I D C I V I E W T R A N S C R I P T

Automated, Standardized, and Private Cloud

Management
August 2010
By Mary Johnston Turner, Research Director, System Management Software
Sponsored by Novell

Overview
IDC forecasts that 69% of enterprise datacenter workloads will be virtualized by 2013. Already 75% of
organizations that use virtualization have made virtual servers their default environment for new
application and workload deployments.

Virtualization has enabled IT organizations to hold the line on capital spending by allowing them to
share physical server resources across multiple workloads and applications. As the utilization of
physical servers increases, IT organizations save money on hardware, power and cooling, and
facilities costs. IDC estimates the average virtualized server currently supports roughly six virtualized
workloads, and that number will only go higher over time.

Virtualization and cloud computing initiatives are creating new types of management challenges
across enterprise datacenters. Unfortunately, what we consistently see is that the rate and pace of
operational change and complexity that results from virtualization makes it very difficult for IT teams
to significantly improve operational effectiveness and IT staff efficiency. Equally disappointing,
extensive use of virtualization does not always translate into better business outcomes, as spinning
up a virtual machine (VM) is only part of what it takes to deliver a usable business service.

As we discuss in this iView, cloud computing represents the next step in delivering enterprise IT and
business services more efficiently and dynamically than has previously been possible. Unlike
virtualization, which is focused on the optimization of physical resources and individual workloads,
cloud computing targets the delivery of end-to-end services used by end users. These services are
generally made up of multiple workloads which frequently reside on different, heterogeneous
computing platforms.

Public cloud services, including infrastructure-as-a-service, compute-on-demand options, and

software-as-a-service solutions provide customers with computing and applications resources as
needed, enabled via shared infrastructure and delivered over Web interfaces. Customers pay based
on the services used.

Private clouds use the same concepts except the resources and services are dedicated to the needs
of a single organization. Private clouds share infrastructure resources across multiple business
services, using self-service portals and service catalogs, as well as policy-driven automation and
configuration management tools to provision workloads on-demand.

Private clouds enable pooling and dynamic assignment of datacenter resources as needed in order to
optimize business service performance and costs. Within private cloud environments, workloads
move transparently across resources as needed, and end users focus more on the cost of the

IDC 1003
resources they consume and the service levels they experience, rather than on where the associated
infrastructure resources are located and or how they are paid for.

This shift away from focusing on the operation of components, and toward the delivery of services
using shared infrastructure, fundamentally changes many aspects of how IT teams must design, plan,
deploy, and operate datacenter environments. IT teams need holistic insight into service levels, and
need to be able to provision and orchestrate services across heterogeneous platforms. The ability to
rapidly migrate services and to automate self-service provisioning of end-to-end services requires
organizations to standardize service offerings and aggressively automate as many operational
activities as possible.

For most established datacenter environments, the move to cloud, and the accompanying
standardization and automation of management processes and tools, will be an evolutionary effort
that takes place over several years. In speaking with IT decision makers who are already
implementing private-cloud architectures, we consistently hear that the effort to define and
standardize services, workflows, and SLAs is worth the effort, as those investments allow the
organization to take full advantage of state-of-the-art cloud automation and management
technologies.

This iView describes the types of operational changes IT teams need to make in order to benefit from
the promise of private-cloud architectures, and identifies a key set of capabilities and attributes to
look for when considering management tools to help implement, automate, and operate these
environments.

From Virtualization to Cloud

At its core, cloud computing is about creating shared-resource pools to dynamically and cost-
effectively provision and support business service requirements. Public cloud services have received
lots of attention, and by IDC’s own estimates, public cloud services represent one of the fastest
growing of all IT spending categories, with an estimated CAGR of 26% over the next five years.

However, many enterprises are also beginning to implement private cloud architectures. In fact, a
recent IDC survey showed that that 73% of enterprise datacenters are evaluating, planning, or have
already implemented private cloud strategies. Most of these initiatives focus on deploying workloads
on pools of shared resources in order to drive up utilization, improve service levels, and create more
flexible business environments while maintaining control over security and compliance requirements.

Eventually, many organizations expect to make use of hybrid cloud approaches where they opt to use
public services for some functions where the public nature and global reach of the business make
sense, but still rely on internal private cloud architectures for critical business activities. Concerns
about security, compliance, and maintaining competitive business advantage have significant impact
on whether organizations feel comfortable using public cloud services today.

In most organizations, private cloud architectures build on the dynamic workload migration and server
consolidation capabilities provided by virtualization. In today’s environments, individual VMs can often
be spun up quickly, but end users may wait weeks to get their services while different IT teams and
disconnected processes work on deploying the application, the security, the network connections,
and the storage resources needed to deliver a fully functioning service. With a private cloud, the goal
is to standardize and automate most configuration, deployment, and workload migration activity and
eventually get to the point where business users can self-provision, or at least IT can quickly order
and automate deployment of standard services.

As a result, implementing cloud requires IT operations to go to a whole new level of business life-
cycle integration and awareness. Rather than each silo focusing on the availability, performance, or
cost of its own specific domain, private cloud computing focuses the entire organization on the rapid,

2 ©2010 IDC
flexible delivery of business services. This new approach to datacenter planning, provisioning, and
operation requires new approaches to IT workflows and demands widespread use of highly
automated infrastructure management tools.

Operational Transformations
IDC's work with organizations that are starting to implement private cloud strategies shows that
operational transformation is crucial to the success of these projects. Specifically, these organizations
find that, in addition to virtualizing the infrastructure, in order to enable dynamic resource sharing,
many IT workflows, service definitions, workload and system configurations, security and compliance
policies, and SLA metrics need to be highly standardized in order to facilitate extensive use of
automation. Decision-making and governance processes need to become tightly focused on business
outcomes, and business departments and stakeholders need to be trained to focus on the cost and
quality of services rather than on ownership of infrastructure components.

