A Pipeline Implementation for Dynamic Programming on GPU

Miyazaki, Makoto; Matsumae, Susumu

doi:10.1109/candarw.2018.00063

Cited by 2 publications

(13 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Step 1, the head position i of the thread group is 5. In this step the only one thread is activated and executes ST [5] ← ST[0]. In Step 2, the head position is incremented to 6, and two threads are activated.…”

Section: A Pipeline Implementation For S-dp Problemmentioning

confidence: 99%

“…In Step 2, the head position is incremented to 6, and two threads are activated. The first thread treats ST [6] and the second thread works on ST [5]. In Step 3, the head position becomes 7, and now all k = 3 threads actively execute operations for ST [7], ST [6], and ST [5] respectively.…”

Section: A Pipeline Implementation For S-dp Problemmentioning

confidence: 99%

“…For such a case, in [5], we proposed a 2-by-2 pipeline implementation technique where each thread invoked in the inner loop executes two computations for each element of array ST. The details can be found in [5]. Fig.…”

Section: A Pipeline Implementation For S-dp Problemmentioning

confidence: 99%

“…This paper is an extended version of our conference paper [5] where we proposed a pipeline-implementation for the S-DP problem and discussed the memory access conflict issues. In this paper, we provide the MCM algorithm in details and formally prove lemmas and theorem in Section IV.…”

Section: Acknowledgmentmentioning

confidence: 99%

“…Our implementation determines one output value per one computational step with n threads in a pipeline fashion and constructs the solution table totally in O(n 2 ) steps with n threads. This paper is an extended version of our conference paper [5].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Solving Dynamic Programming Problem by Pipeline Implementation on GPU

Matsumae¹,

Miyazaki²

2018

ijacsa

Self Cite

View full text Add to dashboard Cite

In this paper, we show the effectiveness of a pipeline implementation of Dynamic Programming (DP) on GPU. As an example, we explain how to solve a matrix-chain multiplication (MCM) problem by DP on GPU. This problem can be sequentially solved in O(n 3) steps by DP where n is the number of matrices, because its solution table is of size n × n and each element of the table can be computed in O(n) steps. A typical speedup strategy for this is to parallelize the O(n) step computation of each element, which can be easily achieved by parallel prefix computation, i.e., an O(log n) step computation with n threads in a tournament fashion. By such a standard parallelizing method, we can solve the MCM problem in O(n 2 log n) steps with n threads. In our approach, we solve the MCM problem on GPU in a pipeline fashion, i.e., we use GPU cores for supporting pipelinestages so that many elements of the solution table are partially computed in parallel at one time. Our implementation determines one output value per one computational step with n threads in a pipeline fashion and constructs the solution table totally in O(n 2) steps with n threads.

show abstract

Section: A Pipeline Implementation For S-dp Problemmentioning

confidence: 99%

Section: A Pipeline Implementation For S-dp Problemmentioning

confidence: 99%

Section: A Pipeline Implementation For S-dp Problemmentioning

confidence: 99%

Section: Acknowledgmentmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Solving Dynamic Programming Problem by Pipeline Implementation on GPU

Matsumae¹,

Miyazaki²

2018

ijacsa

Self Cite

View full text Add to dashboard Cite

show abstract

Accelerating pipeline implementation of dynamic programming on GPU

Matsumae¹

EPiC Series in Computing

View full text Add to dashboard Cite

In this paper, we show the effectiveness of pipeline implementations of Dynamic Pro- gramming (DP) on Graphics Processing Unit (GPU). We deal with a simplified DP problem where each element of its solution table is calculated in order by semi-group operations among several of already computed elements in the table. We implement the DP program on GPU in a pipeline fashion, i.e., we use GPU cores for supporting pipeline-stages so that several elements of the solution tables are partially computed at one time. Further, to accelerate the pipeline implementation, we propose a p-fold pipeline technique, which enables larger parallelism more than the number of pipeline-stages.

show abstract

A Pipeline Implementation for Dynamic Programming on GPU

Cited by 2 publications

References 3 publications

Solving Dynamic Programming Problem by Pipeline Implementation on GPU

Solving Dynamic Programming Problem by Pipeline Implementation on GPU

Accelerating pipeline implementation of dynamic programming on GPU

Contact Info

Product

Resources

About