As FPGAs gain popularity for on-demand application acceleration in data center computing, dynamic partial reconfiguration (DPR) has become an effective fine-grained sharing technique for FPGA multiplexing. However, current FPGA sharing encounters partial reconfiguration contention and task execution blocking problems introduced by the DPR, which significantly degrade application performance. In this paper, we propose VersaSlot, an efficient spatio-temporal FPGA sharing system with novel Big.Little slot architecture that can effectively resolve the contention and task blocking while improving resource utilization. For the heterogeneous Big.Little architecture, we introduce an efficient slot allocation and scheduling algorithm, along with a seamless cross-board switching and live migration mechanism, to maximize FPGA multiplexing across the cluster. We evaluate the VersaSlot system on an FPGA cluster composed of the latest Xilinx UltraScale+ FPGAs (ZCU216) and compare its performance against four existing scheduling algorithms. The results demonstrate that VersaSlot achieves up to 13.66x lower average response time than the traditional temporal FPGA multiplexing, and up to $2.19 x$ average response time improvement over the state-of-the-art spatio-temporal sharing systems. Furthermore, VersaSlot enhances the LUT and FF resource utilization by 35% and 29% on average, respectively.

Federated Learning (FL) is an emerging machine learning paradigm that enables the collaborative training of a shared global model across distributed clients while keeping the data decentralized. Recent works on designing systems for efficient FL have shown that utilizing serverless computing technologies, particularly Function-as-a-Service (FaaS) for FL, can enhance resource efficiency, reduce training costs, and alleviate the complex infrastructure management burden on data holders. However, current serverless FL systems still suffer from the presence of stragglers, i.e., slow clients that impede the collaborative training process. While strategies aimed at mitigating stragglers in these systems have been proposed, they overlook the diverse hardware resource configurations among FL clients. To this end, we present Apodotiko, a novel asynchronous training strategy designed for serverless FL. Our strategy incorporates a scoring mechanism that evaluates each client’s hardware capacity and dataset size to intelligently prioritize and select clients for each training round, thereby minimizing the effects of stragglers on system performance. We comprehensively evaluate Apodotiko across diverse datasets, considering a mix of CPU and GPU clients, and compare its performance against five other FL training strategies. Results from our experiments demonstrate that Apodotiko outperforms other FL training strategies, achieving an average speedup of 2.75x and a maximum speedup of 7.03x. Furthermore, our strategy significantly reduces cold starts by a factor of four on average, demonstrating suitability in serverless environments.

Serverless computing (FaaS) has been extensively utilized for deep learning (DL) inference due to the ease of deployment and pay-per-use benefits. However, existing FaaS platforms utilize GPUs in a coarse manner for DL inferences, without taking into account spatio-temporal resource multiplexing and isolation, which results in severe GPU under-utilization, high usage expenses, and SLO (Service Level Objectives) violation. There is an imperative need to enable an efficient and SLO-aware GPU-sharing mechanism in serverless computing to facilitate cost-effective DL inferences. In this paper, we propose FaST-GShare, an efficient FaaS-oriented Spatio-Temporal GPU Sharing architecture for deep learning inferences. In the architecture, we introduce the FaST-Manager to limit and isolate spatio-temporal resources for GPU multiplexing. In order to realize function performance, the automatic and flexible FaST-Profiler is proposed to profile function throughput under various resource allocations. Based on the profiling data and the isolation mechanism, we introduce the FaST-Scheduler with heuristic auto-scaling and efficient resource allocation to guarantee function SLOs. Meanwhile, FaST-Scheduler schedules function with efficient GPU node selection to maximize GPU usage. Furthermore, model sharing is exploited to mitigate memory contention. Our prototype implementation on the OpenFaaS platform and experiments on MLPerf-based benchmark prove that FaST-GShare can ensure resource isolation and function SLOs. Compared to the time sharing mechanism, FaST-GShare can improve throughput by 3.15x, GPU utilization by 1.34x, and SM (Streaming Multiprocessor) occupancy by 3.13x on average.

