Item 1. Business

Overview

We are a leading provider of high-density compute servers based on an open architecture approach, targeting scale out datacenter deployments. In addition, we also design, market and sell a range of storage appliances and high-capacity storage systems which leverage many of our core server technologies. Our products are designed to provide benefits in the areas of density, thermal management, remote management, ease of serviceability and flexible and efficient power distribution. We also offer a high degree of flexibility and control in component selection to match the specific environmental and application requirements of our customers. Our products are designed to reduce total cost of ownership through strategic sourcing of components and reduced deployment and operating expenses. We base our products on open standard components such as processors from Advanced Micro Devices, or “AMD” and Intel, and operating systems such as Linux and Windows in order to leverage the continuing price-performance improvements associated with high-volume computer components.

Our flagship Foundation Series compute servers are high-density, rack-mounted server systems designed specifically for scale out datacenter environments. The Foundation Series servers utilize our patented half-depth, back-to-back chassis design to increase the physical server density, reducing floor space requirements. When deployed in our cabinets, we are generally able to offer approximately twice the server or processor density of traditional rack-mount solutions. We provide a range of power and heat management techniques that enable our servers to operate effectively at these density levels. The Foundation Series servers also provide configurable components, front-facing cable connections for enhanced serviceability, and our proprietary Roamer™ remote management solution. In August 2004, we introduced our Scale Out Series of compute servers as an additional offering in our product line, which are designed to offer higher density levels for certain configurations as well as improve thermal and cable management. Our Scale Out Series is complementary to our Foundation Series, as we believe there will continue to be many customers for which the traditional rack mount size or system configurations available with the Foundation Series compute servers are more important than the additional

1

Table of Contents

features of the Scale Out Series. In addition, we selectively use reseller and original equipment manufacturer or OEM relationships to provide additional server offerings that our customers may request from time to time.

We also offer a number of storage solutions. In the third quarter of 2006, we began offering our RapidScale storage appliances that provide scalable shared storage solutions. RapidScale permits all stored data to reside under a single namespace and be simultaneously accessed by up to thousands of servers. The solution delivers high performance while improving capacity and performance scalability, ease of data management, and capacity utilization. RapidScale appliances leverage our own hardware in addition to our proprietary RapidScale file system appliance software. We also sell low-cost, high-capacity storage servers, which leverage many of our core compute server technologies, to help enterprises cost-effectively meet their increasing data storage requirements. If requested by our customers, we also provide storage products that combine our hardware with software layers provided by third parties under OEM software license arrangements. To complement our RapidScale and storage server product lines, we selectively leverage reseller and OEM relationships to provide additional storage offerings that our customers may request from time to time, such as external fiber channel, serial attached small computer system interface, or SAS, and redundant array of independent disks or RAID solutions.

We market our systems primarily through our direct sales force predominantly to enterprises within the United States. We are in the early stages of developing an international sales and operational presence, focusing initially on Europe and China. We focus our sales and marketing activities on enterprises that typically purchase hundreds of servers and tens to thousands of terabytes of storage per year. To date, we have sold our products to over 300 customers. We have concentrated our marketing efforts on leading Internet companies, as well as customers with high-performance computing requirements in vertical markets such as semiconductor design, enterprise software, federal government, entertainment, financial services, oil and gas exploration and biotechnology and pharmaceuticals.

Corporate Information

We were incorporated in the State of Delaware in December 2002 under the name of Rackable Corporation, at which time we received financing from an entity affiliated with Parthenon Capital and purchased substantially all of the assets and assumed substantially all of the liabilities of our predecessor corporation, now named GNJ, Inc. GNJ, Inc., formerly known as Rackable Systems, Inc. and referred to as Old Rackable in this report, was incorporated in October 2000. In December 2000, GNJ, Inc. purchased all of the assets and assumed all of the liabilities of Rackable Systems LLC, the predecessor entity. Rackable Systems LLC was later merged into GNJ, Inc. in January 2001. We changed our corporate name from Rackable Corporation to Rackable Systems, Inc. in December 2002. In September 2006, we acquired Terrascale Technologies, Inc., a Canadian based company, whose products are now sold under the RapidScale name. Our principal executive offices are located at 1933 Milmont Drive, Milpitas, California and our telephone number is (408) 240-8300.

