Download as PDF, PPTX
![Extract features Feature extracIon Original dataset 17 PredicIve model EvaluaIon Text Label Words Features I bought the game... 4 “i", “bought”,... [1, 0, 3, 9, ...] Do NOT bother try... 1 “do”, “not”,... [0, 0, 11, 0, ...] this shirt is aweso... 5 “this”, “shirt” [0, 2, 3, 1, ...] never got it. Seller... 1 “never”, “got” [1, 2, 0, 0, ...] I ordered this to... 3 “i”, “ordered” [1, 0, 0, 3, ...]](https://image.slidesharecdn.com/16-08-14kddbigmine16talk-160815145004/75/Foundations-for-Scaling-ML-in-Apache-Spark-17-2048.jpg)
![Fit a model Feature extracIon Original dataset 18 PredicIve model EvaluaIon Text Label Words Features Prediction Probability I bought the game... 4 “i", “bought”,... [1, 0, 3, 9, ...] 4 0.8 Do NOT bother try... 1 “do”, “not”,... [0, 0, 11, 0, ...] 2 0.6 this shirt is aweso... 5 “this”, “shirt” [0, 2, 3, 1, ...] 5 0.9 never got it. Seller... 1 “never”, “got” [1, 2, 0, 0, ...] 1 0.7 I ordered this to... 3 “i”, “ordered” [1, 0, 0, 3, ...] 4 0.7](https://image.slidesharecdn.com/16-08-14kddbigmine16talk-160815145004/75/Foundations-for-Scaling-ML-in-Apache-Spark-18-2048.jpg)
![Evaluate Feature extracIon Original dataset 19 PredicIve model EvaluaIon Text Label Words Features Prediction Probability I bought the game... 4 “i", “bought”,... [1, 0, 3, 9, ...] 4 0.8 Do NOT bother try... 1 “do”, “not”,... [0, 0, 11, 0, ...] 2 0.6 this shirt is aweso... 5 “this”, “shirt” [0, 2, 3, 1, ...] 5 0.9 never got it. Seller... 1 “never”, “got” [1, 2, 0, 0, ...] 1 0.7 I ordered this to... 3 “i”, “ordered” [1, 0, 0, 3, ...] 4 0.7](https://image.slidesharecdn.com/16-08-14kddbigmine16talk-160815145004/75/Foundations-for-Scaling-ML-in-Apache-Spark-19-2048.jpg)
![Tungsten in ML Tungsten: off-heap memory management • Avoids JVM GC • Uses efficient storage formats • Code generation 26 Issue in ML: object creation during each iteration Issue in ML: Array[(Int,Double,Double)] Issue in ML: Volcano iterator model in MR/RDDs](https://image.slidesharecdn.com/16-08-14kddbigmine16talk-160815145004/75/Foundations-for-Scaling-ML-in-Apache-Spark-26-2048.jpg)
Uploaded byDatabricks
Foundations for Scaling ML in Apache Spark
The document discusses the foundations for scaling machine learning (ML) in Apache Spark, highlighting its capabilities, such as scalability using resilient distributed datasets (RDDs) and integration with DataFrames. Key focuses include the performance challenges of RDDs, the advantages of using DataFrames for ML pipelines, and future improvements with Catalyst query optimization and Tungsten memory management. The author emphasizes the importance of transitioning ML algorithms to DataFrames for better scalability and usability.
Foundations for Scaling ML in Apache Spark
- 1. Foundations for ScalingML in Apache Spark Joseph K. Bradley August 14, 2016 ® ™
- 2. Who am I? ApacheSpark committer & PMC member Software Engineer @ Databricks Machine Learning Department @ Carnegie Mellon 2
- 3. • General enginefor big data computing • Fast • Easy to use • APIs in Python, Scala, Java & R 3 Apache Spark Spark SQL Streaming MLlib GraphX Largest cluster: 8000 Nodes (Tencent) Open source • Apache Software Foundation • 1000+ contributors • 200+ companies & universities
- 4. NOTABLE USERS THATPRESENTED AT SPARK SUMMIT 2015 SAN FRANCISCO Source: Slide 5 of Spark Community Update
- 5. MLlib: Spark’s MLlibrary 5 0 500 1000 v0.8 v0.9 v1.0 v1.1 v1.2 v1.3 v1.4 v1.5 v1.6 v2.0 commits/release Learning tasks Classification Regression Recommendation Clustering Frequent itemsets Data utilities Featurization Statistics Linear algebra Workflow utilities Model import/export Pipelines DataFrames Cross validation Goals Scale-out ML Standard library Extensible API
- 6. MLlib: original design RDDs Challengesfor scalability 6
- 7. Resilient Distributed Datasets(RDDs) 7 Map Reduce master
- 8. Resilient Distributed Datasets(RDDs) 8
- 9. Resilient Distributed Datasets(RDDs) 9 Resiliency • Lineage • Caching & checkpointing
- 10. ML on RDDs:the good Flexible: GLMs, trees, matrix factorization, etc. Scalable: E.g., Alternating Least Squares on Spotify data • 50+ million users x 30+ million songs • 50 billion ratings Cost ~ $10 • 32 r3.8xlarge nodes (spot instances) • For rank 10 with 10 iterations, ~1 hour running time. 10
- 11. ML on RDDs:the challenges Partitioning • Data partitioning impacts performance. • E.g., for Alternating Least Squares Lineage • Iterative algorithms à long RDD lineage • Solvable via careful caching and checkpointing JVM • Garbage collection (GC) • Boxed types 11
- 12. MLlib: current status DataFrame& Dataset integration Pipelines API 12
- 13. Spark DataFrames &Datasets 13 dept age name Bio 48 H Smith CS 34 A Turing Bio 43 B Jones Chem 61 M Kennedy Data grouped into named columns DSL for common tasks • Project, filter, aggregate, join, … • 100+ functions available • User-Defined Functions (UDFs) data.groupBy(“dept”).avg(“age”) Datasets: Strongly typed DataFrames
- 14. DataFrame optimizations Catalyst queryoptimizer Project Tungsten • Memory management • Code generation 14 Predicate pushdown Join selection … Off-heap Avoid JVM GC Compressed format Combine operations into single, efficient code blocks
- 15. ML Pipelines • DataFrames:unified ML dataset API • Flexible types • Add & remove columns during Pipeline execution 15
- 16. Load data Feature extracIon Original dataset 16 PredicIve model EvaluaIon Text Label Ibought the game... 4 Do NOT bother try... 1 this shirt is aweso... 5 never got it. Seller... 1 I ordered this to... 3
- 17. Extract features Feature extracIon Original dataset 17 PredicIve model EvaluaIon Text LabelWords Features I bought the game... 4 “i", “bought”,... [1, 0, 3, 9, ...] Do NOT bother try... 1 “do”, “not”,... [0, 0, 11, 0, ...] this shirt is aweso... 5 “this”, “shirt” [0, 2, 3, 1, ...] never got it. Seller... 1 “never”, “got” [1, 2, 0, 0, ...] I ordered this to... 3 “i”, “ordered” [1, 0, 0, 3, ...]
