Parallel Programming in Python: Speeding up your analysis
2 likes1,211 views
The document presents an overview of parallel programming in Python, emphasizing its significance in enhancing the efficiency of data science tasks. It details the motivation for parallelism, including the prevalence of medium data problems and the limitations of clusters, while providing practical tools and methods for implementation. The author also highlights various machine learning applications, demonstrations, and mentions using GPUs for accelerated deep learning model training.
![Don't oversaturate your resources
Domino Data Lab November 10, 2015
itemIDs = [1, 2, … , n]
parallel-for-each(i = itemIDs){
item = fetchData(i)
result = computeSomething(item)
saveResult(result)
}](https://image.slidesharecdn.com/parallelprogrammingpydata-151107234945-lva1-app6891/85/Parallel-Programming-in-Python-Speeding-up-your-analysis-8-320.jpg)
![Parallelize tasks to match your resources
Domino Data Lab November 10, 2015
items = fetchData([1, 2, … , n])
results = parallel-for-each(i = items){
computeSomething(item)
}
saveResult(results)](https://image.slidesharecdn.com/parallelprogrammingpydata-151107234945-lva1-app6891/85/Parallel-Programming-in-Python-Speeding-up-your-analysis-9-320.jpg)
![Avoid modifying global state
Domino Data Lab November 10, 2015
itemIDs = [0, 0, 0, 0]
parallel-for-each(i = 1:4) {
itemIDs[i] = i
}
A = [0,0,0,0]Array initialized in process 1
[0,0,0,0] [0,0,0,0][0,0,0,0][0,0,0,0]Array copied to each sub-process
[0,0,0,0] [0,0,0,3][0,0,2,0][0,1,0,0]The copy is modified
[0,0,0,0]
When all parallel tasks finish, array in original process remained unchanged](https://image.slidesharecdn.com/parallelprogrammingpydata-151107234945-lva1-app6891/85/Parallel-Programming-in-Python-Speeding-up-your-analysis-10-320.jpg)
Parallel Programming in Python: Speeding up your analysis
- 1. Domino Data Lab November 10, 2015 Faster data science — without a cluster Parallel programming in Python Manojit Nandi [email protected] @mnandi92
- 2. Who am I? Domino Data Lab November 10, 2015 • Data Scientist at STEALTHBits Technologies • Data Science Evangelist at Domino Data Lab • BS in Decision Science
- 3. Domino Data Lab November 10, 2015 Agenda and Goals • Motivation • Conceptual intro to parallelism, general principles and pitfalls • Machine learning applications • Demos Goal: Leave you with principles, and practical concrete tools, that will help you run your code much faster
- 4. Motivation Domino Data Lab November 10, 2015 • Lots of “medium data” problems • Can fit in memory on one machine • Lots of naturally parallel problems • Easy to access large machines • Clusters are hard • Not everything fits map-reduce CPUs with multiple cores have become the standard in the recent development of modern computer architectures and we can not only find them in supercomputer facilities but also in our desktop machines at home, and our laptops; even Apple's iPhone 5S got a 1.3 GHz Dual-core processor in 2013. - Sebastian Rascka
- 5. Parallel programing 101 Domino Data Lab November 10, 2015 • Think about independent tasks (hint: “for” loops are a good place to start!) • Should be CPU-bound tasks • Warning and pitfalls • Not a substitute for good code • Overhead • Shared resource contention • Thrashing Source: Blaise Barney, Lawrence Livermore National Laboratory
- 6. Can parallelize at different “levels” Domino Data Lab November 10, 2015 Will focus on algorithms, with some brief comments on Experiments Run against underlying libraries that parallelize low-level operations, e.g., openBLAS, ATLAS Write your code (or use a package) to parallelize functions or steps within your analysis Run different analyses at once Math ops Algorithms Experiments
- 7. Parallelize tasks to match your resources Domino Data Lab November 10, 2015 Computing something (CPU) Reading from disk/database Writing to disk/database Network IO (e.g., web scraping) Saturating a resource will create a bottleneck
- 8. Don't oversaturate your resources Domino Data Lab November 10, 2015 itemIDs = [1, 2, … , n] parallel-for-each(i = itemIDs){ item = fetchData(i) result = computeSomething(item) saveResult(result) }
- 9. Parallelize tasks to match your resources Domino Data Lab November 10, 2015 items = fetchData([1, 2, … , n]) results = parallel-for-each(i = items){ computeSomething(item) } saveResult(results)
- 10. Avoid modifying global state Domino Data Lab November 10, 2015 itemIDs = [0, 0, 0, 0] parallel-for-each(i = 1:4) { itemIDs[i] = i } A = [0,0,0,0]Array initialized in process 1 [0,0,0,0] [0,0,0,0][0,0,0,0][0,0,0,0]Array copied to each sub-process [0,0,0,0] [0,0,0,3][0,0,2,0][0,1,0,0]The copy is modified [0,0,0,0] When all parallel tasks finish, array in original process remained unchanged
- 11. Demo Domino Data Lab November 10, 2015
- 12. Many ML tasks are parallelized Domino Data Lab November 10, 2015 • Cross-Validation • Grid Search Selection • Random Forest • Kernel Density Estimation • K-Means Clustering • Probabilistic Graphical Models • Online Learning • Neural Networks (Backpropagation) Harder to parallelize Intuitive to parallelize
- 13. Cross validation Domino Data Lab November 10, 2015
- 14. Grid search Domino Data Lab November 10, 2015 1 10 100 1000 Linear RBF C Kernel
- 15. Random forest Domino Data Lab November 10, 2015
- 16. Parallel programing in Python Domino Data Lab November 10, 2015 • Joblib pythonhosted.org/joblib/parallel.html • scikit-learn (n_jobs) scikit-learn.org • GridSearchCV • RandomForest • KMeans • cross_val_score • IPython Notebook clusters www.astro.washington.edu/users/vanderplas/Astr599/notebooks/ 21_IPythonParallel
- 17. Demo Domino Data Lab November 10, 2015
- 18. Parallel Programming using the GPU Domino Data Lab November 10, 2015 • GPUs are essential to deep learning because they can yield 10x speed-up when training the neural networks. • Use PyCUDA library to write Python code that executes using the GPU.
- 19. Demo Domino Data Lab November 10, 2015
- 20. Can compose layers of parallelism Domino Data Lab November 10, 2015 c1 c2 cn… c1 c2 cn…c1 c2 cn… Machines (experiments) Cores RF NN GridSearched SVC
- 21. Demo Domino Data Lab November 10, 2015
- 22. FYI: Parallel programing in R Domino Data Lab November 10, 2015 • General purpose • parallel • foreach cran.r-project.org/web/packages/foreach • More specialized • randomForest cran.r-project.org/web/packages/randomForest • caret topepo.github.io/caret • plyr cran.r-project.org/web/packages/plyr
- 23. Domino Data Lab November 10, 2015 dominodatalab.com blog.dominodatalab.com @dominodatalab Check us out!