Theo Ungerer, Eberhard Zehendner · ACM SIGARCH Computer Architecture News · 1991
This paper describes a parallel programming language and a multi-level parallelism computer architecture, which have been developed by an integrated design process.The ASTOR language is an imperative intermediate language that combines parallel control constructs with a concise module concept. The ASTOR architecture is directed towards reliable execution of programs written in the ASTOR language, utilizing five levels of parallelism expressed by the language constructs. Structure preservation has been applied as a major design principle. The architecture can be characterized as a message passing multiprocessor whose nodes consist of decoupled program flow control and data object processing parts, executing by token-passing similar to large-grain data flow architectures.
Cited by 2
Hazrina Yusof Hamdani, Nur’Aini Abdul Rashid · 2010
In this paper we investigate the advantages of multi-level parallelism on PhyloCon algorithm. The PhyloCon algorithm naturally possesses multi-levels of functions which cost the execution time took longer to complete. We showed that by applying parallelism to the selected level of functions may significantly reduce the execution time. The selected level of functions produced the independent outer-level parallelism and the independent inner-level parallelism. The details of each independent level of parallelism applied in PhyloCon algorithm are also were shown. The result of the research shows that multilevel parallelism produced a better performance compared to independent level parallelism which half of sequential execution time can be reduced using multi-level parallelism.
Cited by 0
Qifeng Huang, Yan Huang · 2008
This paper presents the SWFL workflow engine, a general workflow framework that meets the needs of business processes as well as scientific computing processes with fine multi-level parallelism supports. The workflow description language, SWFL, follows a graph-oriented model to specify workflow processes composed of services. The workflow engine provides an efficient enactment environment for SWFL flow model composed of services. It provides multi-level parallelism supports: a server-level parallelism support using flexible server-level schedule algorithms, a flow-level parallelism support by partitioning a workflow into sub-flows and running the sub-flows in multiple job processing servers in parallel, and a message-passing parallelism support by using the MPFL, an extension language to SWFL and its associated tools. The architecture of the workflow engine and some other related implementation issues are also presented in the paper. 1.
Cited by 0
Ming-yu Chen, Lily Mummert, Padmanabhan Pillai et al. · 2010
Video understanding is a computationally challenging task that is critical not only for traditionally throughput-oriented applications such as search but also latency-sensitive interactive applications such as surveillance, gaming, videoconferencing, and vision-based user interfaces. Enabling these types of video processing applications will require not only new algorithms and techniques, but new runtime systems that optimize latency as well as throughput. In this paper, we present a runtime system called Sprout that achieves low latency by exploiting the parallelism inherent in video understanding applications. We demonstrate the utility of our system on an activity recognition application that employs a robust new descriptor called MoSIFT, which explicitly augments appearance features with motion information. MoSIFT outperforms previous recognition techniques, but like other state-of-the-art techniques, it is computationally expensive — a sequential implementation runs 100 times slower than real time. We describe the implementation of the activity recognition application on Sprout, and show that it can accurately recognize activities at full frame rate (25 fps) and low latency on a challenging airport surveillance video corpus.
Cited by 0
Milfeld, Kent, Kent Milfeld, Bronis R. de Supinski et al. · Lecture notes in computer science · 2020
Cited by 2
Qifeng Huang, Yan Huang · ORCA Online Research @Cardiff (Cardiff University) · 2005
This paper presents the SWFL workflow engine, a general workflow framework that meets the needs of business processes as well as scientific computing processes with fine multi-level parallelism supports. The workflow description language, SWFL, follows a graph-oriented model to specify workflow processes composed of services. The workflow engine provides an efficient enactment environment for SWFL flow model composed of services. It provides multi-level parallelism supports: a server-level parallelism support using flexible server-level schedule algorithms, a flow-level parallelism support by partitioning a workflow into sub-flows and running the sub-flows in multiple job processing servers in parallel, and a message-passing parallelism support by using the MPFL, an extension language to SWFL and its associated tools. The architecture of the workflow engine and some other related implementation issues are also presented in the paper.
Cited by 0Open access
Ferdinando Fioretto, William Yeoh, Enrico Pontelli · 2015
Current DCOP algorithms suffer from a major limiting assumption---each agent can handle only a single variable of the problem---which limits their scalability. This paper proposes a novel Multi-Variable Agent (MVA) DCOP decomposition, which: (i) Exploits co-locality of an agent's variables, allowing us to adopt efficient centralized techniques; (ii) Enables the use of hierarchical parallel models, such us those based on GPGPUs; and (iii) Empirically reduces the amount of communication required in several classes of DCOP algorithms. Experimental results show that our MVA decomposition outperforms non-decomposed DCOP algorithms, in terms of network load and scalability.
