Introduction to Apache Spark and Zeppelin on Google Cloud Dataproc

Want to learn more about using Apache Spark and Zeppelin on Dataproc via the Google Cloud Platform? You’ve come to the right place.

Cloud Dataproc is Google’s answer to Amazon EMR (Elastic MapReduce). Like EMR, Cloud Dataproc provisions and manage Compute Engine-based Apache Hadoop and Spark data processing clusters.
If you are not familiar with Amazon EMR, check out my two-part series about using Apache Spark and Zeppelin on EMR – See Part 1 and Part 2. You may find my notes on IAM helpful, too.

First Impressions: The Key Differences between EMR and Cloud Dataproc

Cloud Dataproc is fairly new. It was first released in beta last September, and is now generally available since February this year. If you have previously used EMR, you may find Cloud Dataproc familiar.
An important difference I have observed is this: In EMR, when you create a cluster, you know exactly what you are installing because you are presented with an option to choose from a list of supported Hadoop components. Cloud Dataproc, on the other hand, will just install all the supported components by default.
EMR is a more mature platform. After all, it has been around since 2009. It has support for many applications, including Tez, Ganglia, Presto, HBase, Pig, Hive, Mahout, Sqoop, and Zeppelin. As for Cloud Dataproc, it only supports Hadoop, Spark, Hive, and Pig (see the supported Cloud Dataproc versions page). Fortunately, you can specify initialization actions when creating a Cloud Dataproc cluster so that you can install the additional software you need.

Installing Zeppelin on Cloud Dataproc

We will go through the steps to do exactly that when we set up Zeppelin with Spark on Cloud Dataproc. Why Zeppelin? It’s an innovative web-based notebook that enables interactive data analytics.
With Zeppelin, you can create data-driven documents based on a variety of different backends, including Hadoop. It’s a great starting project, so let’s jump right into it.

Our Assumptions

Creating a Cloud Dataproc Cluster

Google provided a collection of initialization actions that we can use to install additional (but unsupported) Hadoop components when we create a cluster. For this example, we will use the Zeppelin initialization action.
To use an initialization action, we need to access the initialization action script in a Cloud Storage bucket.
We will not use the publicly-accessible gs://dataproc-initialization-actions Cloud Storage bucket as instructed in the README. At the time of writing, the version of the Zeppelin initialization action script is outdated. If we were to create a cluster with it, we would encounter errors. Let’s upload the script to our own Cloud Storage bucket instead.

$ git clone https://github.com/GoogleCloudPlatform/dataproc-initialization-actions.git
Cloning into 'dataproc-initialization-actions'...
remote: Counting objects: 267, done.
remote: Total 267 (delta 0), reused 0 (delta 0), pack-reused 266
Receiving objects: 100% (267/267), 89.24 KiB | 63.00 KiB/s, done.
Resolving deltas: 100% (88/88), done.
Checking connectivity... done.
$ cd dataproc-initialization-actions/apache-zeppelin/
$ gsutil mb gs://cloudacademy/
Creating gs://cloudacademy/...
$ gsutil cp zeppelin.sh gs://cloudacademy/
Copying file://zeppelin.sh [Content-Type=application/x-sh]...
Uploading   gs://cloudacademy/zeppelin.sh:                       4.47 KiB/4.47 KiB

Next, we will issue the gcloud command to set up a Cloud Dataproc cluster.

