Spark Summit EU talk by Debasish Das and Pramod Narasimha
•
1 like•618 views
This document describes a system called DeviceAnalyzer that builds predictive models in near-real time using Apache Spark and Apache Lucene. It discusses: 1) Integrating Spark and Lucene to enable column search capabilities in Spark and add Spark operations to Lucene. 2) Representing Spark DataFrames as Lucene documents to build a distributed Lucene index from DataFrames. 3) Using the index for tasks like searching devices matching a query, generating statistical and predictive models on retrieved devices, and finding dimensions correlated with selected devices. 4) Architectural components like Trapezium for batch, streaming, and API services and a LuceneDAO for indexing DataFrames and querying the index.
Report
Share
Spark Summit EU talk by Debasish Das and Pramod Narasimha
- 1. FUSING APACHE SPARK AND LUCENE FOR NEAR-REALTIME PREDICTIVE MODEL BUILDING Debasish Das Principal Engineer Verizon Contributors Platform: Pankaj Rastogi, Venkat Chunduru, Ponrama Jegan, Masoud Tavazoei Algorithm: Santanu Das, Debasish Das (Dave) Frontend: Altaff Shaik, Jon Leonhardt Pramod Lakshmi Narasimha Principal Engineer Verizon
- 2. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Data Overview • Location data • Each srcIp defined as unique row key • Provides approximate location of each key • Timeseries containing latitude, longitude, error bound, duration, timezone for each key • Clickstream data • Contains clickstream data of each row key • Contains startTime, duration, httphost, httpuri, upload/download bytes, httpmethod • Compatible with IPFIX/Netflow formats 2
- 3. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Marketing Analytics 3 Lookalike modeling Churn reduction Competitive analysis Increased share of stomach • Anonymous aggregate analysis for customer insights
- 4. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Data Model • Dense dimension, dense measure !Schema: srcip, date, hour, tld, zip, tldvisits, zipvisits! !Data: 10.1.13.120, d1, H2, macys.com, 94555, 2, 4! • Sparse dimension, dense measure !Schema: srcip, date, tld, zip, clickstreamvisits, zipvisits! !Data: 10.1.13.120, d1, {macys.com, kohls.com}, {94555, 94301}, 10, 15! • Sparse dimension, sparse measure !Schema: srcip, date, tld, zip, tldvisits, zipvisits! !Data: 10.1.13.120, d1, {macys.com, kohls.com}, {94555, 94301}, {macys.com:4, kohls.com:6}, {94555:8, 94301:7}! !Schema: srcip, week, tld, zip, tldvisits, zipvisits! !Data: 10.1.13.120, week1, {macys.com, kohls.com}, {94555, 94301}, {macys.com:4, kohls.com:6}, {94555:8, 94301:7}! • Sparse dimension, sparse measure, last N days ! ! Schema: srcip, tld, zip, tldvisits, zipvisits! ! ! Data: 10.1.13.120, {macys.com, kohls.com}, {94555, 94301}, {macys.com:4, kohls.com:6}, {94555:8, 94301:7} ! • Competing technologies: PowerDrill, Druid, LinkedIn Pinot, EssBase 4
- 5. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Document Dataset Representation • Example !Schema: srcip, tld, zip, tldvisits, zipvisits! !Data: 10.1.13.120, {macys.com, kohls.com}, {94555, 94301}, {macys.com:4, kohls.com:6}, {94555:8, 94301:7} • DataFrame row to Lucene Document mapping 5 Store/schema! Row! Document! srcip! primary key! docId! tld! zip! String! Array[String]! SingleValue/MultiValue ! Indexed Fields! tldvisits! zipvisits! Double! Map[String, Double]! SparseVector ! StoredField! • Distributed collection of srcIp as RDD[Document] • ~100M srcip, 1M+ terms (sparse dimensions)
