Proceedings of the 7th Workshop on Python for High-Performance and Scientific Computing 2017
DOI: 10.1145/3149869.3149870
|View full text |Cite
|
Sign up to set email alerts
|

Enabling Python to execute efficiently in heterogeneous distributed infrastructures with PyCOMPSs

Abstract: Python has been adopted as programming language by a large number of scientific communities. Additionally to the easy programming interface, the large number of libraries and modules that have been made available by a large number of contributors, have taken this language to the top of the list of the most popular programming languages in scientific applications. However, one main drawback of Python is the lack of support for concurrency or parallelism. PyCOMPSs is a proved approach to support task-based paral… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
11
0

Year Published

2018
2018
2022
2022

Publication Types

Select...
5
2
1

Relationship

5
3

Authors

Journals

citations
Cited by 9 publications
(11 citation statements)
references
References 17 publications
(17 reference statements)
0
11
0
Order By: Relevance
“…Thus, to parallelise a vector operation, the users must explicitly define the loop nests to apply an operation to all the vector elements. However, in our previous work (Amela et al, 2017(Amela et al, , 2018, we have demonstrated the benefits from combining inter-and intra-node parallelism using PyCOMPSs (Tejedor et al, 2017) and NumPy. Similarly, some NumPy extensions can be integrated with PyCOMPSs to boost the intra-node performance.…”
Section: State Of the Artmentioning
confidence: 99%
See 1 more Smart Citation
“…Thus, to parallelise a vector operation, the users must explicitly define the loop nests to apply an operation to all the vector elements. However, in our previous work (Amela et al, 2017(Amela et al, , 2018, we have demonstrated the benefits from combining inter-and intra-node parallelism using PyCOMPSs (Tejedor et al, 2017) and NumPy. Similarly, some NumPy extensions can be integrated with PyCOMPSs to boost the intra-node performance.…”
Section: State Of the Artmentioning
confidence: 99%
“…Therefore, the performance evaluation of PyCOMPSs and its Runtime is beyond the scope of this paper. For further details, in our previous work (Amela et al, 2017(Amela et al, , 2018, we analysed indepth the performance obtained when executing linear algebra applications when combining PyCOMPSs for inter-node parallelism and NumPy for intra-node parallelism. Also, in Conejero et al (2018), we compared the PyCOMPSs Runtime against Apache Spark.…”
Section: Performance Evaluationmentioning
confidence: 99%
“…... COMPSs provides Java as native programming language and it also provides bindings for Python (PyCOMPSs [2]) and C/C++ [7]. Figure 1 shows an example of a task annotation and COMPSs main program.…”
Section: Compss Overviewmentioning
confidence: 99%
“…Finally, other libraries and frameworks enable Python distributed and multi-threaded computations such as Dask [14], PySpark [15], and PyCOMPSs [16], [17]. Dask is a native Python library that allows both the creation of custom DAG's and the distributed execution of a set of operations on NumPy and pandas [18] objects.…”
Section: State Of the Artmentioning
confidence: 99%