$ gcloud dataproc clusters create spark-zeppelin \
> --initialization-actions gs://cloudacademy/zeppelin.sh \
> --initialization-action-timeout 15m
Waiting on operation [projects/operating-spot-133003/regions/global/operations/cdf1fadd-032d-4261-9520-c2f55f8c46fa].
Waiting for cluster creation operation...done.
Created [https://dataproc.googleapis.com/v1/projects/operating-spot-133003/regions/global/clusters/spark-zeppelin].
clusterName: spark-zeppelin
clusterUuid: f05a9f22-5ee6-48c8-83d3-7079e2d1d834
config:
  configBucket: dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia
  gceClusterConfig:
    networkUri: https://www.googleapis.com/compute/v1/projects/operating-spot-133003/global/networks/default
    serviceAccountScopes:
    - https://www.googleapis.com/auth/bigquery
    - https://www.googleapis.com/auth/bigtable.admin.table
    - https://www.googleapis.com/auth/bigtable.data
    - https://www.googleapis.com/auth/cloud.useraccounts.readonly
    - https://www.googleapis.com/auth/devstorage.full_control
    - https://www.googleapis.com/auth/devstorage.read_write
    - https://www.googleapis.com/auth/logging.write
    zoneUri: https://www.googleapis.com/compute/v1/projects/operating-spot-133003/zones/asia-east1-a
  initializationActions:
  - executableFile: gs://cloudacademy/zeppelin.sh
    executionTimeout: 900.000s
  masterConfig:
    diskConfig:
      bootDiskSizeGb: 500
    imageUri: https://www.googleapis.com/compute/v1/projects/cloud-dataproc/global/images/dataproc-1-0-20160516-190717
    instanceNames:
    - spark-zeppelin-m
    machineTypeUri: https://www.googleapis.com/compute/v1/projects/operating-spot-133003/zones/asia-east1-a/machineTypes/n1-standard-4
    numInstances: 1
  softwareConfig:
    imageVersion: '1.0'
    properties:
      distcp:mapreduce.map.java.opts: -Xmx2457m
      distcp:mapreduce.map.memory.mb: '3072'
      distcp:mapreduce.reduce.java.opts: -Xmx4915m
      distcp:mapreduce.reduce.memory.mb: '6144'
      mapred:mapreduce.map.cpu.vcores: '1'
      mapred:mapreduce.map.java.opts: -Xmx2457m
      mapred:mapreduce.map.memory.mb: '3072'
      mapred:mapreduce.reduce.cpu.vcores: '2'
      mapred:mapreduce.reduce.java.opts: -Xmx4915m
      mapred:mapreduce.reduce.memory.mb: '6144'
      mapred:yarn.app.mapreduce.am.command-opts: -Xmx4915m
      mapred:yarn.app.mapreduce.am.resource.cpu-vcores: '2'
      mapred:yarn.app.mapreduce.am.resource.mb: '6144'
      spark:spark.driver.maxResultSize: 1920m
      spark:spark.driver.memory: 3840m
      spark:spark.executor.cores: '2'
      spark:spark.executor.memory: 5586m
      spark:spark.yarn.am.memory: 5586m
      spark:spark.yarn.am.memoryOverhead: '558'
      spark:spark.yarn.executor.memoryOverhead: '558'
      yarn:yarn.nodemanager.resource.memory-mb: '12288'
      yarn:yarn.scheduler.maximum-allocation-mb: '12288'
      yarn:yarn.scheduler.minimum-allocation-mb: '1024'
  workerConfig:
    diskConfig:
      bootDiskSizeGb: 500
    imageUri: https://www.googleapis.com/compute/v1/projects/cloud-dataproc/global/images/dataproc-1-0-20160516-190717
    instanceNames:
    - spark-zeppelin-w-0
    - spark-zeppelin-w-1
    machineTypeUri: https://www.googleapis.com/compute/v1/projects/operating-spot-133003/zones/asia-east1-a/machineTypes/n1-standard-4
    numInstances: 2
projectId: operating-spot-133003
status:
  state: RUNNING
  stateStartTime: '2016-06-07T09:35:33.783Z'
statusHistory:
- state: CREATING
  stateStartTime: '2016-06-07T09:32:56.398Z'

By default, Cloud Dataproc clusters use n1-standard-4 machine type for the master and worker nodes. These are the standard instances with 4 virtual CPUs and 15GB of memory. You can change the defaults by specifying the relevant flags. See the gcloud dataproc clusters create documentation.

$ gcloud compute instances list
NAME                ZONE          MACHINE_TYPE   PREEMPTIBLE  INTERNAL_IP  EXTERNAL_IP      STATUS
spark-zeppelin-m    asia-east1-a  n1-standard-4               10.140.0.2   104.199.169.41   RUNNING
spark-zeppelin-w-0  asia-east1-a  n1-standard-4               10.140.0.4   104.199.160.183  RUNNING
spark-zeppelin-w-1  asia-east1-a  n1-standard-4               10.140.0.3   130.211.248.2    RUNNING

SSH to the Master Node

Now we can connect to the master node remotely. Instead of running ssh directly, we can issue the gcloud compute ssh spark-zeppelin-m command.