- 6. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! DeviceAnalyzer 6 • DeviceAnalyzer goals – Search and retrieve devices that matched query – Generate statistical and predictive models on retrieved devices
- 7. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! What is Trapezium ? 7 DAIS Open Source framework to build batch, streaming and API services https://github.com/Verizon/trapezium
- 8. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Trapezium Architecture 8 Trapezium D1 D2 D3 O1 O2 O3 Validation D1 V1 V1 O1 D2 O2 D3 O1 VARIOUS TRANSACTIONS
- 9. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Lucene Overview 9 • Scalable, full-text search library • Focus: Indexing + searching documents
- 10. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Trapezium LuceneDAO • SparkSQL and MLlib optimized for full scan, column indexing not supported • Why Spark + Lucene integration • Lucene is battle tested Apache Licensed Open Source Project • Adds column search capabilities to Spark • Adds spark operators (treeAggregate, treeReduce, map) to Lucene • LuceneDAO features • Build distributed lucene shards from Dataframe • Save shards to HDFS for QueryProcessor (CloudSolr) • Access saved shards through LuceneDAO for ML pipelines 10
- 11. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Trapezium Batch 11 runMode = "BATCH" dataSource = “HDFS” dependentWorkflows= { workflows=[aggregate] frequencyToCheck=100 } hdfsFileBatch = { batchTime = 86400 timerStartDelay = 1 batchInfo = [{ name = "DeviceStore" dataDirectory = {saiph-devqa=/aggregates} fileFormat = "parquet" }] } transactions = [{ transactionName = “DeviceIndexer” inputData = [{name = "DeviceStore"}] persistDataName = “indexed" }]
- 12. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! DeviceAnalyzer: Indexing 12 /?ref=1108&?url=http:// www.macys.com&id=5 www.walmart.com%2Fc%2Fep%2Frange- hood-filters&sellermemid=459 http%3A%2F%2Fm.macys.com%2Fshop %2Fproduct%2Fjockey-elance-cotton /?ref=1108&?url=http:// www.macys.com&id=5 m.amazon.com%2Fshop%2Fproduct %2Fjockey-elance-cotton https://www.walmart.com/ip/Women- Pant-Suit-Roundtree walmart://ip/?veh=dsn&wmlspartner m.macys.com%2Fshop%2Fsearch %3Fkeyword%3DDress ip1, macys.com : 2 ip1, walmart.com: 1 ip1, macys.com : 1 ip2, walmart.com: 1 ip1, amazon.com: 1 ip1, macys.com : 2 ip2, walmart.com: 1 0! 1! 2! 5 1! 1! Macys, 0 Walmart, 1 Amazon, 2 object DeviceIndexer extend BatchTransaction { process(dfs: Map[String, DataFrame], batchTime: Time): { df = dfs(“DeviceStore”) dm = generateDictionary(df) vectorizedDf = transform(df, dm) } persist(df: DataFrame, batchTime: Time): { converter = SparkLuceneConverter(dm.size) dao = LuceneDAO(batchTime,…).setConverter(converter) dm.save(path, batchTime) dao.index(df, numShards) } 1! 2 ip1 ip2
- 13. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! LuceneDAO Index Size 13 0.0! 75.0! 150.0! 225.0! 300.0! 1M! 4M! 8M! 16M! 73M! 73M all! rows! InputSize(gb)! IndexSize(gb)! rows InputSize(gb) IndexSize(gb) 1M 4.0 5.1 4M 14.4 19.0 8M 27.9 35.7 16M 58.8 63.2 73M 276.5 228.0 73M all 276.5 267.1
- 14. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! LuceneDAO Shuffle Size 14 0.! 250.! 500.! 750.! 1000.! 1M! 4M! 8M! 16M! 73M! 73M all! Dictionary(mb)! ShuffleWrite(mb)! rows ShuffleWrite(mb) DicIonary(mb) 1M 25 22.0 4M 56 30.0 8M 85 31.6 16M 126 32.2 73M 334 32.4 73M all 921 146.5
- 15. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! LuceneDAO Index Runtime 15 rows RunIme (s) 1M 135 4M 228 8M 434 16M 571 73M 1726 73M all 2456 0! 750! 1500! 2250! 3000! 1M! 4M! 8M! 16M! 73M! 73M all! #rows! Runtime (s)! 20 executors 16 cores! Executor RAM 16 GB! Driver RAM 8g!
