What do Virtualization and Cloud executives think about 2012? Find out in this VMblog.com series exclusive.
Don't Let Next Year PaaS You By – Enterprise Cloud Trends for 2012
Article by Derick Townsend, VP of Product Marketing, ServiceMesh
In 2011, we saw several enterprises launch their initial
cloud deployments. Next year, these will evolve from mostly departmental efforts
to new cloud initiatives that deliver broader enterprise business value. Here
are five enterprise cloud trends that will help drive this in 2012:
1. Enterprises evaluate
and adopt Private PaaS.
Enterprise IT has shown great interest in the benefits
offered by Platform-as-a-Service (PaaS)-faster application development, standardized
platforms at lower cost, scalability, embedded security, etc. But today's public
PaaS offerings also come with serious downsides, including vendor lock-in. If
you develop applications in Saleforce.com, Google AppEngine or a variety of
others, you can't easily migrate them to another platform.
As a result, the momentum behind "private PaaS" offerings is
building as we approach to 2012. With private PaaS, enterprises can assemble their
own PaaS offerings with a cloud management platform and customize them as
desired to specific requirements or their preferred combination of middleware, testing
tools, and utilities. Private PaaS offerings are often initially deployed in a
private cloud, and can take a variety of forms, from classic application
development platforms to more specialized ones such as Hadoop, SAP, or custom
versions of Cloud Foundry. Done right, a private PaaS can deliver the best of
both worlds: agility and scalability benefits similar to a public PaaS, but
leveraging existing tools and training used in the enterprise today.
figure out that cloud governance must extend all the way out to business units.
In 2012, enterprises will continue extending self-service
access to more IT resources and expand the cloud's reach out to business units.
In the process, they'll discover that current cloud "policies" are too limited and
need to go beyond defining role-based access or simple thresholds for scaling
up resources. Instead, comprehensive cloud governance policies must also deal
with the regulatory, compliance, security, and performance issues that drive
the business units.
As a result, IT will adopt policy-driven governance models
and solutions to consistently enforce this across the organization. This includes
both management tools and organizational changes, such as federating governance
and policy responsibilities across groups within the enterprise. For example,
the corporate compliance officer needs to codify and enforce regulatory
constraints using one set of policies. IT needs to enforce another set of
policies related to managing resources and monitoring deployment environments.
Finally, business units need to enforce application-level policies that define
required performance levels and SLAs. All of these policies and their
interactions need to be intelligently enforced across the enterprise's
Next year, more IT organizations will recognize the full
scope and importance of cloud governance. Nearer term, IT will see the benefits
of enabling more end-to-end automation. Longer term, the realization will sink
in that governance is a key driver to ensure that stakeholders derive the cost
and agility benefits they expect out of the cloud-without risk of undercutting their
3. Enterprise cloud
ecosystems will grow fast...and experience growing pains.
As cloud projects expand from departmental efforts to
broader enterprise deployments, teams will have to integrate
a much more complex ecosystem of infrastructure, legacy systems, third-party applications,
custom tools/utilities, etc. Although many of today's cloud vendors claim to
have an end-to-end solution, the reality is that many of these solutions
require significant customization and integration to work out of the box. This
may have been acceptable for limited, proof-of-concept scenarios, but these solutions
will hit a wall when it comes to integrating into large enterprise deployments.
For example, providing security typically requires
integrating solutions for federated identity management, encryption key stores,
disk encryption, firewall products, virtual networks, DHCP, anti-virus, HIDS...the
list goes on. Now expand this list to include an ecosystem with multiple cloud
providers, accounting/chargeback, performance monitoring, and more.
In 2012, enterprises will grow frustrated by vendors that
require seemingly endless integration and customization projects that start to
undercut the expected benefits for their cloud project. Successful cloud
vendors will be the ones with full-featured APIs and support for a broad range of
legacy applications and tools out of the box, allowing enterprises to leverage
their existing investment and dramatically decreasing the time it takes to roll
out a new solution.
4. Back to the
drawing board for organizations that adopted a narrow, tactical view of cloud.
Enterprises that have looked to the cloud simply to provide infrastructure
automation will find their results disappointing. Many times, these enterprises
have made limited attempts to change their IT Ops organization and processes
even though cloud operating models demand fundamental shifts in mindsets,
skills and tools. This approach may incrementally improve some process steps,
but it doesn't address the bigger picture agility benefits of extending a broad
portfolio of self-service IT resources to end users within a fully governed
The reality is that many existing operational processes and
legacy tools don't meet the needs of the self-service, on-demand world of the
cloud. A key problem is that traditional ops processes are vertically siloed
around domains, rather than spanning more horizontally across the software lifecycle.
In 2012, these IT organizations will go back to the drawing board and rethink their
cloud strategy, including new technologies and organizational changes. A
variety of options await them, including the possibility of creating new
horizontal overlay groups or new Operations Teams that incorporate DevOps and automated
migration to the cloud gets serious... and complex.
To date, most cloud projects have encompassed new greenfield
applications or existing apps that are easy to migrate like websites, blog
engines, wikis, and simple departmental apps. The advantage is that these
workloads are not mission-critical and can survive some hiccups. Now that these
efforts have proven successful, companies will pursue the heavy lifting to move
more business-critical apps to the cloud.
In 2012, companies will begin developing Reference
Application Architectures that set the precedent for how to migrate and
optimize more complex applications to the cloud. These application architectures
will define how to ensure high availability, security, disaster recovery and
other enterprise-level application requirements. Much of this can be achieved
in the cloud more efficiently and cost-effectively; however, each organization
will need to go through its own learning pathways to make it happen. The use of
published reference guides and consultants will help, but ultimately each
organization will need to invest with their own significant, hands-on effort.
Here's to a promising and exciting 2012 for enterprise cloud
adoption. I'm already looking forward to future success stories.
ServiceMesh's Agility Platform enterprise cloud platform enables Global 2000 customers to implement
cloud-based “everything-as-a-service” IT delivery models that span internal and
external IaaS, PaaS, and SaaS providers to provide quantum improvements in
business agility and operating costs.