high performance computing on graphics processing units: hgpu.org

We study the problem of semantic matching in product search, that is, given a customer query, retrieve all semantically related products from the catalog. Pure lexical matching via an inverted index falls short in this respect due to several factors: a) lack of understanding of hypernyms, synonyms, and antonyms, b) fragility to morphological variants (e.g. […]

The presented thesis establishes simulations on modern massively parallel computing hardware to investigate relativistic laser-driven plasmas. The latter are of special interest as they may provide a compact source for energetic ion beams. Computer simulations provide valuable insight into ultrafast plasma processes, evolving in the ultrahigh intensity (l_0 >> 1018 W/cm^2) focus of the ultrashort […]

Modern distributed machine learning (ML) training workloads benefit significantly from leveraging GPUs. However, significant contention ensues when multiple such workloads are run atop a shared cluster of GPUs. A key question is how to fairly apportion GPUs across workloads while ensuring overall cluster efficiency. We find that established cluster scheduling disciplines that provide instantaneous fair […]

When vectorizing programs for modern processors with SIMD extensions, IF-statements pose a challenge: existing vectorization approaches often introduce redundant computations or they resort to inefficient masked instructions. In this paper, we introduce a new notion of warp-coherence for conditions that exhibit coherent run-time behavior on different lanes of a vector register. We demonstrate that warp-coherent […]

The paper proposes the further development of algebra-algorithmic design and synthesis tools towards the development of OpenCL programs. The method for semi-automatic parallelization of cyclic operators is proposed. The particular feature of the approach consists in using high-level algebraalgorithmic program specifications (schemes) and rewriting rules technique. The developed tools provide the construction of parallel algorithm […]

A major contributor to the deployment and operational costs of a large-scale high-performance computing (HPC) clusters is the memory system. In terms of system performance it is one of the most critical aspects of the system’s design. However, next generation of HPC systems poses significant challenges for the main memory, and it is questionable whether […]

The validation and deployment of novel research ideas in the field of Deep Learning is often limited by the availability of efficient compute kernels for certain basic primitives. In particular, operations that cannot leverage existing vendor libraries (e.g., cuBLAS, cuDNN) are at risk of facing poor device utilization unless custom implementations are written by experts […]

Cloud providers usually offer diverse types of hardware for their users. Customers exploit this option to deploy cloud instances featuring GPUs, FPGAs, architectures other than x86 (e.g., ARM, IBM Power8), or featuring certain specific extensions (e.g, Intel SGX). We consider in this work the instances used by customers to deploy containers, nowadays the de facto […]

CUDA while very popular, is not as flexible with respect to target devices as OpenCL. While parallel algorithm research might address problems first with a CUDA C++ solution, those results are not easily portable to a target not directly supported by CUDA. In contrast, a SYCL C++ solution can operate on the larger variety of […]

We present HAM (Heterogeneous Active Messages), a C++-only active messaging solution for heterogeneous distributed systems.Combined with a communication protocol, HAM can be used as a generic Remote Procedure Call (RPC) mechanism. It has been used in HAM-Offload to implement a low-overhead offloading framework for inter- and intra-node offloading between different architectures including accelerators like the […]

This paper presents Mirovia, a benchmark suite developed for modern day heterogeneous computing. Previous benchmark suites such as Rodinia [1] and SHOC [2] are well written and have many desirable features. However, these tools were developed years ago when hardware was less powerful and software had fewer features. For example, unified memory was introduced in […]

Porting code from CPU to GPU is costly and time-consuming; Unless much time is invested in development and optimization, it is not obvious, a priori, how much speed-up is achievable or how much room is left for improvement. Knowing the potential speed-up a priori can be very useful: It can save hundreds of engineering hours, […]