Menu
Apache lights a fire under Hadoop with Spark

Apache lights a fire under Hadoop with Spark

Apache Spark might provide a faster alternative to Hadoop MapReduce

The Apache Software Foundation has announced the first production-ready release of Spark, analysis software that could speed jobs that run on the Hadoop data-processing platform.

Dubbed the "Hadoop Swiss Army knife," Apache Spark provides the ability to create data-analysis jobs that can run 100 times faster than those running on the standard Apache Hadoop MapReduce.

MapReduce has been widely criticized as a bottleneck in Hadoop clusters because it executes jobs in batch mode, which means that real-time analysis of data is not possible.

Spark provides an alternative to MapReduce in that it executes jobs in short bursts of micro-batches that are five seconds or less apart. It also provides more stability than real-time, stream-oriented Hadoop frameworks such as Twitter Storm.

The software can be used for a variety of jobs, such as an ongoing analysis of live data, and, thanks to a software library, more computationally in-depth jobs involving machine learning and graph processing.

Using Spark, developers can write data-analysis jobs in Java, Scala or Python, using a set of more than 80 high-level operators.

With the version 1.0 release, Apache Spark now offers a stable API (application programming interface), which developers can use to interact with Spark though their own applications.

Also new for version 1.0 is a Spark SQL component for accessing structured data, allowing the data to be interrogated alongside unstructured data in analysis work.

Apache Spark is fully compatible with Hadoop's Distributed File System (HDFS), as well as with other Hadoop components such as YARN (Yet Another Resource Negotiator) and the HBase distributed database.

The University of California, Berkeley's AMP (Algorithms, Machines and People) Lab originally developed Spark, and Apache adopted it as an incubator project in June 2013. IT companies such as Cloudera, Pivotal, IBM, Intel and MapR have all folded Spark into their Hadoop stacks. Databricks, a company founded by some of the developers of Spark, offers commercial support for the software.

Both Yahoo and NASA, among others, use the software for daily data operations.

As with all Apache software, Apache Spark has been issued under the Apache License version 2.0.

Follow Us

Join the New Zealand Reseller News newsletter!

Error: Please check your email address.

Tags applicationsDatabricksdata miningsoftware

Slideshows

Top 50 defining moments of the New Zealand channel in 2016

Top 50 defining moments of the New Zealand channel in 2016

Reseller News looks back on a tumultuous 12 months for the New Zealand channel, assessing the fallout from a year of sizeable industry change. Whether it be local or global mergers and acquisitions, distribution deals or job changes, the channel that started the year differs somewhat to the one set to finish it - Reseller News assesses the key moments that made 2016.​

Top 50 defining moments of the New Zealand channel in 2016
​Hewlett Packard Enterprise honours high achieving NZ channel

​Hewlett Packard Enterprise honours high achieving NZ channel

Hewlett Packard Enterprise honoured its top performing Kiwi partners at the second running of its HPE Partner Awards in New Zealand, held at a glitzy ceremony in Auckland. Recognising excellence across eight categories - from distributors to resellers - the tech giant celebrated its first year as a standalone company, following its official split from HP in 2015.

​Hewlett Packard Enterprise honours high achieving NZ channel
Nutanix treats channel partners to Christmas cruise

Nutanix treats channel partners to Christmas cruise

Nutanix recently took to the seas for a Christmas Cruise around Sydney Harbour with its Australia and New Zealand staff, customers and partners to celebrate a stellar year for the vendor. With the sun out, they were all smiles and mingled over drinks and food.

Nutanix treats channel partners to Christmas cruise
Show Comments