Exploiting heterogeneous parallelism with the Heterogeneous Programming Library

Viñas, Moisés; Bozkuş, Zeki; Fraguela, Basilio B.

doi:10.1016/j.jpdc.2013.07.013

Cited by 31 publications

(49 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This has led to the development of libraries that simplify the use of this paradigm by automating many of tasks it requires and hiding from the developer several of its related concepts, while usually providing minimal overheads with respect to the straight use of OpenCL. We have developed our parallelization using one of these libraries, namely, the Heterogeneous Programming Library (HPL) [4], which has shown to provide excellent programmability for heterogeneous applications while incurring in negligible overheads. Although HPL supports writing the kernels in OpenCL C [5] our implementation relies on the embedded language introduced in [4], which is translated at runtime into OpenCL C by the library.…”

Section: Opencl Parallelizationmentioning

confidence: 99%

Accelerating the HyperLogLog Cardinality Estimation Algorithm

Bozkus

Fraguela

2017

Scientific Programming

Self Cite

View full text Add to dashboard Cite

In recent years, vast amounts of data of different kinds, from pictures and videos from our cameras to software logs from sensor networks and Internet routers operating day and night, are being generated. This has led to new big data problems, which require new algorithms to handle these large volumes of data and as a result are very computationally demanding because of the volumes to process. In this paper, we parallelize one of these new algorithms, namely, the HyperLogLog algorithm, which estimates the number of different items in a large data set with minimal memory usage, as it lowers the typical memory usage of this type of calculation from ( ) to (1). We have implemented parallelizations based on OpenMP and OpenCL and evaluated them in a standard multicore system, an Intel Xeon Phi, and two GPUs from different vendors. The results obtained in our experiments, in which we reach a speedup of 88.6 with respect to an optimized sequential implementation, are very positive, particularly taking into account the need to run this kind of algorithm on large amounts of data.

show abstract

Section: Opencl Parallelizationmentioning

confidence: 99%

Accelerating the HyperLogLog Cardinality Estimation Algorithm

Bozkus

Fraguela

2017

Scientific Programming

Self Cite

View full text Add to dashboard Cite

show abstract

“…Unfortunately, the OpenCL API is too low level [72], resulting in code verbosity and low productivity of programmers, who need to manually perform buffer allocations, data transfers, synchronizations etc. High level annotations [81,2] or embedded languages [97] built on top of OpenCL that automate these tasks and hide their complexity are a better approach. Several authors have considered the programmability of accelerators on top of OpenCL across a cluster [7,35]; however, existing solutions only abstract away some complexities of OpenCL but still expose other low-level details or restrict the non-accelerator part of an application to a single node.…”

Section: Advancing the Programming Environmentsmentioning

confidence: 99%

“…In order to exploit all the available hardware in exascale systems, we need to extend and evolve the highlevel programming of accelerators (e.g., [81,97,2]) into high-productivity seamless approaches, for instance, through the natural integration with the ADT previously described.…”

Section: Advancing the Programming Environmentsmentioning

confidence: 99%

On Processing Extreme Data

Petcu¹,

Iuhasz²,

Pop³

et al. 2016

SCPE

Self Cite

View full text Add to dashboard Cite

Abstract. Extreme Data is an incarnation of Big Data concept distinguished by the massive amounts of data that must be queried, communicated and analyzed in near real-time by using a very large number of memory or storage elements and exascale computing systems. Immediate examples are the scientific data produced at a rate of hundreds of gigabits-per-second that must be stored, filtered and analyzed, the millions of images per day that must be analyzed in parallel, the one billion of social data posts queried in real-time on an in-memory components database. Traditional disks or commercial storage nowadays cannot handle the extreme scale of such application data.Following the need of improvement of current concepts and technologies, we focus in this paper on the needs of data intensive applications running on systems composed of up to millions of computing elements (exascale systems). We propose in this paper a methodology to advance the state-of-the-art. The starting point is the definition of new programming paradigms, APIs, runtime tools and methodologies for expressing data-intensive tasks on exascale systems. This will pave the way for the exploitation of massive parallelism over a simplified model of the system architecture, thus promoting high performance and efficiency, offering powerful operations and mechanisms for processing extreme data sources at high speed and/or real time.

show abstract

“…Vinas et al [204] propose a high-level programming library with OpenCL code generation facility whereas KFusion [134] can help fusing multiple OpenCL kernel calls for performance reasons. Benkner et al [31] propose high level constructs (in the form of pragmas) for modeling pipeline patterns in an application; underneath, they generate StarPU code for making decisions at runtime.…”

Section: Other Approaches For Gpu-based Systemsmentioning

confidence: 99%

Performance-aware component composition for GPU-based systems

Dastgeer¹

2014

View full text Add to dashboard Cite

This thesis adresses issues associated with efficiently programming modern heterogeneous GPU-based systems, containing multicore CPUs and one or more programmable Graphics Processing Units (GPUs). We use ideas from component-based programming to address programming, performance and portability issues of these heterogeneous systems. Specifically, we present three approaches that all use the idea of having multiple implementations for each computation; performance is achieved/retained either a) by selecting a suitable implementation for each computation on a given platform or b) by dividing the computation work across different implementations running on CPU and GPU devices in parallel.In the first approach, we work on a skeleton programming library (SkePU) that provides high-level abstraction while making intelligent implementation selection decisions underneath either before or during the actual program execution. In the second approach, we develop a composition tool that parses extra information (metadata) from XML files, makes certain decisions offline, and, in the end, generates code for making the final decisions at runtime. The third approach is a framework that uses source-code annotations and program analysis to generate code for the runtime library to make the selection decision at runtime. With a generic performance modeling API alongside program analysis capabilities, it supports online tuning as well as complex program transformations.These approaches differ in terms of genericity, intrusiveness, capabilities and knowledge about the program source-code; however, they all demonstrate usefulness of component programming techniques for programming GPU-based systems. With experimental evaluation, we demonstrate how all three approaches, although different in their own way, provide good performance on different GPU-based systems for a variety of applications.This work has been supported by two EU FP7 projects (PEP-PHER, EXCESS) and by SeRC. Populärvetenskaplig sammanfattningAtt få varje generation av datorer att fungera snabbareär viktigt för samhä-llets utveckling och tillväxt. Traditionellt hade de flesta datorer bara en general-purpose processor (den så kallade CPU:n) som bara kunde exekvera en beräkningsuppgift i taget. Under det senasteårtiondet har dock flerkärniga och mångkärniga processorer blivit vanliga, och datorer har också blivit mer heterogena. Ett modernt datorsystem innehåller vanligtvis flerän en CPU, tillsammans med specialprocessorer såsom grafikprocessorer (GPU:er) som ar anpassade för att kunna exekvera vissa typer av beräkningar effektivarë an CPU:er. Vi kallar ett sådant system med en eller flera GPU:er för ett GPU-baserat system. GPU:er i sådana system har sitt eget separata minne, och för att kunna köra en beräkning på en GPU så behöver man vanligtvis flytta all indata till GPU:ns minne och sedan hämta tillbaka resultatet när beräkningenär klar.Programmeringen av GPU-baserade systemär icke-trivialt av flera anledningar: (1) CPU:er och GPU:er kräver olika programmeringsexper...

show abstract

Exploiting heterogeneous parallelism with the Heterogeneous Programming Library

Cited by 31 publications

References 20 publications

Accelerating the HyperLogLog Cardinality Estimation Algorithm

Accelerating the HyperLogLog Cardinality Estimation Algorithm

On Processing Extreme Data

Performance-aware component composition for GPU-based systems

Contact Info

Product

Resources

About