How Does Hadoop Work?

Hadoop is an open-source software framework for storing data and running applications on clusters of commodity hardware. It provides massive storage for any kind of data, enormous processing power and the ability to handle virtually limitless concurrent tasks or jobs.

Hadoop History

As the World Wide Web grew in the late 1900s and early 2000s, search engines and indexes were created to help locate relevant information amid the text-based content. In the early years, search results were returned by humans. But as the web grew from dozens to millions of pages, automation was needed. Web crawlers were created, many as university-led research projects, and search engine start-ups took off (Yahoo, AltaVista, etc.).

hadoop-timeline-infographic

One such project was an open-source web search engine called Nutch – the brainchild of Doug Cutting and Mike Cafarella. They wanted to return web search results faster by distributing data and calculations across different computers so multiple tasks could be accomplished simultaneously. During this time, another search engine project called Google was in progress. It was based on the same concept – storing and processing data in a distributed, automated way so that relevant web search results could be returned faster.

In 2006, Cutting joined Yahoo and took with him the Nutch project as well as ideas based on Google’s early work with automating distributed data storage and processing. The Nutch project was divided – the web crawler portion remained as Nutch and the distributed computing and processing portion became Hadoop (named after Cutting’s son’s toy elephant). In 2008, Yahoo released Hadoop as an open-source project. Today, Hadoop’s framework and ecosystem of technologies are managed and maintained by the non-profit Apache Software Foundation (ASF), a global community of software developers and contributors.

 

Why is Hadoop important?

  • Ability to store and process huge amounts of any kind of data, quickly. With data volumes and varieties constantly increasing, especially from social media and the Internet of Things (IoT), that's a key consideration.
  • Computing power. Hadoop's distributed computing model processes big data fast. The more computing nodes you use, the more processing power you have.
  • Fault tolerance. Data and application processing are protected against hardware failure. If a node goes down, jobs are automatically redirected to other nodes to make sure the distributed computing does not fail. Multiple copies of all data are stored automatically.
  • Flexibility. Unlike traditional relational databases, you don’t have to preprocess data before storing it. You can store as much data as you want and decide how to use it later. That includes unstructured data like text, images and videos.
  • Low cost. The open-source framework is free and uses commodity hardware to store large quantities of data.
  • Scalability. You can easily grow your system to handle more data simply by adding nodes. Little administration is required.

What are the challenges of using Hadoop?

MapReduce programming is not a good match for all problems. It’s good for simple information requests and problems that can be divided into independent units, but it's not efficient for iterative and interactive analytic tasks. MapReduce is file-intensive. Because the nodes don’t intercommunicate except through sorts and shuffles, iterative algorithms require multiple map-shuffle/sort-reduce phases to complete. This creates multiple files between MapReduce phases and is inefficient for advanced analytic computing.

There’s a widely acknowledged talent gap. It can be difficult to find entry-level programmers who have sufficient Java skills to be productive with MapReduce. That's one reason distribution providers are racing to put relational (SQL) technology on top of Hadoop. It is much easier to find programmers with SQL skills than MapReduce skills. And, Hadoop administration seems part art and part science, requiring low-level knowledge of operating systems, hardware and Hadoop kernel settings.

Data security. Another challenge centers around the fragmented data security issues, though new tools and technologies are surfacing. The Kerberos authentication protocol is a great step toward making Hadoop environments secure.

Full-fledged data management and governance. Hadoop does not have easy-to-use, full-feature tools for data management, data cleansing, governance and metadata. Especially lacking are tools for data quality and standardization.



Fun Fact: "Hadoop” was the name of a yellow toy elephant owned by the son of one of its inventors.

Hadoop in Today's World

The promise of low-cost, high-availability storage and processing power has drawn many organizations to Hadoop. Yet for many, a central question remains: How can Hadoop help us with big data and analytics? Learn more here!

tdwi logo - Advancing all things data


A big read: Hadoop for the enterprise

This comprehensive 40-page Best Practices Report from TDWI explains how Hadoop and its implementations are evolving to enable enterprise deployments that go beyond niche applications.

