Embracing Cloud in the Cognitive Era

We all know speed is of the essence
and time means money.

The introduction of cloud computing
provides new tools and techniques to reduce the time a developer needs to go
from idea to application. What used to take weeks or months, can now be done in
days or even hours!

Shortened development time is made
possible because the “as-a-service” model has simplified developers’ workload
and facilitated a transformation in DevOps using the cloud.
Infrastructure-as-a-Service (IaaS) reduces the effort and time required to provision
and manage hardware while Platform-as-a-Service (PaaS) reduces the effort needed
to manage the various middleware from databases to APIs to services. Lastly,
Software-as-a-Service (SaaS) gives end users access to business applications
hosted and maintained by the vendor.

Canadian bank, Tangerine is one such
enterprise client. A challenge they used to face was recreating problems to
provide solutions in response to their customers’ complaints and reviews. With
Bluemix however, customers could easily submit their feedback by simply shaking
their phones which in turn, allowed the Bank to get to the root of the problem
very quickly. Implementation time was found to reduce by 90% to 95%.

To summarize, both IaaS and PaaS cloud
models simplify the workload of developers by managing complex tasks that are
orthogonal to the developers’ goals, thereby freeing up more time for them to
create innovative new applications.

2.    
The second principle is to connect
what you do in the native cloud, seamlessly across your enterprise IT without
compromising governance or security.  

We talk about public clouds and
private clouds, but the reality is that most clients will go to a world that is
hybrid. After all, one size does not fit all. This is the classic bimodal IT
conundrum.  

CIOs spend billions of dollars to
build up their data centre for a reason. It provides a rich source of data that
should and can be leveraged instead of simply letting it stay locked away. For
example, the vibrant API economy today lets you open your business to new
markets and extend your business assets to ecosystem developers.  

Unfortunately, many of our clients today
spend large amounts of time dealing with issues related to deployment and
operations. Consider using deployment scripts that are repeatable, which helps
to reduce errors and variability in this part of the lifecycle, improve
productivity and improve speed time-to-market as simplified. Reusable
deployment processes with built-in actions help to enable faster testing and
deployment of changes.

Last year, IBM partnered with YES
Bank, India’s fifth largest bank, to help them rapidly define the APIs required
for digital banking. Through an interactive portal, the Bank was able to expand
its revenue generating channels by adding multiple partners in a matter of
days, rather than months. Today, YES BANK is even offering its payment
capabilities as APIs to its customers and partnering with e-commerce vendors
and currency exchange houses to offer an integrated financial supply chain
solution.

Leveraging your existing IT
investments will certainly help to lower the risks of any cloud deployment.

3.     Finally,
the third principle to any successful cloud deployment is optimize each
workload to the best fit.

This is especially important when CIOs
are now investing in analytics, mobile, social and related applications. All
these new workloads present different patterns, run at different speeds, but
need to work together.

So it is important to know what runs
where, how close the data is to the application, how fast the analytical
queries can be done, and how secure the integration is done on different
platforms.  It matters what workloads run
where and how.

In short, you should be able to optimize
each workload to the best fit.