$ gcloud compute ssh spark-zeppelin-m
WARNING: The private SSH key file for Google Compute Engine does not exist.
WARNING: You do not have an SSH key for Google Compute Engine.
WARNING: [/usr/bin/ssh-keygen] will be executed to generate a key.
This tool needs to create the directory [/Users/eugeneteo/.ssh] before being
able to generate SSH keys.
Do you want to continue (Y/n)?
Generating public/private rsa key pair.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /Users/eugeneteo/.ssh/google_compute_engine.
Your public key has been saved in /Users/eugeneteo/.ssh/google_compute_engine.pub.
The key fingerprint is:
SHA256:xTjbTqWIfroXr670RWQkscOPfGdY6rzpi14WCoFI5Ko eugeneteo@eugeneteos-MacBook-Pro.local
The key's randomart image is:
+---[RSA 2048]----+
| .o     o..      |
| o . . . *       |
|  o . . * = o    |
| .     + % *     |
|.     o S @ o    |
|.    . ..O +     |
|E     o ooB      |
|     . +.=.o     |
|      +**o=.     |
+----[SHA256]-----+
Updated [https://www.googleapis.com/compute/v1/projects/operating-spot-133003].
Warning: Permanently added 'compute.5792161037633137215' (ECDSA) to the list of known hosts.
The programs included with the Debian GNU/Linux system are free software;
the exact distribution terms for each program are described in the
individual files in /usr/share/doc/*/copyright.
Debian GNU/Linux comes with ABSOLUTELY NO WARRANTY, to the extent
permitted by applicable law.
eugeneteo@spark-zeppelin-m:~$

Spark’s Scala Shell

We will not cover the Spark programming model in this article, but we will learn just enough to start an interpreter on the command-line and to make sure it works. We will launch spark-shell on YARN.

$ spark-shell --master=yarn
16/06/07 09:53:32 INFO org.spark-project.jetty.server.Server: jetty-8.y.z-SNAPSHOT
16/06/07 09:53:32 INFO org.spark-project.jetty.server.AbstractConnector: Started SocketConnector@0.0.0.0:36325
Welcome to
      ____              __
     / __/__  ___ _____/ /__
    _\ \/ _ \/ _ `/ __/  '_/
   /___/ .__/\_,_/_/ /_/\_\   version 1.6.1
      /_/
Using Scala version 2.10.5 (OpenJDK 64-Bit Server VM, Java 1.8.0_72-internal)
Type in expressions to have them evaluated.
Type :help for more information.
16/06/07 09:53:38 INFO akka.event.slf4j.Slf4jLogger: Slf4jLogger started
16/06/07 09:53:38 INFO Remoting: Starting remoting
16/06/07 09:53:38 INFO Remoting: Remoting started; listening on addresses :[akka.tcp://sparkDriverActorSystem@10.140.0.2:40692]
16/06/07 09:53:39 INFO org.spark-project.jetty.server.Server: jetty-8.y.z-SNAPSHOT
16/06/07 09:53:39 INFO org.spark-project.jetty.server.AbstractConnector: Started SelectChannelConnector@0.0.0.0:4040
16/06/07 09:53:39 INFO org.apache.hadoop.yarn.client.RMProxy: Connecting to ResourceManager at spark-zeppelin-m/10.140.0.2:8032
16/06/07 09:53:43 INFO org.apache.hadoop.yarn.client.api.impl.YarnClientImpl: Submitted application application_1465292021581_0001
[...]
scala> val data = sc.parallelize(1 to 100000)
data: org.apache.spark.rdd.RDD[Int] = ParallelCollectionRDD[0] at parallelize at <console>:27
scala> data.filter(_ < 100).collect()
res0: Array[Int] = Array(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99)

It worked!

Access the Zeppelin Notebook

Before we can access the Zeppelin Notebook, we will need to create a SSH tunnel to the master node.

$ gcloud compute ssh --ssh-flag="-D 31337" --ssh-flag="-N" --ssh-flag="-n" spark-zeppelin-m

Configure your web browser to use the SOCKS proxy localhost:31337.
Having done that, we can now access http://localhost:8080/.
zeppelin1
zeppelin2
Great! The notebook works, too!

Terminating the Cloud Dataproc Cluster

Always remember to terminate your cluster after you have completed your work to avoid spending more money than you have planned. We are billed by the minute, based on the size of our cluster and the duration we ran our jobs.

