Finite state machine-based optimization of data parallel regular domain problems applied in low-level image processing

Seinstra, F.J.; Koelma, D.C.; Bagdanov, Andrew D.

doi:10.1109/tpds.2004.55

Cited by 17 publications

(12 citation statements)

References 30 publications

(39 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, it was realized that it is not sufficient to consider parallelization of library operations in isolation. Therefore, the programming systems incorporate a run-time approach for communication minimization (called lazy parallelization) that automatically parallelizes a fully sequential program at runtime by inserting communication primitives and additional memory management operations whenever necessary [43].…”

Section: High-performance Distributed Multimedia Analysis With Jorusmentioning

confidence: 99%

Jungle Computing: Distributed Supercomputing Beyond Clusters, Grids, and Clouds

Seinstra

Maassen

Nieuwpoort

et al. 2011

Computer Communications and Networks

View full text Add to dashboard Cite

In recent years, the application of high-performance and distributed computing in scientific practice has become increasingly wide-spread. Among the most widely available platforms to scientists are clusters, grids, and cloud systems. Such infrastructures currently are undergoing revolutionary change due to the integration of many-core technologies, providing orders-of-magnitude speed improvements for selected compute kernels. With high-performance and distributed computing systems thus becoming more heterogeneous and hierarchical, programming complexity is vastly increased. Further complexities arise because urgent desire for scalability, and issues including data distribution, software heterogeneity, and ad-hoc hardware availability, commonly force scientists into simultaneous use of multiple platforms (e.g. clusters, grids, and clouds used concurrently). A true computing jungle. In this chapter we explore the possibilities of enabling efficient and transparent use of Jungle Computing Systems in every-day scientific practice. To this end, we discuss the fundamental methodologies required for defining programming models that are tailored to the specific needs of scientific researchers. Importantly, we claim that many of these fundamental methodologies already exist today, as integrated in our Ibis high-performance distributed programming system. We also make a case for the urgent need for easy and efficient Jungle Computing in scientific practice, by exploring a set of state-of-the-art application domains. For one of these domains we present results obtained with Ibis on a real-world Jungle Computing System. The chapter concludes by exploring fundamental research questions to be investigated in the years to come.

show abstract

Section: High-performance Distributed Multimedia Analysis With Jorusmentioning

confidence: 99%

Jungle Computing: Distributed Supercomputing Beyond Clusters, Grids, and Clouds

Seinstra

Maassen

Nieuwpoort

et al. 2011

Computer Communications and Networks

View full text Add to dashboard Cite

show abstract

“…Moreover, it was realized that it is not sufficient to consider parallelization of library operations in isolation. Therefore, the library was extended with a run-time approach for communication minimization (called lazy parallelization) that automatically parallelizes a fully sequential program at run-time by inserting communication primitives and additional memory management operations whenever necessary [10].…”

Section: Parallel-horusmentioning

confidence: 99%

Towards User Transparent Parallel Multimedia Computing on GPU-Clusters

Werkhoven

Maassen

Seinstra

2011

Computer Architecture

View full text Add to dashboard Cite

Abstract. The research area of Multimedia Content Analysis (MMCA) considers all aspects of the automated extraction of knowledge from multimedia archives and data streams. To satisfy the increasing computational demands of MMCA problems, the use of High Performance Computing (HPC) techniques is essential. As most MMCA researchers are not HPC experts, there is an urgent need for 'familiar' programming models and tools that are both easy to use and efficient. Today, several user transparent library-based parallelization tools exist that aim to satisfy both these requirements. In general, such tools focus on data parallel execution on traditional compute clusters. As of yet, none of these tools also incorporate the use of many-core processors (e.g. GPUs), however. While traditional clusters are now being transformed into GPU-clusters, programming complexity vastly increases -and the need for easy and efficient programming models is as urgent as ever. This paper presents our first steps in the direction of obtaining a user transparent programming model for data parallel and hierarchical multimedia computing on GPU-clusters. The model is obtained by extending an existing user transparent parallel programming system (applicable to traditional compute clusters) with a set of CUDA compute kernels. We show our model to be capable of obtaining orders-of-magnitude speed improvements, without requiring any additional effort from the application programmer.

show abstract

“…This is explained by the fact that, in the case of horizontal partitioning, most of the data structures to be communicated are stored contiguously in memory. In contrast, vertical partitioning causes data structures to be stored non-contiguously, a problem which is discussed extensively in [16]. As a result, horizontal partitioning is our parallelization strategy of choice for each single iteration of the Householder bidiagonalization (see Figure 2).…”

Section: Parallelizing a Single Iterationmentioning

confidence: 99%

Adaptive Parallel Householder Bidiagonalization

Liu

Seinstra²

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract.With the increasing use of large image and video archives and high-resolution multimedia data streams in many of today's research and application areas, there is a growing need for multimedia-oriented high-performance computing. As a consequence, a need for algorithms, methodologies, and tools that can serve as support in the (automatic) parallelization of multimedia applications is rapidly emerging.This paper discusses the parallelization of Householder bidiagonalization, a matrix factorization method which is an integral part of full Singular Value Decomposition (SVD) -an important algorithm for many multimedia problems. Householder bidiagonalization is hard to parallelize efficiently because the total number of matrix elements taking part in the calculations reduces during runtime. To overcome the growing negative performance impact of load imbalances and overprovisioning of compute resources, we apply adaptive runtime techniques of periodic matrix remapping and process reduction for improved performance. Results show that our adaptive parallel execution approach provides a significant improvement in efficiency, even when applying a set of compute resources which is (initially) very large.

show abstract

Finite state machine-based optimization of data parallel regular domain problems applied in low-level image processing

Cited by 17 publications

References 30 publications

Jungle Computing: Distributed Supercomputing Beyond Clusters, Grids, and Clouds

Jungle Computing: Distributed Supercomputing Beyond Clusters, Grids, and Clouds

Towards User Transparent Parallel Multimedia Computing on GPU-Clusters

Adaptive Parallel Householder Bidiagonalization

Contact Info

Product

Resources

About