Las Vegas, Starnberg, 06. März 2017 – Enterprise-Class Software Defined Storage (SDS) für Hortonworks auf Basis IBM Power8 Server mit paralleler File System Architektur…
Zum Hintergrund: IBM und Hortonworks haben am 14. Februar im Rahmen der IBM-Partnerkonferenz in LasVegas die geplante Verfügbarkeit von Hortonworks Data Platform (HDP) für IBM Elastic Storage Server (ESS) und IBM Spectrum Scale angekündigt. Auf Basis dieser Vereinbarung soll HDP® auf Power mit IBM Spectrum Scale bis Ende April und HDP® auf x86 mit IBM Spectrum Scale bis Ende Mai zertifiziert werden. Durch das neue Agreement können IBM-Kunden ihre Investitionen in IBM Storage nutzen, um Hadoop-basierte Big Data-Anwendungen zu installieren. Zudem wird laut Vereinbarung HDP®-Kunden für ihre Hadoop- und Spark-Workloads eine Enterprise-Class Speicher-Alternative mit IBM Spectrum Scale angeboten. Dieses Agreement ist laut Anbieter das erste IBM Storage-Angebot mit Software-Defined-Storage, die für Hortonworks zertifiziert ist.
Hier ein gekürzter Auszug zur Ankündigung (Quelle: IBM Orginalmeldung): "The agreement with Hortonworks will lead to certification of Hortonworks HDP on Power with IBM Spectrum Scale and Hortonworks HDP on x86 with IBM Spectrum Scale. With the recent announcement availability of Hortonworks HDP on POWER8 platform, IBM clients already have increased choice when selecting the platform for their Hadoop distribution. This new agreement will enable IBM clients to also leverage their existing and future investments in IBM storage in deploying Hadoop based big data applications. Additionally, this agreement will provide existing Hortonworks HDP customers an enterprise-class storage alternative with IBM Spectrum Scale for their Hadoop and Spark workloads.
- The IBM Elastic Storage Server is a modern implementation of software-defined storage, combining IBM Spectrum Scale software with IBM POWER8 processor-based servers and storage solutions. Clearly differentiating IBM Spectrum Scale from other products in the market is its parallel file system architecture, which is ideal for the massive scaling of performance and capacity needed in today’s cognitive and big data analytic workloads. By eliminating the need to copy data from enterprise storage to a separate analytics platform IBM clients can more quickly respond to data based queries.
Key benefits of HDP for IBM Elastic Storage Server (ESS) and IBM Spectrum Scale include:
- Better storage efficiency – unlike HDFS which creates three copies of every piece of data, IBM ESS uses erasure coding, eliminating the need for multiple copies and increasing storage efficiency.
- Hybrid Storage – extends on premise storage to the cloud and therefore realizing a myriad of economic, security, accessibility and other benefits.
- High Performance – as an optimized, highly parallel storage solution ESS is capable of high-speed data throughput able to exceed the performance of distributed HDFS.
IBM is one of the founding partners of Open Data Platform and has helped pave the way to enable any company to join the Open Data Platform initiative (ODPi). The focus is on innovating around the Apache Hadoop open source core, growing the ecosystem and enabling solutions on a standardized Open Data Platform across the ecosystem partners.
IBM Big Insights, IBM's own Hadoop and Spark distribution, is also an option for clients who prefer to run IBM's own analytics software on IBM Spectrum Scale Storage.“
Weitere Informationen
> Hinweis auf IBM Expertentag vom 07. - 09. März 2017 in Ehningen > WhatsUp IBM Spectrum Scale Strategy Days 2017
> Storage Consortium Video: Übersicht zu IBM Spectrum Scale (YouTube)
Abb. 1: Building a Multi-Tenant Big Data Environment for Big Data and Analytics (Bildquelle: IBM)