custom optimizations for different platforms. In this paper, we propose XFaaS, a cross-platform deployment and orchestration engine for FaaS workflows to operate on multiple clouds. XFaaS allows “zero touch” deployment of functions and workflows across AWS and Azure clouds by automatically generating the necessary code wrappers, cloud queues, and coordinating with the native FaaS engine of the cloud providers. It also uses intelligent function fusion and placement logic to reduce the workflow execution latency in a hybrid cloud while mitigating costs, using performance and billing models specific to the providers based in detailed benchmarks. Our empirical results indicate that fusion offers up to ≈75 % benefits in latency and ≈57% reduction in cost, while placement strategies reduce the latency by ≈ 24%, compared to baselines in the best cases.

Internet of Things (loT) is leading to the pervasive availability of streaming data about the physical world, coupled with edge computing infrastructure deployed as part of smart cities and 5G rollout. These constrained, less reliable but cheap resources are complemented by fog resources that offer feder-ated management and accelerated computing, and pay-as-you-go cloud resources. There is a lack of intuitive means to deploy application pipelines to consume such diverse streams, and to execute them reliably on edge and fog resources. We propose an innovative application model to declaratively specify queries to match streams of micro-batch data from stream sources and trigger the distributed execution of data pipelines. We also design a resilient scheduling strategy using advanced reservation on reliable fogs to guarantee dataflow completion within a deadline while minimizing the execution cost. Our detailed experiments on over 100 virtual loT resources and for$\approx 10k$task executions, with comparison against baseline scheduling strategies, illustrates the cost-effectiveness, resilience and scalability of our framework.

Serverless computing and FaaS have gained popularity due to their ease of design, deployment, scaling and billing on clouds. However, when used to compose and orchestrate scientific workflows, they pose limitations due to cold starts, message indirection, vendor lock-in and lack of provenance support. Here, we propose a design for a Ser verless Scientific Workflow Orchestrator that overcomes these challenges using techniques like function fusion, pilot invocations and data fabrics.

Unmanned Aerial Vehicles (UAVs) or drones are increasingly used for urban applications like traffic monitoring and construction surveys. Autonomous navigation allows drones to visit waypoints and accomplish activities as part of their mission. A common activity is to hover and observe a location using on-board cameras. Advances in Deep Neural Networks (DNNs) allow such videos to be analyzed for automated decision making. UAVs also host edge computing capability for on-board inferencing by such DNNs. To this end, for a fleet of drones, we propose a novel Mission Scheduling Problem (MSP) that co-schedules the flight routes to visit and record video at waypoints, and their subsequent on-board edge analytics. The proposed schedule maximizes the utility from the activities while meeting activity deadlines as well as energy and computing constraints. We first prove that MSP is NP-hard and then optimally solve it by formulating a mixed integer linear programming (MILP) problem. Next, we design two efficient heuristic algorithms, jsc and vrc, that provide fast sub-optimal solutions. Evaluation of these three schedulers using real drone traces demonstrate utility-runtime trade-offs under diverse workloads.

Advances in deep neural networks (DNN) and computer vision (CV) algorithms have made it feasible to extract meaningful insights from large-scale deployments of urban cameras. Tracking an object of interest across the camera network in near real-time is a canonical problem. However, current tracking platforms have two key limitations: 1) They are monolithic, proprietary and lack the ability to rapidly incorporate sophisticated tracking models, and 2) They are less responsive to dynamism across wide-area computing resources that include edge, fog, and cloud abstractions. We address these gaps using Anveshak, a runtime platform for composing and coordinating distributed tracking applications. It provides a domain-specific dataflow programming model to intuitively compose a tracking application, supporting contemporary CV advances like query fusion and re-identification, and enabling dynamic scoping of the camera network's search space to avoid wasted computation. We also offer tunable batching and data-dropping strategies for dataflow blocks deployed on distributed resources to respond to network and compute variability. These balance the tracking accuracy, its real-time performance, and the active camera-set size. We illustrate the concise expressiveness of the programming model for four tracking applications. Our detailed experiments for a network of 1000 camera-feeds on modest resources exhibit the tunable scalability, performance, and quality trade-offs enabled by our dynamic tracking, batching, and dropping strategies.

Smart City deployments typically have thousands to even hundreds of thousands of Surveillance cameras. Rapid advancements in computer vision techniques due to Deep Neural Networks enable using these camera feeds for performing non-trivial analytics. Tracking a moving object of interest using a large network of cameras, also known as object reidentification, is one such analytic that empowers city administration with capabilities such as finding missing people or prioritizing emergency vehicles. We have built Anveshak, a framework for distributed wide-area tracking. Anveshak fills in the shortcomings of existing Big Data and Deep Learning frameworks by - exposing an intuitive and composable programming model; automating application deployment and orchestration across edge, fog and cloud resources and providing knobs to the user for managing the application performance. The knobs lend the application the ability to scale potentially to thousands of cameras. In this proposal we have designed two representative applications; missing person tracking and priority signalling for emergency vehicles. We empirically verify that the application scales to 1000 cameras on a Cloud-only deployment of 10 Azure VMs with 8 cores and 32GB RAM each. Alternatively, it scales to 500 cameras on a simulated setup of 100 edge, 30 fog, and 1 Cloud VM. We also highlight the effect of the knobs on the application performance. designed two representative applications; missing person tracking and priority signalling for emergency vehicles. We empirically verify that the application scales to 1000 cameras on a Cloud-only deployment of 10 Azure VMs with 8 cores and 32GB RAM each. Alternatively, it scales to 500 cameras on a simulated setup of 100 edge, 30 fog, and 1 Cloud VM. We also highlight the effect of the knobs on the application performance.