Research

Publication Date: 23 November 2010 ID Number: G00209000

Private Cloud Computing: An Essential Overview

Thomas J. Bittman

Private cloud computing requires strong leadership and a strategic plan to navigate the hype, a dynamic vendor landscape, and difficult organizational and cultural changes. We discuss the forms of private cloud computing, the benefits and challenges, the architecture of the private cloud, the vendor landscape, and how to get started. Key Findings
Private cloud computing can come in many different forms, and is not necessarily onpremises, insourced or based on virtual machines. Technology is often the easiest part of private cloud computing culture, politics, process and funding are all much harder. Many enterprises (especially midmarket) will build partial, but "good enough," private cloud solutions based on their requirements. The private cloud market will be very dynamic and notable for acquisitions during the next few years.

Recommendations
Implement private cloud computing when public cloud services do not meet your requirements for service levels, security, compliance, etc. Start with strong leadership, including CIO commitments an organization's ability to evolve culturally and politically is critical. Develop a portfolio of your services (inventory, SLAs, costs), and build strategies and road maps for each. Redesign applications using platform as a service (PaaS) before planning to leverage infrastructure as a service (IaaS) for every application. Enable hybrid capabilities when developing a private cloud architecture, to avoid lock-in and enable choice in the future.

2010 Gartner, Inc. and/or its affiliates. All rights reserved. Gartner is a registered trademark of Gartner, Inc. or its affiliates. This publication may not be reproduced or distributed in any form without Gartner's prior written permission. The information contained in this publication has been obtained from sources believed to be reliable. Gartner disclaims all warranties as to the accuracy, completeness or adequacy of such information and shall have no liability for errors, omissions or inadequacies in such information. This publication consists of the opinions of Gartner's research organization and should not be construed as statements of fact. The opinions expressed herein are subject to change without notice. Although Gartner research may include a discussion of related legal issues, Gartner does not provide legal advice or services and its research should not be construed or used as such. Gartner is a public company, and its shareholders may include firms and funds that have financial interests in entities covered in Gartner research. Gartner's Board of Directors may include senior managers of these firms or funds. Gartner research is produced independently by its research organization without input or influence from these firms, funds or their managers. For further information on the independence and integrity of Gartner research, see "Guiding Principles on Independence and Objectivity" on its website, http://www.gartner.com/technology/about/ombudsman/omb_guide2.jsp

TABLE OF CONTENTS
Strategic Planning Assumption(s) ................................................................................................. 3 Analysis ....................................................................................................................................... 3 Forms of Private Cloud Computing................................................................................... 3 Benefits, Challenges and the Path Forward...................................................................... 4 The Architecture of Private Cloud Computing ................................................................... 4 Private Cloud Vendors ..................................................................................................... 5 How to Start ..................................................................................................................... 6 Bottom Line ..................................................................................................................... 7 Recommended Reading ............................................................................................................... 7

LIST OF FIGURES
Figure 1. Private-Cloud-Computing-Enabling Vendors .................................................................. 5

Publication Date: 23 November 2010/ID Number: G00209000 2010 Gartner, Inc. and/or its Affiliates. All Rights Reserved.

Page 2 of 8

STRATEGIC PLANNING ASSUMPTION(S)

Through 2014, IT organizations will spend more money on private-cloud-computing investments than on offerings from public cloud providers. By 2015, the majority of private-cloud-computing services will evolve to leverage public cloud services in a hybrid model. Through 2014, fewer than 20% of virtualized deployments will be complete private cloud deployments. By 2015, the majority of virtualized deployments will evolve to support some private-cloudcomputing capabilities, but fewer than 20% will be complete private cloud deployments.

