Web 2.0 News Desk
Is Modern Grid Computing Applicable to Business?
Scaling applications made easy
May. 16, 2008 03:00 PM
Digg This!
Page 2 of 2
« previous page
How Is Modern Grid Computing Applicable to Business?
As physical enterprise IT systems increasingly adopt more
decentralized infrastructure, distributed computing will unlock new business
models, improve business competitiveness, and facilitate rapid change and
structural migration. With the majority of businesses under continuous pressure
to utilize all resources as efficiently as possible in order to maintain
financial competition in crowded markets, distributed systems offer an
attractive solution – centralized control coupled with dispersed physical
assets and overheads.
This route is particularly attractive to software developers
and providers looking to harness the power of the grid model to provide
software as an on-demand service that delivers the same benefits to customers
and end users without the on-site IT headache. The relentlessly increasing cost
and complexity of maintaining IT departments and infrastructure makes on-tap
computing power very attractive to modern enterprises.
There are many compelling reasons why the grid represents
the future of business computing, and it’s worth taking a closer look at five
key areas: scalability, speed, deployment, pervasiveness and cost.
Scalability
Scalability is the very essence of distributed or grid
computing. Businesses are understandably attracted to the idea of gaining the
same benefits of commercially licensed applications supported by in-house IT
departments, without having to deal with any of the associated costs and
back-end mess. Consumers used to seamless transparency in online applications
have no trouble using this model in their Web 2.0 world and now it’s the turn
of businesses to follow suit by adopting Platform as a Service, or PaaS.
PaaS is the idea that a platform can be run as a service,
which is very similar to a SaaS environment. The key difference between SaaS
and PaaS, however, is that the platform itself is a service that hosts SaaS
applications. Think of it as meta-SaaS. PaaS solutions are built from the
ground up to harness the benefits of grid computing coupled with the simplicity
of web access. SaaS applications that run on top of PaaS will benefit from easy
deployment due to the inherent similarity in architecture. A good portion of
businesses wear themselves out deploying, maintaining, and upgrading software
systems using skills way outside their businesses’ core function of interest
simply because the resulting application is indispensable. It’s these services
that fit perfectly into the PaaS model. Over a grid, the power of existing
systems can be scaled exponentially, maintained seamlessly, upgraded
transparently, and redeployed almost instantly.
Speed
Distributed computing models exploit the computing resources
of multiple networks to solve problems quickly and increase the speed and
availability of applications to market. Grids have no particular relevance to
any one business sector and can help increase productivity and competitiveness
by facilitating the deployment of new systems or accelerating existing system
performance. Grid-powered PaaS and the SaaS applications that run on top of it
can be tapped on-demand anywhere across an organization via simple web
technology by unskilled, untrained users who don’t need to know anything about
where or how their application is being provided, only that it is continuously
available, secure, and fast. For them, technology is becoming a utility.
Deployment
Software as a Service benefits greatly from an elastic PaaS
Web 2.0 delivery and management solution, which means software developers can
save weeks of integration and configuration time by creating a suitable hosting
environment for their SaaS application. Dynamic networked computing power and
storage resources provide flexibility on a computing-on-demand model, eliminating
the need for businesses to invest heavily in static hardware while at the same
time benefiting from that partner’s grid experience.
Pervasiveness
Grid computing was created to throw almost limitless
computational power and storage at problems that were too big for individual
computers or small clusters to take on. Increasingly, one of the repercussions
of this approach is that businesses and consumers alike are provided with so
many relevant applications on-tap around them all the time that technology
becomes almost ubiquitous and computing becomes pervasive. Grid computing will
facilitate the growth of pervasive computing that will support daily tasks of
large organizations and communities without any conscious awareness on behalf
of individuals. This means that not only can software be delivered as a service,
applications can be provided unobtrusively in the working environment, enabling
constant connectivity and intuitive computation without any consideration for
delivery, maintenance, or the more tiresome side of software deployment.
Cost
Perhaps the single biggest cited advantage of grid computing
is its cost savings benefit. Indeed, analysts at Gartner have previously
defined the value of the grid as "creating a more powerful, larger, single
virtual system, or producing a less expensive alternative of the same size as
the system it is replacing.” Grids pull together application, database and
storage resources so that they may be shared across projects to provide more
predictability at less cost with fewer software licenses. PaaS combined with
SaaS finishes the picture by providing a payment model that allows businesses
to be billed intelligently and only for the resources that they actually use,
providing major savings on top of those made by avoiding initial investment in
expensive hardware.
Implications of the Grid on Future Applications
Platform as a Service, distributed computing infrastructure,
and virtualization are combining to create increasingly powerful and rapid
networks.
The responsibility now lies with providers of on-demand
application delivery platforms, managed services, and end-user applications to
deliver robust services that are able to support developing distributed computing
infrastructure.
If applications play nicely with grids, organizations will
benefit from true business continuity through distributed, shared computing
environments.
Page 2 of 2
« previous page
About David AbramowskiDavid Abramowski is CEO of Morph Labs. Prior to joining Morph Labs, David was Director of Product Marketing for Symantec, where he was responsible for introducing and enabling acquired endpoint technologies to Symantec's worldwide sales and partner organizations. In addition, David has extensive international experience enabling sales and technical teams, most recently with an assignment in Singapore, where he was involved with enterprise sales organizations across Japan, India, China, South Asia and Australia.