The Application-Driven Computing and Storage Platform Movement

Posted 7/19/2016 4:02:43 PM by STUART PARKERSON, Publisher Emeritus

The Application-Driven Computing and Storage Platform Movement
We recently had a visit with Sushil Kumar, CMO of Robin Systems, about the application-driven compute and storage platform movement within the application-defined data center era.

ADM: What challenges are data centers, private clouds, and enterprise application users and administrators faced with today?

Kumar: In modern data centers & private clouds, enterprise application users and administrators are typically faced with unique challenges around data velocity, volume and variance.

These challenges become augmented with distributed clustered applications; especially those operating in multi-tenant self-service mode, rely on well managed and optimized infrastructure. Typically, the challenges will be attributed to the infrastructure and less on the applications running atop.

ADM: What sort of solutions do containers bring to bear on these challenges?

Kumar: Container-based, compute and storage platforms fit snugly in, by making the boundaries among servers, virtual machines, and storage invisible to the applications.  This maximizes hardware utilization, eliminates virtual machine and data sprawl, and dramatically simplifies the application lifecycle. 

ADM: That sounds complicated? Is it?

Kumar: Being able to deploy an application into the best-suited infrastructure, while guaranteeing QoS and predicted performance thru entire application life cycle - can not be easily achieved (if at all) by using multitude of scripts and tools stitched together. However, an innovative end-2-end platform with its holistic coherent coverage of the entire stack - from app-2-spindle enables effective CAPX and efficient OPEX without compromise.

ADM: How does a platform like the one you mention pay off?

Kumar: This platform has enabled leading retailers to accelerate data ingest 10x faster which is key to multi-data source and retail analysis. In another case, the ease and simplicity of defining and deploying complex distributed applications such as Cassandra and Elasticsearch went from days to minutes.

Being able to match the right analytics job to the best suited infrastructure while guaranteeing QoS and predicted performance cannot be easily achieved (if at all) by using a multitude of scripts and tools stitched together. Robin’s holistic end-2-end platform coverage from app-2-spindle enables effective CAPX and efficient OPEX without compromise.

ADM: How is this technology solution designed?

Kumar: The design roots are in block storage enablement of the entire application stack scaling, snapshot and cloning. This process occurs through advanced BI orchestration matching best available resources (CPU, memory, network, IOps) to compute resources.

It brings together container technology with super dynamic application and data flow reflected by distributed digital world and hybrid clouds. This technology enables both optimization of stateless based analytics tools "born in the cloud", as well as modernization and optimization of “legacy” applications migrating from any virtualized environment to containers.

ADM: What is the impact for end-users or the industry at large? What has been their ROI?

Kumar: This type of platform provides the application users (business analysts, data scientists, app developers) with what they want - a simple and agile way to define, deploy and manage analytics applications and entire stack including supporting infrastructure.

The platform is centered around rationalization of applications atop containerized infrastructure. The complete application lifecycle, from provisioning to continuous Integration (CI) and continuous delivery (CD) are the foundation which demonstrate  new operating & implementation models previously not possible.

With this solution, defining and creating a multi-node distributed analytics clustered application should not take more than a few minutes, snapshotting/cloning should be accessible in a couple mouse clicks, scaling compute and/or storage should be done automagically.  

Using this technology, users can expect up to 50% savings on overall storage, up to 10x faster deployment times, simplified and guaranteed QoS in the form of IOps so critical for analytics apps.

Read More


About the author: STUART PARKERSON, Publisher Emeritus

Stuart Parkerson has an extensive background in niche technology publishing.

Subscribe to App Developer Daily

Latest headlines delivered to you daily.