ANALYSIS
Private cloud computing is gaining interest, especially among larger enterprises (see "Private Cloud Computing Plans From Conference Polls"). Vendors are aggressively targeting the privatecloud-computing opportunity by repositioning existing products (e.g., IBM's CloudBurst), acquiring smaller vendors (e.g., CA Technologies' acquisition of 3Tera and Cassatt), delivering new functionality (e.g., VMware's vCloud Director), and offering on-premises versions of public cloud offerings (e.g., Microsoft's Windows Azure Platform Appliance). Due to concerns with public cloud offerings, the momentum and growing maturity of virtualization deployments in enterprises, the growing number of products that can help enable a private cloud architecture, and enterprise interest in the use of the cloud-computing style, through 2014, IT organizations will spend more money on private-cloud-computing investments than on offerings from public cloud providers (see "Predicts 2010: Cloud Computing Emerges From the Hype, Scope and Issues Demand Clarification" and "Private Cloud Computing Plans From Conference Polls").

Forms of Private Cloud Computing

Cloud computing is a style of computing in which scalable and elastic IT-enabled capabilities are delivered as a service to customers using Internet technologies. Cloud-computing services can be delivered by an internal IT organization (insourced) or by an external service provider (outsourced). The underlying infrastructure can be hosted within an organization's data center or in an external data center. That underlying infrastructure can be dedicated to a single customer ("private cloud"), shared between a consortium of customers ("community cloud") or shared with a service provider's customer base in general ("public cloud"). Private cloud computing can come in many forms: A private cloud will usually be insourced and run on-premises using equipment owned by the enterprise, but not always. Private cloud computing can be outsourced and externally hosted (this is often called "virtual private cloud," but that term is used to describe a broad spectrum from dedicated equipment to virtual private networks). It can also be outsourced, but internally hosted (managed by a third party). Private cloud services will usually be IaaS, but not always. For example, some PaaS offerings will be available for private cloud services (e.g., Microsoft's Windows Azure Platform Appliance), and likewise with software-as-a-service (SaaS) offerings.

Publication Date: 23 November 2010/ID Number: G00209000 2010 Gartner, Inc. and/or its Affiliates. All Rights Reserved.

Page 3 of 8

A private cloud that is IaaS will usually leverage virtual machines, but not always. Private cloud solutions that leverage rapid provisioning (e.g., IBM's CloudBurst) will be available to support physical and virtual devices.

Benefits, Challenges and the Path Forward

The primary benefit of private cloud computing is speed. Implementing a service catalog that offers standard services through a self-service interface, and automating the delivery of those offerings, can increase the speed of delivery dramatically. By themselves, standards, automation, and some form of resource pooling or virtualization will also reduce costs but these can be done without building a complete private cloud. Private cloud computing introduces a new way to use IT, and changes the relationship between the customer and the IT organization to a business relationship, based on service delivery and usage metrics. For most organizations, this is a difficult change. Process changes, management changes, funding changes, service standardization, culture and politics consistently come up in polls as more-difficult challenges than technology itself (see "Private Cloud Computing Plans From Conference Polls"). However, navigating these changes for private cloud computing also smoothes the path for the enterprise to consider public-cloud-computing alternatives in the future if and when they begin to meet enterprise requirements for security, service levels, compliance, etc. Private cloud computing is a steppingstone to public cloud computing, and it should be designed to enable future sourcing choices. Hybrid cloud computing refers to a blend of cloud-computing delivery models, most commonly a blend of a public cloud model with a private cloud model. Once an organization has made the leap in terms of leveraging a self-service interface with a service catalog, and automated service delivery behind that, adding an alternative sourcing option (an external service provider) for some services or some instances of services becomes easier. Terms like "overdrafting" and "cloudbursting" are used to describe the potential here, but few end-to-end services will span both public and private clouds. However, the ability to locate new development and test requests with a third party, or to move low-priority workloads during quarterly peaks, is a common request heard from Gartner clients. While hybrid cloud computing is still a challenge, based on today's technologies, it is evolving quickly (e.g., VMware's vCloud Director is designed specifically to enable the hybrid cloud). The majority of private-cloud-computing services will evolve to leverage public cloud services in a hybrid model by 2015. For some organizations, the final goal will become a full migration to public cloud services; however, most organizations that build a private cloud architecture will maintain either a private or hybrid architecture in the long term. Most private-cloud-computing architectures should be designed with the future use of hybrid cloud computing in mind, providing more choices and a potential migration path to public cloud services as they mature, in an evolutionary way.

