Hortonworks plans to revamp Hadoop and its big data tools with cloud best practi ...
Arun Murthy, co-founder and chief product officer, Hortonworks. (LinkedIn Photo)

One big disadvantage that comes with a hybrid cloud strategy is forcing your developers to learn and understand the different techniques required by cloud providers and on-premises software vendors for lots of applications. Hortonworks , the company behind several tools for big-data darling Hadoop, plans to revamp its software over the next few years in order to make modern cloud-native development practices part of its on-premises tools, giving hybrid cloud developers one less thing to worry about.

Hortonworks plans to announce the Open Hybrid Architecture Initiative Monday, kicking off the project that will allow customers running Hadoop and Hortonworks tools on their own servers to take advantage of newer infrastructure ideas that have become popular since the big-data analysis software was created, said Arun Murthy, co-founder and chief technical officer of Hortonworks. It’s yet another sign that while self-managed servers aren’t disappearing as fast as people once thought they might , the infrastructure concepts of the cloud-native era are going to eventually become de facto standards.

A very popular tool within enterprise data teams, Hadoop was built at Yahoo more than a decade ago as a big data analytics tool, and it was originally designed for a world of on-premises servers in which storage and compute are tightly coupled, Murthy said. However, one of the primary benefits of cloud computing is that storage and compute are decoupled from each other, which means you can upgrade your storage needs without having to change anything on the compute side, or vice versa.


Hortonworks plans to revamp Hadoop and its big data tools with cloud best practi ...
Hadoop was built for a world in which storage and compute were tightly coupled (left), but modern cloud computing separates the two. (Hortonworks Image)

The Open Hybrid Architecture Initiative will build on existing code written by the Hadoop community to create a way for on-premises Hadoop users to enjoy the benefits from decoupling storage and compute, Murthy said. It also will add support for containerized application development shops through partnerships with Red Hat and IBM around Kubernetes, the open-source container-management project enjoying its time in the sun just as Hadoop did earlier this decade.

“It becomes hard for the enterprise to learn multiple systems, one, two, three, four systems, beyond the ones they already know,” Murthy said, describing how the rise of hybrid cloud and multicloud strategies can give companies some operational flexibility but often makes life harder for the developers. The idea here is that developers will be able to manage Hadoop and Hortonworks tools across both on-premises and cloud servers using the techniques they need to learn to make everything work on the cloud, rather that flitting back and forth between two (or more) different application models.

Hortonworks is kicking off this project on Monday, but expects it to take well over a year before all the steps are completed.

本文数据库(综合)相关术语:系统安全软件

tags: cloud,Hadoop,Hortonworks,storage,compute,premises,Murthy,tools
分页:12
转载请注明
本文标题:Hortonworks plans to revamp Hadoop and its big data tools with cloud best practi ...
本站链接:https://www.codesec.net/view/597115.html


1.凡CodeSecTeam转载的文章,均出自其它媒体或其他官网介绍,目的在于传递更多的信息,并不代表本站赞同其观点和其真实性负责;
2.转载的文章仅代表原创作者观点,与本站无关。其原创性以及文中陈述文字和内容未经本站证实,本站对该文以及其中全部或者部分内容、文字的真实性、完整性、及时性,不作出任何保证或承若;
3.如本站转载稿涉及版权等问题,请作者及时联系本站,我们会及时处理。
登录后可拥有收藏文章、关注作者等权限...
技术大类 技术大类 | 数据库(综合) | 评论(0) | 阅读(28)