Federated Learning (FL) is a machine learning paradigm that enables the training of a shared global model across distributed clients while keeping the training data local. While most prior work on designing systems for FL has focused on using stateful always running components, recent work has shown that components in an FL system can greatly benefit from the usage of serverless computing and Function-as-a-Service technologies. To this end, distributed training of models with severless FL systems can be more resource-efficient and cheaper than conventional FL systems. However, serverless FL systems still suffer from the presence of stragglers, i.e., slow clients due to their resource and statistical heterogeneity. While several strategies have been proposed for mitigating stragglers in FL, most methodologies do not account for the particular characteristics of serverless environments, i.e., cold-starts, performance variations, and the ephemeral stateless nature of the function instances. Towards this, we propose FedLesScan, a novel clustering-based semi-asynchronous training strategy, specifically tailored for serverless F L. FedLesScan dynamically adapts to the behavior of clients and minimizes the effect of stragglers on the overall system. We implement our strategy by extending an open-source serverless FL system called FedLess. Moreover, we comprehensively evaluate our strategy using the 2ndgeneration Google Cloud Functions with four datasets and varying percentages of stragglers. Results from our experiments show that compared to other approaches FedLesScan reduces training time and cost by an average of 8% and 20% respectively while utilizing clients better with an average increase in the effective update ratio of 17.75%.

The traditional cloud-centric approach for Deep Learning (DL) requires training data to be collected and processed at a central server which is often challenging in privacy-sensitive domains like healthcare. Towards this, a new learning paradigm called Federated Learning (FL) has been proposed that brings the potential of DL to these domains while addressing privacy and data ownership issues. FL enables clients to learn a shared ML model while keeping the data local. However, conventional FL systems face challenges such as scalability, complex infrastructure management, and wasted compute and incurred costs due to idle clients. These challenges of FL systems closely align with the core problems that serverless computing and Function-as-a-Service (FaaS) platforms aim to solve. These include rapid scalability, no infrastructure management, automatic scaling to zero for idle clients, and a pay-per-use billing model. To this end, we present a novel system and framework for serverless FL, called FedLess. Our system supports multiple commercial and self-hosted FaaS providers and can be deployed in the cloud, on-premise in institutional data centers, and on edge devices. To the best of our knowledge, we are the first to enable FL across a large fabric of heterogeneous FaaS providers while providing important features like security and Differential Privacy. We demonstrate with comprehensive experiments that the successful training of DNNs for different tasks across up to 200 client functions and more is easily possible using our system. Furthermore, we demonstrate the practical viability of our methodology by comparing it against a traditional FL system and show that it can be cheaper and more resource-efficient.

Active LiDAR and stereo vision are the most commonly used depth sensing techniques in autonomous vehicles. Each of them alone has weaknesses in terms of density and reliability and thus cannot perform well on all practical scenarios. Recent works use deep neural networks (DNNs) to exploit their complementary properties, achieving a superior depth-sensing. However, these state-of-the-art solutions are not satisfactory on real-time responsiveness due to the high computational complexities of DNNs. In this paper, we present FastFusion, a fast deep stereo-LiDAR fusion framework for real-time high-precision depth estimation. FastFusion provides an efficient two-stage fusion strategy that leverages binary neural network to integrate stereo-LiDAR information as input and use cross-based LiDAR trust aggregation to further fuse the sparse LiDAR measurements in the back-end of stereo matching. More importantly, we present a GPU-based acceleration framework for providing a low latency implementation of FastFusion, gaining both accuracy improvement and real-time responsiveness. In the experiments, we demonstrate the effectiveness and practicability of FastFusion, which obtains a significant speedup over state-of-the-art baselines while achieving comparable accuracy on depth sensing.

Recent approaches for 3D object detection have made tremendous progresses due to the development of deep learning. However, previous researches are mostly based on individual frames, leading to limited exploitation of information between frames. In this paper, we attempt to leverage the temporal information in streaming data and explore 3D streaming based object detection as well as tracking. Toward this goal, we set up a dual-way network for 3D object detection based on keyframes, and then propagate predictions to non-key frames through a motion based interpolation algorithm guided by temporal information. Our framework is not only shown to have significant improvements on object detection compared with frame-by-frame paradigm, but also proven to produce competitive results on KITTI Object Tracking Benchmark, with 76.68% in MOTA and 81.65% in MOTP respectively.

Precise segmentation of road areas using cheap Lidar is a tough and critical task due to data sparsity problem. With sparse point clouds, reliable perception of environment is difficult due to the lack of available information and loss of object features. This paper presents a new approach to use sparse 3D Lidar data for road segmentation by fusing multiple frames of point cloud. With registration of multiple frames into a same coordinate system, reliable data can be provided for later ground segmentation and edge extraction. The accuracies of extensive experiments on three kinds of roads demonstrate that the proposed approach obtains high precision and reliability.