Distributed Deep Neural Network (DDNN) training on cloud spot instances is increasingly compelling as it can significantly save the user budget. To handle unexpected instance revocations, provisioning a heterogeneous cluster using the asynchronous parallel mechanism becomes the dominant method for DDNN training with spot instances. However, blindly provisioning a cluster of spot instances can easily result in unpre-dictable DDNN training performance, mainly because bottlenecks occur on the parameter server network bandwidth and PCIe bandwidth resources, as well as the inadequate cluster heterogeneity. To address the challenges above, we propose spotDNN, a heterogeneity-aware spot instance provisioning framework that provides predictable performance for DDNN training in the cloud. By explicitly considering the contention for bottle-neck resources, we first build an analytical performance model of DDNN training in heterogeneous clusters. It leverages the weighted average batch size and convergence coefficient to quantify the DDNN training loss in heterogeneous clusters. Through a lightweight workload profiling, we further design a cost-efficient instance provisioning strategy which incorporates the bounds calculation and sliding window techniques to effectively guarantee the training performance service level objectives (SLOs). We have implemented a prototype of spotDNN and conducted extensive experiments on Amazon EC2. Experiment results show that spotDNN can deliver predictable DDNN training performance while reducing the monetary cost by up to 68.1% compared to the existing solutions, yet with acceptable runtime overhead.

GPUs are essential to accelerating the latency-sensitive deep neural network (DNN) inference workloads in cloud datacenters. To fully utilize GPU resources,spatial sharingof GPUs among co-located DNN inference workloads becomes increasingly compelling. However, GPU sharing inevitably bringssevere performance interferenceamong co-located inference workloads, as motivated by an empirical measurement study of DNN inference on EC2 GPU instances. While existing works on guaranteeing inference performance service level objectives (SLOs) focus on eithertemporal sharingof GPUs orreactiveGPU resource scaling and inference migration techniques, how toproactivelymitigate such severe performance interference has received comparatively little attention. In this paper, we proposeiGniter, aninterference-awareGPU resource provisioning framework for cost-efficiently achieving predictable DNN inference in the cloud.iGniteris comprised of two key components: (1) alightweightDNN inference performance model, which leverages the system and workload metrics that are practically accessible to capture the performance interference; (2) Acost-efficientGPU resource provisioning strategy thatjointlyoptimizes the GPU resource allocation and adaptive batching based on our inference performance model, with the aim of achieving predictable performance of DNN inference workloads. We implement a prototype ofiGniterbased on the NVIDIA Triton inference server hosted on EC2 GPU instances. Extensive prototype experiments on four representative DNN models and datasets demonstrate thatiGnitercan guarantee the performance SLOs of DNN inference workloads with practically acceptable runtime overhead, while saving the monetary cost by up to$25\%$in comparison to the state-of-the-art GPU resource provisioning strategies.

Optimizing performance for Distributed Deep Neural Network (DDNN) training has recently become increasingly compelling, as the DNN model gets complex and the training dataset grows large. While existing works on communication scheduling mostly focus on overlapping the computation and communication to improve DDNN training performance, the GPU and network resources are still under-utilized in DDNN training clusters. To tackle this issue, in this paper, we design and implement a predictable communication scheduling strategy named Prophet to schedule the gradient transfer in an adequate order, with the aim of maximizing the GPU and network resource utilization. Leveraging our observed stepwise pattern of gradient transfer start time, Prophet first uses the monitored network bandwidth and the profiled time interval among gradients to predict the appropriate number of gradients that can be grouped into blocks. Then, these gradient blocks can be transferred one by one to guarantee high utilization of GPU and network resources while ensuring the priority of gradient transfer (i.e., low-priority gradients cannot preempt high-priority gradients in the network transfer). Prophet can make the forward propagation start as early as possible so as to greedily reduce the waiting (idle) time of GPU resources during the DDNN training process. Prototype experiments with representative DNN models trained on Amazon EC2 demonstrate that Prophet can improve the DDNN training performance by up to 40% compared with the state-of-the-art priority-based communication scheduling strategies, yet with negligible runtime performance overhead.