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.
- We will assume that you have created a new Cloud Platform Console project and that you have enabled billing for this project.
- We will need both Google Cloud Dataproc and Compute Engine APIs enabled.
- And finally, we will assume that you have installed the Google Cloud SDK and initialized gcloud.
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 10.140.0.2 220.127.116.11 RUNNING spark-zeppelin-w-0 asia-east1-a n1-standard-4 10.140.0.4 18.104.22.168 RUNNING spark-zeppelin-w-1 asia-east1-a n1-standard-4 10.140.0.3 22.214.171.124 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 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)
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
Having done that, we can now access http://localhost:8080/.
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:
- How to set up a Cloud Dataproc cluster with Zeppelin.
- We have launched
- We have also set up a SSH tunnel to access the Zeppelin Notebook from the master node.
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!
How to Unlock Complimentary Access to Cloud Academy
Are you looking to get trained or certified on AWS, Azure, Google Cloud Platform, DevOps, Cloud Security, Python, Java, or another technical skill? Then you'll want to mark your calendars for August 23, 2019. Starting Friday at 12:00 a.m. PDT (3:00 a.m. EDT), Cloud Academy is offering c...
What Exactly Is a Cloud Architect and How Do You Become One?
One of the buzzwords surrounding the cloud that I'm sure you've heard is "Cloud Architect." In this article, I will outline my understanding of what a cloud architect does and I'll analyze the skills and certifications necessary to become one. I will also list some of the types of jobs ...
Boto: Using Python to Automate AWS Services
Boto allows you to write scripts to automate things like starting AWS EC2 instances Boto is a Python package that provides programmatic connectivity to Amazon Web Services (AWS). AWS offers a range of services for dynamically scaling servers including the core compute service, Elastic...
Content Roadmap: AZ-500, ITIL 4, MS-100, Google Cloud Associate Engineer, and More
Last month, Cloud Academy joined forces with QA, the UK’s largest B2B skills provider, and it put us in an excellent position to solve a massive skills gap problem. As a result of this collaboration, you will see our training library grow with additions from QA’s massive catalog of 500+...
DevSecOps: How to Secure DevOps Environments
Security has been a friction point when discussing DevOps. This stems from the assumption that DevOps teams move too fast to handle security concerns. This makes sense if Information Security (InfoSec) is separate from the DevOps value stream, or if development velocity exceeds the band...
Test Your Cloud Knowledge on AWS, Azure, or Google Cloud Platform
Cloud skills are in demand | In today's digital era, employers are constantly seeking skilled professionals with working knowledge of AWS, Azure, and Google Cloud Platform. According to the 2019 Trends in Cloud Transformation report by 451 Research: Business and IT transformations re...
Disadvantages of Cloud Computing
If you want to deliver digital services of any kind, you’ll need to estimate all types of resources, not the least of which are CPU, memory, storage, and network connectivity. Which resources you choose for your delivery — cloud-based or local — is up to you. But you’ll definitely want...
Google Cloud vs AWS: A Comparison (or can they be compared?)
The "Google Cloud vs AWS" argument used to be a common discussion among our members, but is this still really a thing? You may already know that there are three major players in the public cloud platforms arena: Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP)...
Deployment Orchestration with AWS Elastic Beanstalk
If you're responsible for the development and deployment of web applications within your AWS environment for your organization, then it's likely you've heard of AWS Elastic Beanstalk. If you are new to this service, or simply need to know a bit more about the service and the benefits th...
How to Use & Install the AWS CLI
What is the AWS CLI? | The AWS Command Line Interface (CLI) is for managing your AWS services from a terminal session on your own client, allowing you to control and configure multiple AWS services and implement a level of automation. If you’ve been using AWS for some time and feel...
Cloud Academy’s Blog Digest: July 2019
July has been a very exciting month for us at Cloud Academy. On July 10, we officially joined forces with QA, the UK’s largest B2B skills provider (read the announcement). Over the coming weeks, you will see additions from QA’s massive catalog of 500+ certification courses and 1500+ ins...
AWS Fundamentals: Understanding Compute, Storage, Database, Networking & Security
If you are just starting out on your journey toward mastering AWS cloud computing, then your first stop should be to understand the AWS fundamentals. This will enable you to get a solid foundation to then expand your knowledge across the entire AWS service catalog. It can be both d...