- 18. Fit a model Feature extracIon Original dataset 18 PredicIve model EvaluaIon TextLabel Words Features Prediction Probability I bought the game... 4 “i", “bought”,... [1, 0, 3, 9, ...] 4 0.8 Do NOT bother try... 1 “do”, “not”,... [0, 0, 11, 0, ...] 2 0.6 this shirt is aweso... 5 “this”, “shirt” [0, 2, 3, 1, ...] 5 0.9 never got it. Seller... 1 “never”, “got” [1, 2, 0, 0, ...] 1 0.7 I ordered this to... 3 “i”, “ordered” [1, 0, 0, 3, ...] 4 0.7
- 19. Evaluate Feature extracIon Original dataset 19 PredicIve model EvaluaIon Text Label WordsFeatures Prediction Probability I bought the game... 4 “i", “bought”,... [1, 0, 3, 9, ...] 4 0.8 Do NOT bother try... 1 “do”, “not”,... [0, 0, 11, 0, ...] 2 0.6 this shirt is aweso... 5 “this”, “shirt” [0, 2, 3, 1, ...] 5 0.9 never got it. Seller... 1 “never”, “got” [1, 2, 0, 0, ...] 1 0.7 I ordered this to... 3 “i”, “ordered” [1, 0, 0, 3, ...] 4 0.7
- 20. ML Pipelines DataFrames: unifiedML dataset API • Flexible types • Add & remove columns during Pipeline execution • Materialize columns lazily • Inspect intermediate results 20
- 21. Under the hood:optimizations Current use of DataFrames • API • Transformations & predictions 21 Feature transformation & model prediction are phrased as User- Defined Functions (UDFs) à Catalyst query optimizer à Tungsten memory management + code generation
- 22. MLlib: future scaling DataFramesfor training Potential benefits • Spilling to disk • Catalyst • Tungsten Challenges remaining 22
- 23. Implementing ML onDataFrames 23 Map Reduce master
- 24. Scalability DataFrames automatically spillto disk à Classic pain point of RDDs 24 java.lang.OutOfMemoryError Goal: Smoothly scale, without custom per-algorithm optimizations
- 25. Catalyst in ML Keyidea: automatic query (ML algorithm) optimization • DataFrame operations are lazy. • Express entire algorithm as DataFrame operations. • Let Catalyst reorganize the algorithm, data, etc. à Fewer manual optimizations 25
- 26. Tungsten in ML Tungsten:off-heap memory management • Avoids JVM GC • Uses efficient storage formats • Code generation 26 Issue in ML: object creation during each iteration Issue in ML: Array[(Int,Double,Double)] Issue in ML: Volcano iterator model in MR/RDDs
- 27. Prototyping ML onDataFrames Currently: • Belief propagation • Connected components Current challenges: • DataFrame query plans do not have iteration as a top-level concept • ML/Graph-specific optimizations for Catalyst query planner Eventual goal: Port all ML algorithms to run on top of DataFrames à speed & scalability 27
- 28. To summarize... MLlib onRDDs • Required custom optimizations MLlib with a DataFrame-based API • Friendly API • Improvements for prediction MLlib on DataFrames • Potential for even greater scaling for training • Simpler for non-experts to write new algorithms 28
- 29. Get started Get involved • JIRA http://issues.apache.org • mailing lists http://spark.apache.org • Github http://github.com/apache/spark • Spark Packages http://spark-packages.org Learn more • New in Apache Spark 2.0 http://databricks.com/blog/2016/06/01 • MOOCs on EdX http://databricks.com/spark/training 29 Try out Apache Spark 2.0 in Databricks Community Edition http://databricks.com/ce Many thanks to the community for contributions & support!
- 30. Databricks Founded by thecreators of Apache Spark Offers hosted service • Spark on EC2 • Notebooks • Visualizations • Cluster management • Scheduled jobs 30 We’re hiring!
- 31.