Se ha denunciado esta presentación.
Utilizamos tu perfil de LinkedIn y tus datos de actividad para personalizar los anuncios y mostrarte publicidad más relevante. Puedes cambiar tus preferencias de publicidad en cualquier momento.

Spark Job Server and Spark as a Query Engine (Spark Meetup 5/14)

11.440 visualizaciones

Publicado el

This was a talk that Kelvin Chu and I just gave at the SF Bay Area Spark Meetup 5/14 at Palantir Technologies.

We discussed the Spark Job Server (, its history, example workflows, architecture, and exciting future plans to provide HA spark job contexts.

We also discussed the use case of the job server at Ooyala to facilitate fast query jobs using shared RDD and a shared job context, and how we integrate with Apache Cassandra.

Spark Job Server and Spark as a Query Engine (Spark Meetup 5/14)

  1. 1. Date Spark Job Server Evan Chan and Kelvin Chu
  2. 2. Overview • REST API for Spark jobs and contexts. Easily operate Spark from any language or environment. • Runs jobs in their own Contexts or share 1 context amongst jobs • Great for sharing cached RDDs across jobs and low-latency jobs • Works with Standalone, Mesos, any Spark config • Jars, job history and config are persisted via a pluggable API • Async and sync API, JSON job results
  3. 3. Open Source!!
  4. 4. History
  5. 5. CONFIDENTIAL—DO NOT DISTRIBUTE 5 Founded in 2007 Commercially launched in 2009 300+ employees in Silicon Valley, LA, NYC, 
 London, Paris, Tokyo, Sydney & Guadalajara Global footprint, 200M unique users,
110+ countries, and more than 6,000 websites Over 1 billion videos played per month 
and 2 billion analytic events per day 25% of U.S. online viewers watch video 
 powered by Ooyala Ooyala, Inc.
  6. 6. Spark at Ooyala • Started investing in Spark beginning of 2013 • Developers loved it, promise of a unifying platform • 2 teams of developers building on Spark • Actively contributing to the Spark community • Largest Spark cluster has > 100 nodes • Spark community very active, huge amount of interest
  7. 7. From raw logs to fast queries Processing C*
 columnar store Raw Log Files Raw Log Files Raw Log Files Spark Spark Spark View 1 View 2 View 3 Spark Shark Predefined queries Ad-hoc HiveQL
  8. 8. Our Spark/Shark/Cassandra Stack Node1 Cassandra SerDe Spark Worker Shark Node2 Cassandra SerDe Spark Worker Shark Node3 Cassandra SerDe Spark Worker Shark Spark Master Job Server
  9. 9. WhyWe Needed a Job Server • Our vision for Spark is as a multi-team big data service • What gets repeated by every team: • Bastion box for running Hadoop/Spark jobs • Deploys and process monitoring • Tracking and serializing job status, progress, and job results • Job validation • No easy way to kill jobs • Polyglot technology stack - Ruby scripts run jobs, Go services
  10. 10. ExampleWorkflow
  11. 11. Creating a Job Server Project ✤ sbt assembly -> fat jar -> upload to job server! ✤ "provided" is used. Don’t want SBT assembly to include the whole job server jar.! ✤ Java projects should be possible too resolvers += "Ooyala Bintray" at "" ! libraryDependencies += "ooyala.cnd" % "job-server" % "0.3.1" % "provided" ✤ In your build.sbt, add this
  12. 12. Example Job Server Job /**! * A super-simple Spark job example that implements the SparkJob trait and! * can be submitted to the job server.! */! object WordCountExample extends SparkJob {! override def validate(sc: SparkContext, config: Config): SparkJobValidation = {! Try(config.getString(“input.string”))! .map(x => SparkJobValid)! .getOrElse(SparkJobInvalid(“No input.string”))! }! ! override def runJob(sc: SparkContext, config: Config): Any = {! val dd = sc.parallelize(config.getString(“input.string”).split(" ").toSeq)!, 1)).reduceByKey(_ + _).collect().toMap! }! }!
  13. 13. What’s Different? • Job does not create Context, Job Server does • Decide when I run the job: in own context, or in pre-created context • Upload new jobs to diagnose your RDD issues: • POST /contexts/newContext • POST /jobs .... context=newContext • Upload a new diagnostic jar... POST /jars/newDiag • Run diagnostic jar to dump into on cached RDDs
  14. 14. Submitting and Running a Job ✦ curl --data-binary @../target/mydemo.jar localhost:8090/jars/demo OK[11:32 PM] ~ ! ✦ curl -d "input.string = A lazy dog jumped mean dog" 'localhost:8090/jobs? appName=demo&classPath=WordCountExample&sync=true' { "status": "OK", "RESULT": { "lazy": 1, "jumped": 1, "A": 1, "mean": 1, "dog": 2 } }