Cited by 0
Xavier Martorell, Eduard Ayguadé, Nacho Navarro et al. · 1999
This paper presents some techniques for efficient thread forking and joining in parallel execution environments, taking into consideration the physical structure of NUMA machines and the support for multi-level parallelization and processor grouping. Two work generation schemes and one join mechanism are designed, implemented, evaluated and compared with the ones used in the IRIX MP library, an efficient implementation which supports a single level of parallelism. Supporting multiple levels of parallelism is a current research goal, both in shared and distributed memory machines. Our proposals include a first work generation scheme (GWD, or global work descriptor) which supports multiple levels of parallelism, but not processor grouping. The second work generation scheme (LWD, or local work descriptor) has been designed to support multiple levels of parallelism and processor grouping. Processor grouping is needed to distribute processors among different parts of the computation and maintain the working set of each processor across different parallel constructs. The mechanisms are evaluated using synthetic benchmarks, two SPEC95fp applications and one NAS application. The performance evaluation concludes that: i) the overhead of the proposed mechanisms is similar to the overhead of the existing ones when exploiting a single level of parallelism, and ii) a remarkable improvement in performance is obtained for applications that have multiple levels of parallelism. The comparison with the traditional single-level parallelism exploitation gives an improvement in the range of 30-65 % for these applications. 1.
Cited by 0
Zeyu Xia, Canqun Yang, Chenchen Peng et al. · BMC Bioinformatics · 2025
Performance evaluations show that ParaHAT got a 10.03x speedup in base-level alignment, with a parallel acceleration ratio and weak scalability metric of 94.61 and 98.98% on 128 nodes, respectively.
Cited by 5Open access
Xia, Zeyu, Yang, Canqun, Peng, Chenchen et al. · Figshare · 2025
Abstract Background The advent of Single Molecule Real-Time (SMRT) sequencing has overcome many limitations of second-generation sequencing, such as limited read lengths, PCR amplification biases. However, longer reads increase data volume exponentially and high error rates make many existing alignment tools inapplicable. Additionally, a single CPU’s performance bottleneck restricts the effectiveness of alignment algorithms for SMRT sequencing. Results To address these challenges, we introduce ParaHAT, a parallel alignment algorithm for noisy long reads. ParaHAT utilizes vector-level, thread-level, process-level, and heterogeneous parallelism. We redesign the dynamic programming matrices layouts to eliminate data dependency in the base-level alignment, enabling effective vectorization. We further enhance computational speed through heterogeneous parallel technology and implement the algorithm for multi-node computing using MPI, overcoming the computational limits of a single node. Conclusions Performance evaluations show that ParaHAT got a 10.03x speedup in base-level alignment, with a parallel acceleration ratio and weak scalability metric of 94.61 and 98.98% on 128 nodes, respectively.
Cited by 0Open access
Alessandro De Gloria, Paolo Faraboschi · Integration · 1992
Cited by 4
Zeyu Xia, Canqun Yang, Chenchen Peng et al. · Research Square · 2024
Cited by 0Open access
Xia, Zeyu, Yang, Canqun, Peng, Chenchen et al. · Figshare · 2025
Abstract Background The advent of Single Molecule Real-Time (SMRT) sequencing has overcome many limitations of second-generation sequencing, such as limited read lengths, PCR amplification biases. However, longer reads increase data volume exponentially and high error rates make many existing alignment tools inapplicable. Additionally, a single CPU’s performance bottleneck restricts the effectiveness of alignment algorithms for SMRT sequencing. Results To address these challenges, we introduce ParaHAT, a parallel alignment algorithm for noisy long reads. ParaHAT utilizes vector-level, thread-level, process-level, and heterogeneous parallelism. We redesign the dynamic programming matrices layouts to eliminate data dependency in the base-level alignment, enabling effective vectorization. We further enhance computational speed through heterogeneous parallel technology and implement the algorithm for multi-node computing using MPI, overcoming the computational limits of a single node. Conclusions Performance evaluations show that ParaHAT got a 10.03x speedup in base-level alignment, with a parallel acceleration ratio and weak scalability metric of 94.61 and 98.98% on 128 nodes, respectively.
Cited by 0Open access
Gabriel Okša, Marián Vajteršic · 2002
We analyse the fine-grained parallelism of the two-sided block-Jacobi algorithm for the singular value decomposition (SVD) of matrix A/spl isin/R/sup m/spl times/n/, m/spl ges/n. The algorithm involves the class CO of parallel orderings on the two-dimensional toroidal mesh with p processors. The mathematical background is based on the QR decomposition (QRD) of local data matrices and on the triangular Kogbetliantz algorithm (TKA) for local SVDs in the diagonal mesh processors. Subsequent updates of local matrices in the diagonal as well as nondiagonal mesh processors are required. WE show that all updates can be realized by orthogonal modified Givens rotations. These rotations can be efficiently pipelined in parallel in the horizontal and vertical rings of /spl radic/p processors through the toroidal mesh. For one mesh processor our solution requires O[(m+n)/sup 2///sub p/] systolic processing elements (PEs). O(m/sup 2//p) local memory registers and O[(m+n)/sup 2//p] additional delay elements. The time complexity of our solution is O[(m+n/sup 3/2//p/sup 3/4/)/spl Delta/] time steps per one global iteration where /spl Delta/ is the length of the global synchronization time step that is given by evaluation and application of two modified Givens rotations in TKA.