In organizations that have deployed virtualization, but have not made substantive changes to
processes, workflows, configurations, and SLAs, IDC finds that the ratio of administrators to managed
machines improves only slightly – hovering in the range of about 29 physical servers per
administrator versus 33 virtual machines per administrator. By comparison, IT organizations that
make the effort to negotiate with business partners to define standard images and configurations, to
eradicate one-off implementations, and to expose the true cost of providing customized service levels
find they can achieve dramatic staff productivity improvements by automating the provisioning,
migration, and ongoing management of those workloads and resources.

Although there are relatively few private clouds operational today, we have recently seen examples of
organizations that are operating with ratios of 150 - 200 or more physical and virtual machines per
administrator. A few leading-edge organizations that have aggressively standardized, integrated, and
automated physical and virtual system and workload management activities are achieving ratios of
350 - 500 or more servers per administrator.

Organizations experiencing these kinds of productivity improvements typically work aggressively to

share infrastructure across multiple workloads, implement governance systems to promote
standardization of services, configurations, security and related policies, and invest in automated,
integrated tools to provision, migrate, and maintain the health of workloads across these shared
environments. These organizations are also working to make business groups more aware of their
resource consumption and help them to better understand the cost of changing SLAs or assigning
dedicated rather than shared resources.

Cloud Management Priorities

Most early, private cloud customers are making significant investments in new automated planning,
provisioning, and monitoring tools to enable effective cloud operations. When it comes to evaluating
solutions, decision-makers should look for the following capabilities:

The ability to manage, provision, and maintain standardized service models that include security,
compliance, and performance attributes. These models need to recognize the dependencies and
interconnections across the multiple workloads that compose a service. They are needed to drive
automated service provisioning activities

The ability to migrate running workloads across heterogeneous physical, virtual, and public cloud
environments in order to continually optimize resource consumption, reduce VM sprawl, and
maintain security and end-to-end service levels

Automated support for routine patch management and service provisioning activities, which
includes the ability to discover configurations and patch levels, evaluate them against the gold

©2010 IDC 3
 image, and enforce updates and compliance on an ongoing basis using software libraries and
templates

The ability to define and maintain templates for standardized, automated approval and change
control workflows to streamline approval, change control, audit and compliance reporting
processes

Self-service provisioning capabilities, as well as integration with service catalogs, service

fulfillment systems, and service desks

The ability to monitor and analyze hardware and workload characteristics, and to match them
appropriately in real time in order to maintain optimal service performance and costs

Usage and consumption metering and reporting to support capacity planning, governance, and
chargeback/showback programs

The most effective tools will support role-based interfaces and dashboards that can be used by both
IT professionals and end user organizations. By enabling all stakeholders to interact using the same
monitoring, provisioning, and workflow platforms, enterprises can reduce the time it takes to provision
and support services while increasing IT's credibility and visibility with business stakeholders.

Measuring the Benefits

Most organizations planning to deploy private clouds expect it will be a multi-year effort that begins
with a few selected services and applications or departments, and is then extended as the benefits of
the concept are proven and IT develops increased confidence in its ability to maintain SLAs while
increasing staff productivity and increasing resource utilization.

In order to build the business case for private cloud solutions, IT decision makers need to develop
solid data about the costs and benefits of these initiatives. The cost side of the equation includes
spending on new tools and platforms, as well as staff time devoted to developing and implementing
standards and service definitions.

The benefit side of the discussion needs to include both hard- and soft-dollar impacts, including the
following:

Hardware and power/cooling cost savings resulting from improved utilization due to eliminating
VM sprawl, sharing resources across multiple workloads, improving capacity planning, and
implementing dynamic workload balancing

Software license and maintenance savings due to elimination of assigned but unused licenses
and elimination of redundant management tools

Improved security and compliance due to more automated enforcement of policies and standards

Staff cost savings due to improved productivity and operational efficiencies

Improved end-user productivity, business performance, and customer satisfaction due to faster
provisioning, reductions in downtime, and consistent service levels

As datacenters become more virtualized and enterprises take greater advantage of cloud services, IT
organizations will need to invest in standardization and automation initiatives designed to deliver
high-quality, cost-effective services as needed to meet business requirements. Policy-based cloud
management solutions will be important to getting IT and business stakeholders aligned and moving
forward to support the transformation of their organization's infrastructure environments.

4 ©2010 IDC
IT cannot implement cloud strategies on its own – business stakeholders must buy in and participate
at every step. Effective measurement and reporting on costs, services levels, security, and
compliance will facilitate this type of interactive decision making and build support for cloud strategies
across business and IT leadership teams.

A B O U T T H I S P U B L I C A T I O N

This publication was produced by IDC Go-to-Market Services. The opinion, analysis, and research results presented herein
are drawn from more detailed research and analysis independently conducted and published by IDC, unless specific vendor
sponsorship is noted. IDC Go-to-Market Services makes IDC content available in a wide range of formats for distribution by
various companies. A license to distribute IDC content does not imply endorsement of or opinion about the licensee.

C O P Y R I G H T A N D R E S T R I C T I O N S

Any IDC information or reference to IDC that is to be used in advertising, press releases, or promotional materials requires
prior written approval from IDC. For permission requests contact the GMS information line at 508-988-7610 or [email protected].
Translation and/or localization of this document requires an additional license from IDC.

For more information on IDC visit www.idc.com. For more information on IDC GMS visit www.idc.com/gms.

Global Headquarters: 5 Speen Street Framingham, MA 01701 USA P.508.872.8200 F.508.935.4015 www.idc.com

©2010 IDC 5