$ gcloud dataproc clusters delete spark-zeppelin
The cluster 'spark-zeppelin' and all attached disks will be deleted.
Do you want to continue (Y/n)?  y
Waiting on operation [projects/operating-spot-133003/regions/global/operations/e2b88d41-e215-480e-a970-e6b49c0de574].
Waiting for cluster deletion operation...done.
Deleted [https://dataproc.googleapis.com/v1/projects/operating-spot-133003/regions/global/clusters/spark-zeppelin].
$ gsutil -m rm -r gs://cloudacademy/ gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/
Removing gs://cloudacademy/zeppelin.sh#1465291266953000...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/google-cloud-dataproc-metainfo/f05a9f22-5ee6-48c8-83d3-7079e2d1d834/cluster.properties#1465291982908000...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/google-cloud-dataproc-metainfo/f05a9f22-5ee6-48c8-83d3-7079e2d1d834/spark-zeppelin-m/dataproc-initialization-script-0_output#1465292128609000...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/google-cloud-dataproc-metainfo/f05a9f22-5ee6-48c8-83d3-7079e2d1d834/spark-zeppelin-m/dataproc-startup-script_SUCCESS#1465292128628000...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/google-cloud-dataproc-metainfo/f05a9f22-5ee6-48c8-83d3-7079e2d1d834/spark-zeppelin-m/dataproc-startup-script_output#1465292038813000...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/google-cloud-dataproc-metainfo/f05a9f22-5ee6-48c8-83d3-7079e2d1d834/spark-zeppelin-w-0/dataproc-initialization-script-0_output#1465292016529000...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/google-cloud-dataproc-metainfo/f05a9f22-5ee6-48c8-83d3-7079e2d1d834/spark-zeppelin-w-0/dataproc-startup-script_SUCCESS#1465292016563000...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/google-cloud-dataproc-metainfo/f05a9f22-5ee6-48c8-83d3-7079e2d1d834/spark-zeppelin-w-0/dataproc-startup-script_output#1465292015001000...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/google-cloud-dataproc-metainfo/f05a9f22-5ee6-48c8-83d3-7079e2d1d834/spark-zeppelin-w-1/dataproc-initialization-script-0_output#1465292016544000...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/google-cloud-dataproc-metainfo/f05a9f22-5ee6-48c8-83d3-7079e2d1d834/spark-zeppelin-w-1/dataproc-startup-script_SUCCESS#1465292016575000...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/google-cloud-dataproc-metainfo/f05a9f22-5ee6-48c8-83d3-7079e2d1d834/spark-zeppelin-w-1/dataproc-startup-script_output#1465292014612000...
Removing gs://cloudacademy/...
Removing gs://dataproc-6dfb6ea8-847d-438c-befd-2883003f61e5-asia/...

What We’ve Learned

In this article, we have learned:

  1. How to set up a Cloud Dataproc cluster with Zeppelin.
  2. We have launched spark-shell on YARN.
  3. We have also set up a SSH tunnel to access the Zeppelin Notebook from the master node.

Related Courses

Data Management on Google Cloud Platform might be a place to start. It’s 26 minutes of data management goodness from David Clinton, an expert Linux Sysadmin.

For a guided instructional experience, check out a Cloud Academy Learning Path. We offer a free 7-day trial subscription with access all video courses, self-test quizzes, and labs. Our labs are a great tool for applying what you learned and testing your understanding in a live environment. Get started today!

Avatar

Written by

Eugene Teo

Eugene Teo is a director of security at a US-based technology company. He is interested in applying machine learning techniques to solve problems in the security domain.


Related Posts

Valery Calderón Briz
Valery Calderón Briz
— October 22, 2019

How to Go Serverless Like a Pro

So, no servers? Yeah, I checked and there are definitely no servers. Well...the cloud service providers do need servers to host and run the code, but we don’t have to worry about it. Which operating system to use, how and when to run the instances, the scalability, and all the arch...

Read more
  • AWS
  • Lambda
  • Serverless
Avatar
Stuart Scott
— October 16, 2019

AWS Security: Bastion Host, NAT instances and VPC Peering

Effective security requires close control over your data and resources. Bastion hosts, NAT instances, and VPC peering can help you secure your AWS infrastructure. Welcome to part four of my AWS Security overview. In part three, we looked at network security at the subnet level. This ti...

Read more
  • AWS
Avatar
Sudhi Seshachala
— October 9, 2019

Top 13 Amazon Virtual Private Cloud (VPC) Best Practices

Amazon Virtual Private Cloud (VPC) brings a host of advantages to the table, including static private IP addresses, Elastic Network Interfaces, secure bastion host setup, DHCP options, Advanced Network Access Control, predictable internal IP ranges, VPN connectivity, movement of interna...

