DeepAI AI Chat
Log In Sign Up

Lightning: Scaling the GPU Programming Model Beyond a Single GPU

by   Stijn Heldens, et al.
eScience Center

The GPU programming model is primarily aimed at the development of applications that run one GPU. However, this limits the scalability of GPU code to the capabilities of a single GPU in terms of compute power and memory capacity. To scale GPU applications further, a great engineering effort is typically required: work and data must be divided over multiple GPUs by hand, possibly in multiple nodes, and data must be manually spilled from GPU memory to higher-level memories. We present Lightning: a framework that follows the common GPU programming paradigm but enables scaling to large problems with ease. Lightning supports multi-GPU execution of GPU kernels, even across multiple nodes, and seamlessly spills data to higher-level memories (main memory and disk). Existing CUDA kernels can easily be adapted for use in Lightning, with data access annotations on these kernels allowing Lightning to infer their data requirements and the dependencies between subsequent kernel launches. Lightning efficiently distributes the work/data across GPUs and maximizes efficiency by overlapping scheduling, data movement, and kernel execution when possible. We present the design and implementation of Lightning, as well as experimental results on up to 32 GPUs for eight benchmarks and one real-world application. Evaluation shows excellent performance and scalability, such as a speedup of 57.2x over the CPU using Lighting with 16 GPUs over 4 nodes and 80 GB of data, far beyond the memory capacity of one GPU.


page 1

page 2

page 3

page 4


Protecting real-time GPU kernels on integrated CPU-GPU SoC platforms

Integrated CPU-GPU architecture provides excellent acceleration capabili...

Effective GPU Sharing Under Compiler Guidance

Modern computing platforms tend to deploy multiple GPUs (2, 4, or more) ...

Cooperative Kernels: GPU Multitasking for Blocking Algorithms (Extended Version)

There is growing interest in accelerating irregular data-parallel algori...

Targeting GPUs with OpenMP Directives on Summit: A Simple and Effective Fortran Experience

We use OpenMP directives to target hardware accelerators (GPUs) on Summi...

AMOEBA: A Coarse Grained Reconfigurable Architecture for Dynamic GPU Scaling

Different GPU applications exhibit varying scalability patterns with net...

A Simple Model for Portable and Fast Prediction of Execution Time and Power Consumption of GPU Kernels

Characterizing compute kernel execution behavior on GPUs for efficient t...

Leveraging GPU batching for scalable nonlinear programming through massive Lagrangian decomposition

We present the implementation of a trust-region Newton algorithm ExaTron...