Nothing Special   »   [go: up one dir, main page]

Skip to content

hrchlhck/k8s-bigdata

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

90 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

k8s-bigdata

Apache Spark with HDFS cluster within Kubernetes.

Overview

As the description says, this repository is an Apache Spark with an HDFS cluster within Kubernetes. Although it contains Intel HiBench benchmark suite for testing CPU, IO, and network usage, the cluster can run as a regular one.

Supported HiBench Workloads

  • Micro
  • Machine Learning
  • Websearch

Building k8s-bigdata

You can just execute the build.sh file.

$ ./build.sh

Submitting the cluster

To submit the cluster and prepare it, you must type the following ./scripts/init-cluster.sh <WORKLOAD> <BENCHMARK> <INPUT_SIZE> Where:

  1. WORKLOAD represents a workload from HiBench
  2. BENCHMARK represents the benchmark
  3. INPUT_SIZE means the size of the workload for the benchmark

Running HiBench

To run a HiBench benchmark, you can run ./scripts/run.sh <WORKLOAD> <BENCHMARK> The report saved will be in the base directory with the name hibench.report.

Features

  • k8s-bigdata currently uses Apache Spark 2.4 with Hadoop 2.7 binary
  • No need to register manually each datanode
  • Kubernetes will create a datanode for each node registered in the cluster
  • Can specify which node namenode, resourcemanager, and historyserver will be launched by assigning the label type=master. If you are new to Kubernetes, type kubectl label nodes YOUR NODE type=master

Future works

  • 🗴 Support data streaming frameworks such as Apache Kafka
  • ✓ Switch static to dynamic environment variables for containers (avoid building on every change in ./hadoop/base/hadoop.env file)
  • ✓ Implement a configuration parser to run HiBench without changing run.sh
  • ✓ Implement a solution to change the size of input data for HiBench benchmarks without accessing namenode pod directly

Architecture

Based on the locality of reference, HiBench, Hadoop Namenode, and Spark Master are within the same container as processes. Also, HiBench needs the Hadoop and Spark directories located at the namenode pod.

Alt text

References