  15. 15. Retrieve Job Statuses ~/s/jobserver (evan-working-1 ↩=) curl 'localhost:8090/jobs?limit=2' [{ "duration": "77.744 secs", "classPath": "ooyala.cnd.CreateMaterializedView", "startTime": "2013-11-26T20:13:09.071Z", "context": "8b7059dd-ooyala.cnd.CreateMaterializedView", "status": "FINISHED", "jobId": "9982f961-aaaa-4195-88c2-962eae9b08d9" }, { "duration": "58.067 secs", "classPath": "ooyala.cnd.CreateMaterializedView", "startTime": "2013-11-26T20:22:03.257Z", "context": "d0a5ebdc-ooyala.cnd.CreateMaterializedView", "status": "FINISHED", "jobId": "e9317383-6a67-41c4-8291-9c140b6d8459" }]
  16. 16. Use Case: Fast Query Jobs
  17. 17. Spark as a Query Engine ✤ Goal: spark jobs that run in under a second and answers queries on shared RDD data! ✤ Query params passed in as job config! ✤ Need to minimize context creation overhead! ✤ Thus many jobs sharing the same SparkContext! ✤ On-heap RDD caching means no serialization loss! ✤ Need to consider concurrent jobs (fair scheduling)
  18. 18. LOW-LATENCY QUERY JOBS RDDLoad Data Query Job Spark
 Executors Cassandra REST Job Server Query Job Query Result Query Result new SparkContext Create query context Load some data
  19. 19. Sharing Data Between Jobs ✤ RDD Caching! ✤ Benefit: no need to serialize data. Especially useful for indexes etc.! ✤ Job server provides a NamedRdds trait for thread-safe CRUD of cached RDDs by name! ✤ (Compare to SparkContext’s API which uses an integer ID and is not thread safe)! ✤ For example, at Ooyala a number of fields are multiplexed into the RDD name: timestamp:customerID:granularity
  20. 20. Data Concurrency ✤ Single writer, multiple readers! ✤ Managing multiple updates to RDDs! ✤ Cache keeps track of which RDDs being updated! ✤ Example: thread A spark job creates RDD “A” at t0! ✤ thread B fetches RDD “A” at t1 > t0! ✤ Both threads A and B, using NamedRdds, will get the RDD at time t2 when thread A finishes creating the RDD “A”
  21. 21. UsingTachyon Pros Cons Off-heap storage: No GC ByteBuffer API - need to pay deserialization cost Can be shared across multiple processes Data can survive process loss Backed by HDFS Does not support random access writes
  22. 22. Architecture
  23. 23. Completely Async Design ✤ - probably the fastest JVM HTTP microframework! ✤ Akka Actor based, non blocking! ✤ Futures used to manage individual jobs. (Note that Spark is using Scala futures to manage job stages now)! ✤ Single JVM for now, but easy to distribute later via remote Actors / Akka Cluster
  24. 24. Async Actor Flow Spray web API Request actor Local Supervisor Job Manager Job 1 Future Job 2 Future Job Status Actor Job Result Actor
  25. 25. Message flow fully documented
  26. 26. Production Usage
  27. 27. Metadata Store ✤ JarInfo, JobInfo, ConfigInfo! ✤ JobSqlDAO. Store metadata to SQL database by JDBC interface.! ✤ Easily configured by spark.sqldao.jdbc.url! ✤ jdbc:mysql://dbserver:3306/jobserverdb ✤ Multiple Job Servers can share the same MySQL.! ✤ Jars uploaded once but accessible by all servers.! ✤ The default will be JobSqlDAO and H2.! ✤ Single H2 DB file. Serialization and deserialization are handled by H2.
  28. 28. Deployment and Metrics ✤ spark-jobserver repo comes with a full suite of tests and deploy scripts:! ✤ for regular server pushes! ✤ for Mesos and Chronos .tar.gz! ✤ /metricz route for codahale-metrics monitoring! ✤ /healthz route for health check0o
  29. 29. Challenges and Lessons • Spark is based around contexts - we need a Job Server oriented around logical jobs • Running multiple SparkContexts in the same process • Global use of System properties makes it impossible to start multiple contexts at same time (but see pull request...) • Have to be careful with SparkEnv • Dynamic jar and class loading is tricky • Manage threads carefully - each context uses lots of threads
  30. 30. FutureWork
  31. 31. Future Plans ✤ Spark-contrib project list. So this and other projects can gain visibility! (SPARK-1283)! ✤ HA mode using Akka Cluster or Mesos! ✤ HA and Hot Failover for Spark Drivers/Contexts! ✤ REST API for job progress! ✤ Swagger API documentation
  32. 32. HA and Hot Failover for Jobs Job Server 1 Job Server 2 Active Job Context HDFS Standby Job Context Gossip Checkpoint ✤ Job context dies:! ✤ Job server 2 notices and spins up standby context, restores checkpoint
  33. 33. Thanks for your contributions! ✤ All of these were community contributed:! ✤ index.html main page! ✤ saving and retrieving job configuration! ✤ Your contributions are very welcome on Github!
  34. 34. Thank you! And Everybody is Hiring!!