From the Dev Team

Follow the latest developments from our technical team

Introduced in 2008, Apache Hive has been the de-facto SQL solution in Hadoop. By 2012, SQL had become a key battleground for Hadoop and many vendors started to publish benchmarks showing massive performance advantages their solutions had over Hive. Each of these vendors predicted that Hive would eventually be supplanted by the proprietary solution they were pushing.

The concerns about Hive’s performance were real. Hadoop in 2012 was a purely batch platform and no work had ever been done within Hive to address low-latency or interactive workloads.…

A significant reason for the increased adoption of the Hortonworks Data Platform by customers and partners has been Apache Hadoop YARN. This major advance, introduced last October in HDP2, allows Hadoop to move from many single-purpose clusters to a versatile, integrated data platform that hosts multiple business applications.

YARN has become the architectural center of Hadoop. We intend to make it easier for applications to work in a YARN environment, and benefit from the integrated capabilities and technologies that form the blueprint for enterprise Hadoop.…

Apache Ambari has always provided an operator the ability to provision an Apache Hadoop cluster using an intuitive Cluster Install Wizard web interface, guiding the user through a series of steps:

  • confirming the list of hosts
  • assigning master, slave, and client components to configuring services, and
  • installing, starting and testing the cluster.

With Ambari Blueprints, system administrators and dev-ops engineers can expedite the process of provisioning a cluster. Once defined, Blueprints can be re-used, which facilitates easy configuration and automation for each successive cluster creation.…

We recently hosted the fourth of our seven Discover HDP 2.1 webinars, entitled Apache 2.4.0, HDFS and YARN. It was very well attended and a very informative discourse. The speakers outlined the new features in YARN and HDFS in HDP 2.1 including:

  • HDFS Extended ACLs
  • HTTPs support for WebHDFS and for the Hadoop web UIs
  • HDFS Coordinated DataNode Caching
  • YARN Resource Manager High Availability
  • Application Monitoring through the YARN Timeline Server
  • Capacity Scheduler Preemption

Many thanks to our presenters, Rohit Bakhshi (Hortonworks’ senior product manager), Vinod Kumar Vavilapalli (co-author of the YARN Book, PMC, Hadoop YARN Project Lead at Apache and Hortonworks), and Justin Sears (Hortonworks’ Product Marketing Manager).…

Traditionally, HDFS, Hadoop’s storage subsystem, has focused on one kind of storage medium, namely spindle-based disks.  However, a Hadoop cluster can contain significant amounts of memory and with the continued drop in memory prices, customers are willing to add memory targeted at caching storage to speed up processing.

Recently HDFS generalized its architecture to include other kinds of storage media including SDDs and memory [1]. We also added support for caching hot files in memory [2].…

Julian Hyde will present the following talks at the Hadoop Summit:

  • Discardable In-Memory, Materialized Query for Hadoop,”  (June 3rd, 11:15-11:55 am)
  • “Cost-based Query Optimization in Hive,” (June 4th,  4:35 pm-5:15 pm)
  • What to do with all that memory in a Hadoop cluster? The question is frequently heard. Should we load all of our data into memory to process it? Unfortunately the answer isn’t quite that simple.

    The goal should be to put memory into its right place in the storage hierarchy, alongside disk and solid-state drives (SSD).…

    The Apache Ambari community is happy to announce last week’s release of Apache Ambari 1.6.0, which includes exciting new capabilities and resolves 288 JIRA issues.  

    Many thanks to all of the contributors in the Apache Ambari community for the collaboration to deliver 1.6.0, especially with Blueprints, a crucial feature that enables rapid instantiation and replication of clusters.

    Each release of Ambari makes substantial strides in providing functionality to simplify the lives of system administrators and dev-ops engineers to deploy, manage, and monitor large Hadoop clusters, including those running on Hortonworks Data Platform 2.1 (HDP).…

    On Wednesday May 21, Himanshu Bari (Hortonworks’ senior product manager), Venkatesh Seetharam (committer to Apache Falcon), and Justin Sears ( Hortonworks’ Product Marketing Manager), hosted the third of our seven Discover HDP 2.1 webinars. Himanshu and Venkatesh discussed data governance in Hadoop through Apache Falcon that is included in HDP 2.1. As most of you know, ingesting data into Hadoop is one thing; having data governed, by dictating and defining data-pipeline policies, is another thing—a necessity in the enterprise.…

    According to New York Observer, there were couple of major social reasons that spurred the genesis and growth of Meetup.com. First, it was Robert Putman’s book Bowling Alone, in which he talks about the collapse of communities in America. And the second was an event that not only changed the world but changed New York: it was the aftermath of September 11, where strangers cared about greeting, meeting, and talking.…

    On May 15, Owen O’Malley and Carter Shanklin hosted the second of our seven Discover HDP 2.1 webinars. Owen and Carter discussed the Stinger Initiative and the improvements to Apache Hive that are included in HDP 2.1:

    • Faster queries with Hive on Tez, vectorized query execution and a cost-based optimizer
    • New SQL semantics and datatypes
    • SQL-standard authorization
    • The Hive job visualizer in Apache Ambari
    • And many more

    Here is the complete recording of the webinar.…

    Last week Vinay Shukla and Kevin Minder hosted the first of our seven Discover HDP 2.1 webinars. Vinay and Kevin covered three important topics related to new Apache Hadoop security features in HDP 2.1:

    • REST API security with Apache Knox Gateway
    • HDFS security with Access Control Lists (ACLs)
    • SQL security and next-generation Hive authorization

    Here is the complete recording of the webinar.

    Here are the presentation slides: http://www.slideshare.net/hortonworks/discoverhdp21security

    Attend our next Discover HDP 2.1 webinar tomorrow, Thursday, May 15 at 10am Pacific Time: Interactive SQL Query in Hadoop with Apache Hive

    We’re grateful to the many participants who joined and asked excellent questions.…

    I’m a pretty heavy Unix user and I tend to prefer doing things the Unix Way™, which is to say, composing many small command line oriented utilities. With composability comes power and with specialization comes simplicity. Although, sometimes if two utilities are used all the time, sometimes it makes sense for either:

    • A utility that specializes in a very common use-case
    • One utility to provide basic functionality from another utility

    For example, one thing that I find myself doing a lot of is searching a directory recursively for files that contain an expression:

    find /path/to/root -exec grep -l "search phrase" {} \;

    Despite the fact that you can do this, specialized utilities, such as ack have come up to simplify this style of querying.…

    The first use of the term BoF session was used at the Digital Equipment Users’ Society (DECUS) conference in the 1960s. Its essence was to bring together like minds and thought leaders—just as birds of the feather flock together— to share and exchange computing ideas, in an informal yet spirited way. Since then, the organizers and sponsors of most computing conferences have been loyal to its essence and spirit.

    For ideas and innovation happen in collaboration—not in isolation. …

    This is the second in our series on the motivations and architecture for improvements to the Apache Hadoop YARN’s Resource Manager Restart resiliency. Other in the series are:

    Introduction: Phase I – Preserve Application-queues

    In the introductory blog, we previewed what RM Restart Phase I entails. In essence, we preserve the application-queue state into a persistent store and reread it upon RM restart, eliminating the need for users to resubmit their applications.…

    This is the first post in our series on the motivations and architecture for improvements to the Apache Hadoop YARN’s Resource Manager Restart resiliency. Other in the series are:

    Resource Manager (RM) is the central authority of Apache Hadoop YARN for resource management and scheduling. It is responsible for allocation of resources to applications like Hadoop MapReduce jobs, Apache TEZ DAGs, and other applications running atop YARN.…

    Go to page:« First...23456...10...Last »