Browse: Home / Hadoop Summit: Appistry Targets HDFS Market
By Krishnan Subramanian on June 30, 2010
Appistry, the platform player based out of St. Louis, is positioning itself to target companies considering HDFS (See our previous coverage of Appistry here). Yesterday, Yahoo hosted Hadoop Summit 2010. Appistry made an announcement about some strategic alliances they have built with some of the important players in the Hadoop ecosystem. It is pretty clear that they want to offer an alternative to enterprises not convinced about HDFS.
In May this year, Appistry announced CloudIQ Storage offering (Ben covered this announcement here at Cloudave) which is a distributed storage platform with no single point of failure. This platform was highly suitable for petabyte scale and data intensive applications. Unlike many of the other storage solutions where the storage and compute are separated, CloudIQ nodes contain both a storage component and a compute component. It brings the processing relevant to the stored data into the node, thereby overcoming any bottlenecks associated with the transfer of data from the compute component to the storage component. CloudIQ is highly suitable for high throughput computations.
HDFS is fast gaining traction in many fields with more and more enterprises taking it seriously for their use. However, HDFS has a problem of single point of failure at the HDFS NameNode. If it is used with MapReduce, this is not a problem because the uptime requirement of mapreduce kind of batch processing systems is not highly crucial like few other scenarios. Now a days more and more distributed applications are using HDFS as a general purpose storage because they can store large volumes of data. In such use cases, high availability is a requirement and the single point of failure is not a welcome situation.
Appistry is trying to target this drawback with HDFS and trying to convince HDFS users with their CloudIQ storage platform (which doesn’t have this single point of failure problem). CloudIQ offers plug and play compatibility to HDFS which means that critical applications using HDFS can be moved to CloudIQ platform without any significant disruption to the workflow. Today’s announcement is a shot in the arm for Appistry because three significant players in the Hadoop ecosystems, Concurrent, Datameer and Kitenga, have validated their products against Appistry CloudIQ Storage Hadoop Edition. This means that, without any changes to their applications, users of the Concurrent, Datameer and Kitenga products can choose Appistry CloudIQ Storage as a more robust file system for their enterprise-grade Hadoop deployments.
This validation is pretty significant for Appistry and their CloudIQ Storage product. They are currently at the initial stages of talking to potential enterprise customers and this validation will help them gain credibility among the enterprises. Plus, it also offers these companies in the Hadoop ecosystem an opportunity to diversify their offerings.
Director, OpenShift Strategy at Red Hat. Founder of Rishidot Research, a research community focused on services world. His focus is on Platform Services, Infrastructure and the role of Open Source in the services era. Krish has been writing @ CloudAve from its inception and had also been part of GigaOm Pro Analyst Group. The opinions expressed here are his own and are neither representative of his employer, Red Hat, nor CloudAve, nor its sponsors.