The Architecture of Private Cloud Computing

A private cloud service has four specific architectural elements: access management, service management, resource management and pooled resources (see "The Architecture of a Private Cloud Service"). A self-service portal may provide access management and an easy interface, but service automation behind that portal is also critical. A virtualized infrastructure (which could have rapid reprovisioning as a virtualization tool) may provide a fluid and flexible resource pool, but unless those service requirements can be entered in a self-service or programmatic way, and unless the allocation of the virtualized resources to meet service requirements is automated, it is not a private cloud architecture.

Publication Date: 23 November 2010/ID Number: G00209000 2010 Gartner, Inc. and/or its Affiliates. All Rights Reserved.

Page 4 of 8

Almost all large companies, and most small and midmarket enterprises, are virtualizing some portion of their infrastructures, and many of them consider private cloud computing an ultimate goal. However, a full private-cloud-computing architecture will not be worth the investment nor necessary for all organizations. Many smaller organizations will forgo usage metrics and chargeback. Some larger organizations will maintain a human interface between the customer and some of the services (to ensure appropriate service design, etc.), even if the deployment of the service is automated. While these may not be complete private clouds by definition, they are real and provide value, and enterprises should consider which elements of private cloud computing are necessary before deploying them. For many enterprise, implementing 70% of a private cloud architecture is absolutely good enough. By 2015, the majority of virtualized deployments will evolve to support some private-cloud-computing capabilities, but less than 20% will be "complete" private cloud deployments.

Private Cloud Vendors

Vendors are gravitating to private cloud computing from many areas of expertise (see Figure 1). Private cloud computing is also a hotbed for startups, and more-established vendors are often moving into private cloud computing through acquisitions. A few examples of recent important acquisitions include: 3Tera and Cassatt (by CA Technologies), Surgient (by Quest Software), VMLogix (by Citrix), Cast Iron Systems (by IBM), Neptuny (by BMC Software), and SpringSource (by VMware). Through 2015, the private cloud market will be extremely dynamic, and notable for acquisitions of smaller vendors by larger vendors trying to rapidly fill out their portfolios. Figure 1. Private-Cloud-Computing-Enabling Vendors

Source: Gartner (November 2010)

Several vendors stand out:

Publication Date: 23 November 2010/ID Number: G00209000 2010 Gartner, Inc. and/or its Affiliates. All Rights Reserved.

Page 5 of 8

The Big Four traditional management vendors (BMC Software, CA Technologies, HP and IBM) have rich service and data center management tools to bring to the market. All four were latecomers to the server virtualization market, from which private cloud computing is emerging, but all four have a rich set of assets (and consulting services) to be major players. Where there are gaps, these four vendors have traditionally filled them with acquisitions, and will continue to do so. VMware and Citrix are starting from the opposite end of the market growing from a rich base of virtualization and virtualization management capability. While VMware has the larger installed base, it still has a long way to go up the stack in terms of service automation and manageability. Citrix has made its own acquisitions (such as VMLogix), and it has an interesting portfolio that includes NetScaler, but it is also incomplete. Cisco is focusing on expanding its infrastructure influence, and is filling its gaps through alliances with VMware, BMC Software, NetApp and EMC. Microsoft was late to virtualization, but is closing the capability gap quickly. It has also traditionally been slow to focus on management, but the priority of System Center has been rising and Microsoft has become more acquisitive to fill gaps in System Center. Eucalyptus Systems is an interesting small player. It is focused on a complete privatecloud-computing infrastructure with partners rPath and newScale, with a focus on interoperability with Amazon, VMware and others. Through acquisitions and the need for complete solutions, private-cloud-computing enablement will become centered on a relatively small number of vendors or vendor partnerships that supply complete solutions.

