Biological Sequence Comparison on Hybrid Platforms with Dynamic Workload Adjustment

Mendonca, Fernando Machado; Melo, Alba Cristina Magalhães Alves de

doi:10.1109/ipdpsw.2013.28

Cited by 7 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Load balancing. According to formula (9), the load balance deviation degree of node i can be obtained, and the set of all is F. Load balancing has three states: full load balancing STATE1, approximate load balancing STATE2, and non-load balancing STATE3. The mathematical description corresponding to load balancing LB is shown in formula ( 21), where c is the set of calculations for all tasks.…”

Section: Related Experiments and Results Analysis 41 Evaluation Indexmentioning

confidence: 99%

“…Full comparison calculation is a typical calculation mode [1], used to solve a type of calculation associated with pairwise data files. Full-comparative computing, as a special computing model, frequently appears in many disciplines, such as bioinformatics [2], biometrics [3], traditional machine learning field [4][5] [6], natural language processing field [7], the field of traffic big data [8] [9]. Scholars at home and abroad have been researching full-comparative computing, and full-comparative computing is one of the research hotspots.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Particle Swarm Optimization Algorithm for Data Distribution Model Design

Taniar¹

2022

JSER

View full text Add to dashboard Cite

This paper studies the data distribution problem of full comparative computing, proposes a data distribution model and related algorithms based on Particle Swarm Optimization (PSO) algorithm, and conducts experiments on the algorithm. The experimental results show that the data distribution scheme given by the data distribution model of the particle swarm optimization algorithm can realize the complete localization of the data files required by the task, and can reduce the use of storage space in the distributed system. In terms of load balancing, the task scheduling scheme given by this model can achieve load balancing among various nodes in a distributed system. In terms of computing time, the model can find data distribution schemes and task scheduling schemes at a faster rate and can better complete the data distribution of full comparison calculations in a distributed system. The data distribution model of the particle swarm optimization algorithm effectively solves the data distribution problem of large-scale full-comparison computing and will promote the research progress of bioinformatics, natural language processing, and other fields.

show abstract

Section: Related Experiments and Results Analysis 41 Evaluation Indexmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Particle Swarm Optimization Algorithm for Data Distribution Model Design

Taniar¹

2022

JSER

View full text Add to dashboard Cite

show abstract

“…Current problems of great concern in SW algorithm are computations and spaces complexity, which required powerful algorithm(s) to utilize the power of parallel machines. Implementing SW algorithm in parallel platforms plays a key role in sequences comparisons problems in order [16][17][18][19][22][23][24][25][26][27]. However, a serious weakness with this architecture is the limitation and constraint of fixed sizes of memory available for all shared processors.…”

Section: Discussionmentioning

confidence: 99%

“…Most current discussions in local sequence alignment focus on multicore architectures with shared memory such as divide and conquer techniques [16][17][18], striped SW [19][20][21][22], Instruction-set Processor (ASIP) architecture [23], data compression [24,25], genome assembly (re-sequence) algorithms [26,27], and Symmetric Multi-Processing (SMP) architecture [28,29]. This section discusses in detail these algorithms along with their advantage and disadvantage.…”

Section: Fig 3: Dual Core and Quad Core In Multicores Platformsmentioning

confidence: 99%

See 1 more Smart Citation

Shared Memory and Hardware Utilizations for the Parallelization of Local Sequences Alignment using SW Algorithm: A Review

Eltayeeb¹,

Latiff²,

Isnin³

2015

IJCA

View full text Add to dashboard Cite

It is becoming increasingly difficult to ignore the importance of aligning DNA and Protein sequences to infer properties of new sequences from well-known reference sequences established and sorted in genetics databanks. Many studies in recent years have focused on different implementations of Sequences Alignment Problems (SAP). However, researcher confused with the ambiguous classification of the SAP. This paper is set out mainly to review, investigate, and analysis current trends in shared memory and hardware implementation of local SAP using Smith-Waterman algorithm. The literatures are addressing and evaluating in order to highlight their advantages and disadvantages.

show abstract