Enterprise hardware evolve at a scorching pace

GemStone Systems, that claims pioneering work in several enterprise infrastructure technologies, has been steering through this space through enterprise-wide deployments in industries such as financial services, the Federal Government, transportation, telecommunications and energy.

Its proffered areas of forte include distributed resource management, in-memory caching and disk persistence, scalable data distribution, high performance computing operations, object management, etc.

Enterprise infrastructure is a terrain that despite its deep legacy, continues to mushroom with dynamic changes. The next spotlight to watch out for could be elastic data provisioning, event driven architectures and data driven grid computing.

Sudhir Menon, director of engineering - GemStone Systems, was in India recently, where he touched upon some key action happening in the enterprise infrastructure technology and where is it headed next. He also shared the new experiments happening with distributed computing by slicing the problem in a new way and trying collaborative portals, driven by web services and cloud computing services. Read on.

What are the future trends in the enterprise infrastructure industry both from a business and technology perspective?

The Enterprise Infrastructure industry stands on the threshold of massive irreversible change. From a business perspective, there is the constant push to lower TCO for the enterprise. Coupled with this is the ever-growing need to make business decisions with the most relevant and up-to-date information. This is pushing the industry towards lower latency, higher throughput infrastructure software that offers unparalleled scale out and zero downtimes. From a technology perspective, the availability of high performance commodity hardware along with zero copy latency stacks provides an ecosystem that makes it possible to achieve the desired business needs. Driving the convergence of these two trends is the availability of distributed data management software like Gemfire Enterprise, which makes it possible to utilize the hardware much more efficiently while lowering latency, increasing throughput and offering scale out and high availability.

How do you reckon the evolution of enterprise infrastructure so far?

Enterprise hardware continues to evolve at a scorching pace. Enterprise infrastructure too has come a long ways, especially the data management and data sharing aspects of it, starting with relational databases, traversing through a combination of databases and messaging systems, culminating in the data fabric. The emergence of elastic data provisioning, event-driven architectures and data driven grid computing are defining the next generation enterprise applications, and we believe that this technology as implemented by Gemfire is now ready for the mainstream after having proven itself in the extreme financial services vertical.

What are the exciting things to watch out from Gemstone per se in the near future?

In addition to the intense focus on latency and performance that we bring to every Gemfire release, we are tackling more and more items that provide ease of use, ease of integration and ease of monitoring and management to the enterprise architect in the distributed data management segment. Watch out for pluggable solutions with GFE that will make it easier to leverage the power of Gemfire much more easily in a much broader spectrum of applications.

How are your data fabrics progressing in view of demands like user-ease and transparency, integration tools with other infrastructures, virtualization and vertical specifics? Where is distributed computing headed next?

For years, the enterprise infrastructure industry has focused on building a distributed computing platform that can be used to solve large-scale problems. And for years, they have been stymied by the lack of products that can effectively manage data in a scaled-up environment. Gemfire seeks to turn that problem upside down by offering the ability to route execution of behavior where the data resides (rather than trying to fetch the data to a node as part of executing the behavior). This represents a whole new way of slicing the problem, one that is more scalable and we believe that it will change the way distributed computing is done in enterprises large and small. Collaborative portals, driven by web services and cloud computing services, are also dictating the evolution of distributed computing in the real time and near real time market.



Anything latest on Enterprise Data Fabric (EDF) particularly in context to SOA, Grid/cluster computing and OLTP applications?

The upcoming release of Gemfire Enterprise (version 5.7) offers several new features that were built as part of our “Grid Readiness” initiative. As part of this effort, we have fine-tuned our scaling, made several ease of use additions to the product, and improved our handling of network partitioning outages (aka “Split Brain”). We have added new features to make our messaging subsystem and its support for durable messaging much more efficient and resilient. We continue to investigate ways to allow our customers and industry enterprise architects to use Gemfire as part of cloud computing initiatives as well.



Anything else you would like to share from Gemstone's perspective?

In addition to everything that we discussed here, a very significant thing that I believe we have undertaken is building up the eco system of partners here in India who work with us to provide scalable implementation services for Gemfire enabled applications. We have conducted several focused workshops allowing us to build Gemfire skill sets in diverse verticals such as financial services, insurance, core banking, animation, and telco, all of which will yield results for us in the months to come. Our business development operation in India has been at the forefront of spreading awareness about Gemfire Enterprise in the local market, even as our development team here builds out newer and exciting product features for our growing customer base.