Download the TDWI report

Report icon


Ugly duckling turned swan

That’s how the Bloor Group introduces the Hadoop ecosystem in this report that explores the evolution of and deployment options for Hadoop. It includes a detailed history and tips on how to choose a distribution for your needs.

Download report

Webcast icon

Self-service data prep for analytics

Want to learn how to get faster time to insights by giving business users direct access to data? This webinar shows how self-service tools like SAS Data Preparation make it easy for non-technical users to independently access and prepare data for analytics.

Watch now

E-book icon


SAS and Hadoop overview

Get acquainted with Hadoop and SAS concepts so you can understand and use the technology that best suits your needs. Check out this page learn how SAS technology interacts with Hadoop.

Get overview

Data management for Hadoop

Big data skills are in high demand. Now business users can profile, transform and cleanse data – on Hadoop or anywhere else it may reside – using an intuitive user interface.

Learn more about Hadoop data management from SAS

How Is Hadoop Being Used?

Going beyond its original goal of searching millions (or billions) of web pages and returning relevant results, many organizations are looking to Hadoop as their next big data platform. Popular uses today include:

Low-cost storage and data archive

The modest cost of commodity hardware makes Hadoop useful for storing and combining data such as transactional, social media, sensor, machine, scientific, click streams, etc. The low-cost storage lets you keep information that is not deemed currently critical but that you might want to analyze later.

Sandbox for discovery and analysis

Because Hadoop was designed to deal with volumes of data in a variety of shapes and forms, it can run analytical algorithms. Big data analytics on Hadoop can help your organization operate more efficiently, uncover new opportunities and derive next-level competitive advantage. The sandbox approach provides an opportunity to innovate with minimal investment.

Data lake

Data lakes support storing data in its original or exact format. The goal is to offer a raw or unrefined view of data to data scientists and analysts for discovery and analytics. It helps them ask new or difficult questions without constraints. Data lakes are not a replacement for data warehouses. In fact, how to secure and govern data lakes is a huge topic for IT. They may rely on data federation techniques to create a logical data structures.

Complement your data warehouse

We're now seeing Hadoop beginning to sit beside data warehouse environments, as well as certain data sets being offloaded from the data warehouse into Hadoop or new types of data going directly to Hadoop. The end goal for every organization is to have a right platform for storing and processing data of different schema, formats, etc. to support different use cases that can be integrated at different levels.

IoT and Hadoop

Things in the IoT need to know what to communicate and when to act. At the core of the IoT is a streaming, always on torrent of data. Hadoop is often used as the data store for millions or billions of transactions. Massive storage and processing capabilities also allow you to use Hadoop as a sandbox for discovery and definition of patterns to be monitored for prescriptive instruction. You can then continuously improve these instructions, because Hadoop is constantly being updated with new data that doesn’t match previously defined patterns.

 

 

Building a recommendation engine in Hadoop

One of the most popular analytical uses by some of Hadoop's largest adopters is for web-based recommendation systems. Facebook – people you may know. LinkedIn – jobs you may be interested in. Netflix, eBay, Hulu – items you may want. These systems analyze huge amounts of data in real time to quickly predict preferences before customers leave the web page.

How: A recommender system can generate a user profile explicitly (by querying the user) and implicitly (by observing the user’s behavior) – then compares this profile to reference characteristics (observations from an entire community of users) to provide relevant recommendations. SAS provides a number of techniques and algorithms for creating a recommendation system, ranging from basic distance measures to matrix factorization and collaborative filtering – all of which can be done within Hadoop.

Read how to create recommendation systems in Hadoop and more


I believe that Hadoop has matured to a point that people can successfully build large and complex applications atop the platform. Hadoop has met our scalability requirements for handling large and varied types of data.
Bob Zurek, Senior Vice President of Products at Epsilon in Hadoop for the Enterprise, a TDWI Best Practices Report

Hadoop, analytics and the data scientist


Like coding? You’ll love this. An all-in-one, interactive environment gives you everything you need to get accurate insights from data in Hadoop. At lightning-fast speed.

