DataStax & O'Reilly Media: Large Scale Data Analytics with Spark and Cassandra on the DataStax Enterprise Platform - A Hands on Workshop

Large Scale Data Analytics

Large Scale Data Analytics

Ryan Knight @Knight_Cloud

Solution Engineer - DataStax

Paco Nathan @pacoid

Evil Mad Scientist - O’Reilly Media

Demo of Streaming in the Real World - Spark At Scale Project

3© 2015. All Rights Reserved.

•Based on Real World Use Cases

•Simulate a real world streaming use case

•Test throughput of Spark Streaming

•Best Practices for scaling

•https://github.com/retroryan/SparkAtScale

Spark At Scale Demo Application

4© 2015. All Rights Reserved.

DataStax Enterprise Platform

DataStax Enterprise Platform

How do we Scale for Load and Traffic?

Data Modeling using Event Sourcing

6© 2015. All Rights Reserved.

•Append-Only Logging

•Database of Facts

•Snapshots or Roll-Ups

•Why Delete Data any more?

•Replay Events

Scala for Large Scale Data Analytics

7© 2015. All Rights Reserved.

•Functional Paradigm is ideal for Data Analytics

•Strongly Typed - Enforce Schema at Every Later

•Immutable by Default - Event Logging

•Declarative instead of Imperative - Focus on Transformation not Implementation

Key to Scaling - Configuring Kafka Topics

8© 2015. All Rights Reserved.

•Number of Partitions per Topic — Degree of parallelism

•Directly Affects Spark Streaming Parallelism

•bin/kafka-topics.sh --create --zookeeper localhost:2181 --replication-factor 1 --partitions 5 --topic ratings

Populating Kafka Topics

9© 2015. All Rights Reserved.

val record = new ProducerRecord[String, String] (feederExtension.kafkaTopic, partNum, key, nxtRating.toString)

val future = feederExtension.producer.send(record, new Callback {

10© 2015. All Rights Reserved.

Spark Streaming with Kafka Direct Approach

11© 2015. All Rights Reserved.

•Use Kafka Direct Approach (No Receivers)

•Queries Kafka Directly

•Automatically Parallelizes based on Kafka Partitions

•Exactly Once Processing - Only Move Offset after Processing

•Resiliency without copying data

12© 2015. All Rights Reserved.

Spark Streaming Monitoring

13© 2015. All Rights Reserved.

Processing Time

>Batch Duration

=Total Delay Grows

Out Of Memory Errors

© 2014 DataStax, All Rights Reserved.

Confidential

DataStax Enterprise Platform Workload Segregation w/out ETL

14

Cassandra Mode OLTP Database

Analytics Mode Streaming and Analytics

Search Mode All Data Searchable

C*

C

C

S A

A

DataStax Enterprise Platform Workload Segregation w/out ETL

15© 2015. All Rights Reserved.

DataStax Enterprise Platform Integrated Spark Analytics

16© 2015. All Rights Reserved.

DataStax Analytics

17© 2015. All Rights Reserved.

•Simplified Deployment and Management •HA Spark Master with automatic leader election

•Detects when Spark Master is down with gossip •Uses Paxos to elect Spark Master

•Stores Spark Worker metadata in Cassandra •No need to run Zookeeper

Spark Notebook

18© 2015. All Rights Reserved.

C*

C

C A

AANotebook

Notebook

Notebook

Spark Notebook ServerCassandra Cluster with Spark Connector

Apache Spark Notebook

19© 2015. All Rights Reserved.

•Reactive / Dynamic Graphs base on Scala, SQL and DataFrames

•Spark Streaming • Examples notebooks covering visualization, machine

learning, streaming, graph analysis, genomics analysis •SVG / Sliders - interactive graphs •Tune and Configure Each Notebook Separately •https://github.com/andypetrella/spark-notebook

databricks.gitbooks.io/databricks-spark-reference-applications/content/twitter_classifier/README.html

Demo: Twitter Streaming Language Classifier

Streaming:collect tweets

Twitter API

HDFS:dataset

Spark SQL:ETL, queries

MLlib:train classifier

Spark:featurize

HDFS:model

Streaming:score tweets

language filter

Demo: Twitter Streaming Language Classifier

Cassandra

Cassandra

1. extract text from the tweet

https://twitter.com/andy_bf/status/16222269370011648

"Ceci n'est pas un tweet"

2. sequence text as bigrams

tweet.sliding(2).toSeq ("Ce", "ec", "ci", …, )

3. convert bigrams into numbers

seq.map(_.hashCode()) (2178, 3230, 3174, …, )

4. index into sparse tf vector

seq.map(_.hashCode() % 1000) (178, 230, 174, …, )

5. increment feature count

Vector.sparse(1000, …) (1000, [102, 104, …], [0.0455, 0.0455, …])

Demo: Twitter Streaming Language Classifier

From tweets to ML features, approximated as sparse vectors:

KMeans: Formal Definition (ignore this)

KMeans: How it really works…

KMeans: How it really works…

Demo: Twitter Streaming Language Classifier

Sample Code + Output: https://github.com/retroryan/twitter_classifier

val sc = new SparkContext(new SparkConf())

val ssc = new StreamingContext(conf, Seconds(5))

val tweets = TwitterUtils.createStream(ssc, Utils.getAuth)

val statuses = tweets.map(_.getText)

val model = new KMeansModel(ssc.sparkContext.objectFile[Vector]

(modelFile.toString).collect())

val filteredTweets = statuses

.filter(t =>

model.predict(Utils.featurize(t)) == clust)

filteredTweets.print()

ssc.start()

ssc.awaitTermination()

CLUSTER 1:TLあんまり見ないけど@くれたっらいつでもくっるよ٩(δωδ)۶

そういえばディスガイアも今日か CLUSTER 4:قالوا العروبه روحت بعد صدامواقول مع سلمان تحيى العروبهRT @vip588: √ للمتواجدين االن √ زيادة متابعني √ فولو مي vip588

فولو باك √ رتويت للتغريدة √ فولو للي عمل رتويت √ اللي ما يلتزم ما √… بيستفيدن سورة

Thank you