How to Start
The challenges with private cloud computing are primarily about culture, politics, services, processes, business relationship and funding technology is not the right place to start (see "Getting Started With Private Cloud: Services First"). Leadership. Because private cloud computing affects the relationship between IT and the business, and it affects the day-to-day jobs of people inside the IT organization, a private-cloudcomputing direction requires strong leadership and executive buy-in first, foremost and throughout. A periodically updated strategic plan with a road map and metrics is critical. Service. Service is the key to cloud computing (see "Getting Started With Private Cloud: Services First"). A self-service interface requires a thorough understanding of the services that IT offers. Which ones do we deliver today? Which ones can be fully standardized? Are SLAs in place today? What service levels do we achieve today? What are the customer requirements for this service? What options should be offered? What is our cost to deliver this service at these service levels today? Once all of that is understood, each service at least the critical ones requires its own road map. Which services make sense for the cloud-computing style, and which ones do not? When will public cloud services meet our needs, if ever? What should we do between now and then? Business Case. Private cloud services should be built with return on investment in mind not just cost, but value (in terms of speed for the IT organization's customers, or as an initial investment toward a cloud computing migration). In some cases, it might be better to wait for a public cloud service to evolve to fill your needs. In other cases, it will make good business sense to build a private cloud service.

Publication Date: 23 November 2010/ID Number: G00209000 2010 Gartner, Inc. and/or its Affiliates. All Rights Reserved.

Page 6 of 8

Benchmark. An enterprise that has built a private-cloud-computing service should constantly benchmark public offerings, to consider leveraging them in a hybrid model, or to perhaps migrate completely to a public cloud service in the future. Once service requirements and internal service delivery costs are fully understood, public cloud offerings can be an objectively compared alternative choice.

Bottom Line
In many ways, the core of private cloud computing is the intersection of a maturing virtualization trend and the rapidly growing cloud-computing trend. Some organizations see private cloud computing as a way to defend the organization against cloud computing. However, the private cloud is all about changing the status quo, in terms of service orientation, culture, process and funding. Done well, private cloud computing will help an IT organization deliver services much faster to its customers, and will open up an easier path to leverage public-cloud-computing solutions, if and when they mature. The market is young, and vendors are rapidly trying to fill out their portfolios to address the opportunities. In the near term, it will be important to create your own strategy, and be careful not to be led down a path that is determined by the technologies and business models that vendors offer today, because those will change rapidly. Instead, start on the inside, with strong leadership and a service focus.

RECOMMENDED READING
"The Architecture of a Private Cloud Service" "Getting Started With Private Cloud: Services First" "Private Cloud Computing Plans From Conference Polls" "From Secure Virtualization to Secure Private Clouds" "Q&A: The Many Aspects of Private Cloud Computing" "Cloud Infrastructure as a Service: An Essential Overview" This research is part of a set of related research pieces. See "Private Cloud Computing: Clearing the Air" for an overview.

Publication Date: 23 November 2010/ID Number: G00209000 2010 Gartner, Inc. and/or its Affiliates. All Rights Reserved.

Page 7 of 8

REGIONAL HEADQUARTERS
Corporate Headquarters 56 Top Gallant Road Stamford, CT 06902-7700 U.S.A. +1 203 964 0096 European Headquarters Tamesis The Glanty Egham Surrey, TW20 9AW UNITED KINGDOM +44 1784 431611 Asia/Pacific Headquarters Gartner Australasia Pty. Ltd. Level 9, 141 Walker Street North Sydney New South Wales 2060 AUSTRALIA +61 2 9459 4600 Japan Headquarters Gartner Japan Ltd. Aobadai Hills, 6F 7-7, Aobadai, 4-chome Meguro-ku, Tokyo 153-0042 JAPAN +81 3 3481 3670 Latin America Headquarters Gartner do Brazil Av. das Naes Unidas, 12551 9 andarWorld Trade Center 04578-903So Paulo SP BRAZIL +55 11 3443 1509

Publication Date: 23 November 2010/ID Number: G00209000 2010 Gartner, Inc. and/or its Affiliates. All Rights Reserved.

Page 8 of 8