Learn more about analytics on Hadoop from SAS

How It Works and a Hadoop Glossary

Currently, four core modules are included in the basic framework from the Apache Foundation:

Hadoop Common – the libraries and utilities used by other Hadoop modules.

Hadoop Distributed File System (HDFS) – the Java-based scalable system that stores data across multiple machines without prior organization.

YARN – (Yet Another Resource Negotiator) provides resource management for the processes running on Hadoop.

MapReduce – a parallel processing software framework. It is comprised of two steps. Map step is a master node that takes inputs and partitions them into smaller subproblems and then distributes them to worker nodes. After the map step has taken place, the master node takes the answers to all of the subproblems and combines them to produce output.

Other software components that can run on top of or alongside Hadoop and have achieved top-level Apache project status include:

AmbariA web interface for managing, configuring and testing Hadoop services and components.
CassandraA distributed database system.
FlumeSoftware that collects, aggregates and moves large amounts of streaming data into HDFS.
HBaseA nonrelational, distributed database that runs on top of Hadoop. HBase tables can serve as input and output for MapReduce jobs.
HCatalogA table and storage management layer that helps users share and access data.
HiveA data warehousing and SQL-like query language that presents data in the form of tables. Hive programming is similar to database programming.
OozieA Hadoop job scheduler.
PigA platform for manipulating data stored in HDFS that includes a compiler for MapReduce programs and a high-level language called Pig Latin. It provides a way to perform data extractions, transformations and loading, and basic analysis without having to write MapReduce programs.
SolrA scalable search tool that includes indexing, reliability, central configuration, failover and recovery.
SparkAn open-source cluster computing framework with in-memory analytics.
SqoopA connection and transfer mechanism that moves data between Hadoop and relational databases.
Zookeeper   
An application that coordinates distributed processing.

Commercial Hadoop distributions

Open-source software is created and maintained by a network of developers from around the world. It's free to download, use and contribute to, though more and more commercial versions of Hadoop are becoming available (these are often called "distros.") With distributions from software vendors, you pay for their version of the Hadoop framework and receive additional capabilities related to security, governance, SQL and management/administration consoles, as well as training, documentation and other services. Popular distros include Cloudera, Hortonworks, MapR, IBM BigInsights and PivotalHD.

 


Getting data into Hadoop

Here are just a few ways to get your data into Hadoop.

  • Use third-party vendor connectors (see Data Connectors).
  • Use Sqoop to import structured data from a relational database to HDFS, Hive and HBase. It can also extract data from Hadoop and export it to relational databases and data warehouses.
  • Use Flume to continuously load data from logs into Hadoop.
  • Load files to the system using simple Java commands.
  • Create a cron job to scan a directory for new files and “put” them in HDFS as they show up. This is useful for things like downloading email at regular intervals.
  • Mount HDFS as a file system and copy or write files there.

Big Data, Hadoop and SAS

SAS support for big data implementations, including Hadoop, centers on a singular goal – helping you know more, faster, so you can make better decisions. Regardless of how you use the technology, every project should go through an iterative and continuous improvement cycle. And that includes data preparation and management, data visualization and exploration, analytical model development, model deployment and monitoring. So you can derive insights and quickly turn your big Hadoop data into bigger opportunities.

Because SAS is focused on analytics, not storage, we offer a flexible approach to choosing hardware and database vendors. We can help you deploy the right mix of technologies, including Hadoop and other data warehouse technologies.

And remember, the success of any project is determined by the value it brings. So metrics built around revenue generation, margins, risk reduction and process improvements will help pilot projects gain wider acceptance and garner more interest from other departments. We've found that many organizations are looking at how they can implement a project or two in Hadoop, with plans to add more in the future.

More on SAS and Hadoop