Read more
  • AWS
  • best practices
  • VPC
Avatar
Stuart Scott
— October 2, 2019

Big Changes to the AWS Certification Exams

With AWS re:Invent 2019 just around the corner, we can expect some early announcements to trickle through with upcoming features and services. However, AWS has just announced some big changes to their certification exams. So what’s changing and what’s new? There is a brand NEW ...

Read more
  • AWS
  • Certifications
Alisha Reyes
Alisha Reyes
— October 1, 2019

New on Cloud Academy: ITIL® 4, Microsoft 365 Tenant, Jenkins, TOGAF® 9.1, and more

At Cloud Academy, we're always striving to make improvements to our training platform. Based on your feedback, we released some new features to help make it easier for you to continue studying. These new features allow you to: Remove content from “Continue Studying” section Disc...

Read more
  • AWS
  • Azure
  • Google Cloud Platform
  • ITIL® 4
  • Jenkins
  • Microsoft 365 Tenant
  • New content
  • Product Feature
  • Python programming
  • TOGAF® 9.1
Avatar
Stuart Scott
— September 27, 2019

AWS Security Groups: Instance Level Security

Instance security requires that you fully understand AWS security groups, along with patching responsibility, key pairs, and various tenancy options. As a precursor to this post, you should have a thorough understanding of the AWS Shared Responsibility Model before moving onto discussi...

Read more
  • AWS
  • instance security
  • Security
  • security groups
Avatar
Jeremy Cook
— September 17, 2019

Cloud Migration Risks & Benefits

If you’re like most businesses, you already have at least one workload running in the cloud. However, that doesn’t mean that cloud migration is right for everyone. While cloud environments are generally scalable, reliable, and highly available, those won’t be the only considerations dri...

Read more
  • AWS
  • Azure
  • Cloud Migration
Joe Nemer
Joe Nemer
— September 12, 2019

Real-Time Application Monitoring with Amazon Kinesis

Amazon Kinesis is a real-time data streaming service that makes it easy to collect, process, and analyze data so you can get quick insights and react as fast as possible to new information.  With Amazon Kinesis you can ingest real-time data such as application logs, website clickstre...

Read more
  • amazon kinesis
  • AWS
  • Stream Analytics
  • Streaming data
Joe Nemer
Joe Nemer
— September 6, 2019

Google Cloud Functions vs. AWS Lambda: The Fight for Serverless Cloud Domination

Serverless computing: What is it and why is it important? A quick background The general concept of serverless computing was introduced to the market by Amazon Web Services (AWS) around 2014 with the release of AWS Lambda. As we know, cloud computing has made it possible for users to ...

Read more
  • AWS
  • Azure
  • Google Cloud Platform
Joe Nemer
Joe Nemer
— September 3, 2019

Google Vision vs. Amazon Rekognition: A Vendor-Neutral Comparison

Google Cloud Vision and Amazon Rekognition offer a broad spectrum of solutions, some of which are comparable in terms of functional details, quality, performance, and costs. This post is a fact-based comparative analysis on Google Vision vs. Amazon Rekognition and will focus on the tech...

Read more
  • Amazon Rekognition
  • AWS
  • Google Cloud Platform
  • Google Vision
Alisha Reyes
Alisha Reyes
— August 30, 2019

New on Cloud Academy: CISSP, AWS, Azure, & DevOps Labs, Python for Beginners, and more…

As Hurricane Dorian intensifies, it looks like Floridians across the entire state might have to hunker down for another big one. If you've gone through a hurricane, you know that preparing for one is no joke. You'll need a survival kit with plenty of water, flashlights, batteries, and n...

Read more
  • AWS
  • Azure
  • Google Cloud Platform
  • New content
  • Product Feature
  • Python programming
Joe Nemer
Joe Nemer
— August 27, 2019

Amazon Route 53: Why You Should Consider DNS Migration

What Amazon Route 53 brings to the DNS table Amazon Route 53 is a highly available and scalable Domain Name System (DNS) service offered by AWS. It is named by the TCP or UDP port 53, which is where DNS server requests are addressed. Like any DNS service, Route 53 handles domain regist...

Read more
  • Amazon
  • AWS
  • Cloud Migration
  • DNS
  • Route 53