- 16. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Trapezium Api runMode = "BATCH" dataSource = "HDFS" httpServer = { provider = "akka" hostname = "localhost" port = 19999 contextPath = "/" endPoints = [{ path = “analyzer-api" className = "TopKEndPoint" }] } 16
- 17. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! DeviceAnalyzer: Topk • Given a query select * from devices where tld=‘macys.com’ OR ‘nordstorm.com’ AND (city=‘SanFrancisco’ OR ‘Brussels’) AND (device=‘Android’) … • ML: Find topk dimensions highly correlated with selected device • BI: group by tld order by sum(visits) as tldVisits limit topk 17 class TopkController(sc: SparkContext) extends SparkServiceEndPoint(sc) { override def route : topkRoute converter = SparkLuceneConverter(dm.size) batchTime = Trapezium.getSyncTime(“indexer”) dao = LuceneDAO(batchTime…) .setConverter(converter).load(sc, indexPath) dict = loadDictionary(sc, indexPath, batchTime) def topkRoute : { post { request => { devices = dao.search(request) response = getCorrelates(devices, dict, topk) } } df[deviceId, vector] sum, support mean, median, stddev
- 18. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Trapezium Stream 18 runMode = "STREAM" dataSource = “KAFKA" kafkaTopicInfo = { consumerGroup = "KafkaStreamGroup" maxRatePerPartition = 970 batchTime = "5" streamsInfo = [{ name = "queries" topicName = “deviceanalyzer” }] } transactions = [{ transactionName = DeviceAnalyzer" inputStreams = [{name: "queries"}] persistStreamName = "deviceanalyzer" isPersist = "true" }]
- 19. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! DeviceAnalyzer: Compare 19 • Given two queries select * from Devices where tld=‘macys.com’ OR ‘nordstorm.com’ AND (city=‘SanFrancisco’) AND (device=‘Android’) select * from Devices where tld=‘macys.com’ OR ‘nordstorm.com’ AND (city=‘Brussels’) AND (device=‘Android’) • Find the dimensions that discriminate the devices associated with two groups def processStream(streams: Map[String, DStream[Row]], workflowTime: Time): { streams(“queries”).collect().map{ requests => group1 = dao.search(requests(0)) group2 = dao.search(requests(1)) response = runLDA(aud1, aud2, dict) } def persistStream(responses: RDD[Row], batchTime: Time) { HBaseDAO.write(responses) } • Sparse weighted least squares using Breeze QuadraticMinimizer • L1 Regularized logistic regression
- 20. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! DeviceAnalyzer: Augment 20 • Given a query select * from Devices where tld=‘macys.com’ OR ‘nordstorm.com’ AND (city=‘SanFrancisco’ OR ‘Brussels’) AND (device=‘Android’)… • Find devices similar to seed as lookalikes • Find dimensions that represent lookalikes object DeviceAnalyzer extends StreamingTransaction { converter = SparkLuceneConverter(dm.size) batchTime = Trapezium.getSyncTime(“indexer”) dao = LuceneDAO(batchTime…) .setConverter(converter).load(sc, indexPath) dict = loadDictionary(sc, indexPath, batchTime) all = dao.search(“*:*”) def processStream(streams: Map[String, DStream[Row]]) : { streams(“queries”).collect().map{ request => audience = dao.search(request) response = getLookalikeDimensions(all, audience, dict) } • Sparse weighted least squares using Breeze QuadraticMinimizer • L2 regularized linear regression
- 21. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! FastSummarizer • Statistical and predictive operators • sum: sum over numeric measures • support: sum over distinct docID • sumSquared: L2 norm • gram: Uses BLAS sspr • solve: Uses Breeze QuadraticMinimizer to support L1 • Implemented using Array[Float] for shuffle opt • Scala/Java for Level1 operations • OpenBLAS for Level3 operations 21
- 22. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Sync API Benchmark 73M rows 1M+ search terms 1 measure on 250K sparse dimensions 20 executors 8 cores 32 GB driver RAM 16 GB executor RAM akka-http cores: 24 default 22 topk! qps! runtime(s)! 1! 1.389! 5! 1.663! 10! 3.214! 20! 5.992! 40! 12.174! 0.! 3.5! 7.! 10.5! 14.! 1! 5! 10! 20! 40! qps! runtime(s)!
- 23. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! Async API Benchmark 73M rows, 1M+ search terms 1 measure on 250K sparse dimensions 20 executors 8 cores 32 GB driver RAM 16 GB executor RAM forkjoinpool = 40 Kafka Fetch + compare/augment + HBase Persist 23 predictions! qps! compare(s)! augment(s)! 1! 9! 16! 5! 13! 36! 10! 23! 70! 20! 42! 142! 0! 40! 80! 120! 160! 1! 5! 10! 20! qps! compare! augment!
- 24. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! topk tld + apps 24
- 25. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! 25 Augment: Auto Enthusiastic
- 26. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! 26 Augment Model Performance
- 27. © Verizon 2016 All Rights Reserved! Information contained herein is provided AS IS and subject to change without notice. All trademarks used herein are property of their respective owners.! 27 Compare: Leisure vs Business Travellers
- 28. THANK YOU. Q&A Join us and make machines intelligent Data & Artificial Intelligence Systems 499 Hamilton Ave, Palo Alto California