The integration of graphics processing units (GPUs) on high-end compute nodes has established a new accelerator-based heterogeneous computing model, which now permeates high-performance computing. The same paradigm nevertheless has limited adoption in cloud computing or other large-scale distributed computing paradigms. Heterogeneous computing with GPUs can benefit the Cloud by reducing operational costs and improving resource and energy efficiency. However, such a paradigm shift would require effective methods for virtualizing GPUs, as well as other accelerators. In this survey article, we present an extensive and in-depth survey of GPU virtualization techniques and their scheduling methods. We review a wide range of virtualization techniques implemented at the GPU library, driver, and hardware levels. Furthermore, we review GPU scheduling methods that address performance and fairness issues between multiple virtual machines sharing GPUs. We believe that our survey delivers a perspective on the challenges and opportunities for virtualization of heterogeneous computing environments.

/pdf/gpu-virtualization-and-scheduling-methods-a-comprehensive-1o9s4xu75a.pdf

GPU Virtualization and Scheduling Methods: A Comprehensive Survey

Recent work on real-world graph analytics has sought to leverage the massive amount of parallelism offered by GPU devices, but challenges remain due to the inherent irregularity of graph algorithms and limitations in GPU-resident memory for storing large graphs. We present GraphReduce, a highly efficient and scalable GPU-based framework that operates on graphs that exceed the device's internal memory capacity. GraphReduce adopts a combination of edge- and vertex-centric implementations of the Gather-Apply-Scatter programming model and operates on multiple asynchronous GPU streams to fully exploit the high degrees of parallelism in GPUs with efficient graph data movement between the host and device. GraphReduce-based programming is performed via device functions that include gatherMap, gatherReduce, apply, and scatter, implemented by programmers for the graph algorithms they wish to realize. Extensive experimental evaluations for a wide variety of graph inputs and algorithms demonstrate that GraphReduce significantly outperforms other competing out-of-memory approaches.

GraphReduce: processing large-scale graphs on accelerator-based systems

Deep Neural Networks (DNNs) have reinvigorated real-world applications that rely on learning patterns of data and are permeating into different industries and markets. Cloud infrastructure and accelerators that offer INFerence-as-a-Service (INFaaS) have become the enabler of this rather quick and invasive shift in the industry. To that end, mostly accelerator-based INFaaS (Google’s TPU [1], NVIDIA T4 [2], Microsoft Brainwave [3], etc.) has become the backbone of many real-life applications. However, as the demand for such services grows, merely scaling-out the number of accelerators is not economically cost-effective. Although multi-tenancy has propelled datacenter scalability, it has not been a primary factor in designing DNN accelerators due to the arms race for higher speed and efficiency. This paper sets out to explore this timely requirement of multi-tenancy through a new dimension: dynamic architecture fission. To that end, we define Planaria1 that can dynamically fission (break) into multiple smaller yet full-fledged DNN engines at runtime. This microarchitectural capability enables spatially co-locating multiple DNN inference services on the same hardware, offering simultaneous multi-tenant DNN acceleration. To realize this dynamic reconfigurability, we first devise breakable omni-directional systolic arrays for DNN acceleration that allows omni-directional flow of data. Second, it uses this capability and a unique organization of on-chip memory, interconnection, and compute resources to enable fission in systolic array based DNN accelerators. Architecture fission and its associated flexibility enables an extra degree of freedom for task scheduling, that even allows breaking the accelerator with regard to the server load, DNN topology, and task priority. As such, it can simultaneously co-locate DNNs to enhance utilization, throughput, QoS, and fairness. We compare the proposed design to PREMA [4], a recent effort that offers multi-tenancy by time-multiplexing the DNN accelerator across multiple tasks. We use the same frequency, the same amount of compute and memory resources for both accelerators. The results show significant benefits with (soft, medium, hard) QoS requirements, in throughput (7.4×, 7.2×, 12.2×), SLA satisfaction rate (45%, 15%, 16%), and fairness (2.1×, 2.3×, 1.9×).

Planaria: Dynamic Architecture Fission for Spatial Multi-Tenant Acceleration of Deep Neural Networks

The massive explosion in social networks has led to a significant growth in graph analytics and specifically in dynamic, time-varying graphs. Most prior work processes dynamic graphs by first storing the updates and then repeatedly running static graph analytics on saved snapshots. To handle the extreme scale and fast evolution of real-world graphs, we propose a dynamic graph analytics framework, GraphIn, that incrementally processes graphs on-the-fly using fixed-sized batches of updates. As part of GraphIn, we propose a novel programming model called I-GAS based on gather-apply-scatter programming paradigm that allows for implementing a large set of incremental graph processing algorithms seamlessly across multiple CPU cores. We further propose a property-based, dual-path execution model to choose between incremental or static computation. Our experiments show that for a variety of graph inputs and algorithms, GraphIn achieves upi?źto 9.3 million updates/sec and over 400$$\times $$ speedup when compared to static graph recomputation.

GraphIn: An Online High Performance Incremental Graph Processing Framework

GPUs have become the primary choice of accelerators for high-end data centers and cloud servers, which can host thousands of disparate applications. With the growing demands for GPUs on clusters, there arises a need for efficient co-execution of applications on the same accelerator device. However, the resource contention among co-executing applications causes interference which leads to degradation in execution performance, impacts QoS requirements of applications and lowers overall system throughput. While previous work has proposed techniques for detecting interference, the existing solutions are either developed for CPU clusters, or use static profiling approaches which can be computationally intensive and do not scale well. We present Mystic, an interference-aware scheduler for efficient co-execution of applications on GPU-based clusters and cloud servers. The most important feature of Mystic is the use of learning-based analytical models for detecting interference between applications. We leverage a collaborative filtering framework to characterize an incoming application with respect to the interference it may cause when co-executing with other applications while sharing GPU resources. Mystic identifies the similarities between new applications and the executing applications, and guides the scheduler to minimize the interference and improve system throughput. We train the learning model with 42 CUDA applications, and consider another separate set of 55 diverse, real-world GPU applications for evaluation. Mystic is evaluated on a live GPU cluster with 32 NVIDIA GPUs. Our framework achieves performance guarantees for 90.3% of the evaluated applications. When compared with state-of-the art interference-oblivious schedulers, Mystic improves the system throughput by 27.5% on average, and achieves a 16.3% improvement on average in GPU utilization.

Mystic: Predictive Scheduling for GPU Based Cloud Servers Using Machine Learning

Accelerator-based systems are making rapid inroads into becoming platforms of choice for high end cloud services. There is a need therefore, to move from the current model in which high performance applications explicitly and programmatically select the GPU devices on which to run, to a dynamic model where GPUs are treated as first class schedulable entities. The Strings scheduler realizes this vision by decomposing the GPU scheduling problem into a combination of load balancing and per-device scheduling. (i) Device-level scheduling efficiently uses all of a GPU's hardware resources, including its computational and data movement engines, and (ii) load balancing goes beyond obtaining high throughput, to ensure fairness through prioritizing GPU requests that have attained least service. With its methods, Strings achieves improvements in system throughput and fairness of up to 8.70x and 13%, respectively, compared to the CUDA runtime.

Krishna Pallavi

Papers

Scheduling multi-tenant cloud workloads on accelerator-based systems