Cited by 1
Singh, Dilip · NORMA · 2021
With the rapid growth of Agile software delivery trends Continuous Integration and Continuous Deployment(CICD) has gained a wide adoption across IT organizations. Continuous Integration and Continuous Deployment(CICD) are the basic pillars of DevOps which enable rapid software deployment with quick feedback. It also bridges the gap between operations and the developer team. Recent studies show that adopting DevOps practices for software development sometimes creates additional overhead for developers due to overly long build time. It becomes even worse when waiting for a build to successfully finish gets fail. To overcome this barrier, the paper emphasis improving Continuous Deployment(CD) performance to reduce the application deployment makespan. The motive is to optimize the performance of AWS Elastic BeanStalk using multi-level thread parallelism in Python. I have used a custom script written in python 3.7 to invoke multiple threads for deploying application files from the AWS S3 bucket to the AWS EC2 instance. The entire software delivery can be improved while reducing the costs by lowering the execution time for Continuous Deployment (CD). For the experiment, we have used AWS cloud and evaluated various matrices such as execution time, CPU utilization, a cost that occurred before and after optimization.
Cited by 0Open access
Daniele Paolo Scarpazza, Praveen Raghavan, David Novo et al. · Lecture notes in computer science · 2006
Cited by 11
Alessandro Bria, Massimo Bernaschi, Massimiliano Guarrasi et al. · Frontiers in Neuroinformatics · 2019
Due to the limited field of view of the microscopes, acquisitions of macroscopic specimens require many parallel image stacks to cover the whole volume of interest. Overlapping regions are introduced among stacks in order to make it possible automatic alignment by means of a 3D stitching tool. Since state-of-the-art microscopes coupled with chemical clearing procedures can generate 3D images whose size exceeds the Terabyte, parallelization is required to keep stitching time within acceptable limits. In the present paper we discuss how multi-level parallelization reduces the execution times of TeraStitcher, a tool designed to deal with very large images. Two algorithms performing dataset partition for efficient parallelization in a transparent way are presented together with experimental results proving the effectiveness of the approach that achieves a speedup close to 300×, when both coarse- and fine-grained parallelism are exploited. Multi-level parallelization of TeraStitcher led to a significant reduction of processing times with no changes in the user interface, and with no additional effort required for the maintenance of code.
Cited by 14Open access
Alessandro Bria (6793319), Massimo Bernaschi (75214), Massimiliano Guarrasi (6793322) et al. · Figshare · 2019
<p>Due to the limited field of view of the microscopes, acquisitions of macroscopic specimens require many parallel image stacks to cover the whole volume of interest. Overlapping regions are introduced among stacks in order to make it possible automatic alignment by means of a 3D stitching tool. Since state-of-the-art microscopes coupled with chemical clearing procedures can generate 3D images whose size exceeds the Terabyte, parallelization is required to keep stitching time within acceptable limits. In the present paper we discuss how multi-level parallelization reduces the execution times of TeraStitcher, a tool designed to deal with very large images. Two algorithms performing dataset partition for efficient parallelization in a transparent way are presented together with experimental results proving the effectiveness of the approach that achieves a speedup close to 300×, when both coarse- and fine-grained parallelism are exploited. Multi-level parallelization of TeraStitcher led to a significant reduction of processing times with no changes in the user interface, and with no additional effort required for the maintenance of code.</p>
Cited by 0Open access
I D E Titesis, George Allen Sawyer, I D~?~t&apos;wnsaimment A et al. · 1990
Cited by 0
Xavier Martorell, Eduard Ayguade , Jesus Labarta · 2000
In this paper we describe the main components of the NanosCompiler, an OpenMP compiler whose implementation is oriented towards the efficient exploitation of nested parallelism. Program parallelization relies both on the automatic parallelization capabilities of the base compiler and the information obtained from user--supplied directives. The compiler uses a hierarchical internal representation that unifies both sources of parallelism, proceeds with a task identification phase that adapts the granularity of the final tasks to the target architecture and then generates parallel code. The paper also presents an analysis of the special support needed from the threads library level to support this kind of parallelism. These requirements are analyzed in our current implementation named NthLib.
Cited by 0