Additional information about Rackable Systems, Inc. is available on our website at www.rackable.com . We make available free of charge, on or through our website, our Annual Report on Form 10-K, our Quarterly Reports on Form 10-Q, our Current Reports on Form 8-K, and any amendments to those reports filed or furnished pursuant to Section 13(a) or 15(d) of the Exchange Act, as soon as reasonably practicable after electronically filing such reports with, or furnishing such reports to, the Securities and Exchange Commission. Information contained on our website is not incorporated by reference in, or made part of this Annual Report on Form 10-K or our other filings with or reports furnished to the Securities and Exchange Commission.

The terms “Rackable,” “Rackable Systems,” “we,” “us” and “our” as used in this Annual Report on Form 10-K refer to Rackable Systems, Inc. and its subsidiary and, with respect to events occurring prior to December 23, 2002, our predecessor company formerly named “Rackable Systems, Inc.” and now named “GNJ, Inc.”

2

Table of Contents

Products

We design, market and support a comprehensive suite of compute and storage server systems that are optimized for scale out datacenter deployments. Our modular, rack-mounted systems are based on industry- standard components and open-standard or lower-cost operating systems. We offer a broad selection of core system component technologies, which enables us to tailor systems to specific customer requirements. Our innovative designs enable our customers to maximize server density while optimizing thermal and power management within the datacenter. We design our solutions to be installed on a turnkey basis with remote management capabilities that minimize ongoing maintenance expenses. By embracing the benefits of modular, open-standards computing and storage solutions, we enable our customers to expand capacity and improve performance in the datacenter at a lower total cost of ownership.

Key benefits of our systems include:

Innovative form factors to improve density and thermal management.  Our systems are designed to alleviate inefficient use of floor and rack space and thermal management issues, problems that are found in major datacenter installations today. Our Foundation Series compute servers are approximately half as deep as traditional rack-mount systems. By deploying these servers in our cabinets with our patented “back-to-back” mounting technology, we are able to offer twice the density of traditional rack-mount solutions. Our Scale Out Series offer high density levels by making use of a new server form factor and our “back-to-back” and “side-to-side” mounting technologies. We can achieve higher density levels for some configurations using the Scale Out Series line as opposed to the Foundation Series line.

With increased density comes increasing needs for optimized thermal management. We use a range of thermal management technologies to vent heat from the cabinet as well as reduce heat inside the server. These thermal management technologies include our patented back-to-back mounting with a central air space, or plenum, to draw heat to the center of the cabinet and enable exhaust from the top of the cabinet, as well as reduced server-level heat associated with our optional DC power solution. Our Scale Out Series has also been designed for thermal efficiency, including a taller chassis form factor to increase air flow and enable the use of larger heat dissipation sinks, and deflected air flow to create an updraft in the cabinet air plenum powerful enough to effectively evacuate heat.

Remote monitoring and improved serviceability.  Our Roamer lights-out remote management solution enables the administration of large-scale server farms from any network-connected off-site location. Roamer enables rapid identification and location of failed servers and results in a significant reduction in the need for onsite maintenance technicians. In addition, our products are designed for improved serviceability through a variety of means including front facing input-output, or I/O, ports and integrated cable troughs. Our Scale Out Series includes additional benefits such as completely internalized cable and power connections and “toolless” serviceability for simplified maintenance. These design innovations allow us to significantly reduce installation and maintenance time and to decrease the need for multiple technicians to service our systems.

Improved power management. Our server and storage lines include a range of design innovations to reduce power consumption and enable flexible power management schemes in scale out deployments. Because we build servers to order, we are sometimes able to eliminate components that draw power but may not be needed by the customer and to match the selection of a particular power supply to the load drawn by the system, which can improve power conversion efficiency. Our remote management technology offers the ability to stagger the start-up of systems, which reduces the aggregate power draw at system boot time and allows the customer to increase the number of systems attached to a power circuit without tripping the circuit breaker. All server components use DC power, and our DC power options allow cabinet- and datacenter-level bulk power conversion to enable higher overall power efficiency versus individual server-level AC power supplies, as well as improved redundancy schemes, such as the ability to withstand complete loss of an AC circuit. Our DC power solutions are designed to be compatible with datacenters that have either AC- or DC-based power distribution

3

Table of Contents

infrastructure. We also provide high-efficiency AC power supply options to provide the flexibility that customers demand.

Flexible platforms that address customer-specific requirements.  Our compute and storage server systems are based on highly flexible platforms that enable our customers to select the configuration, operating system, processor and other key components based on their requirements. Because our solutions are build-to-order at the component level, they meet our customers’ performance requirements without bundling unnecessary features or components that increase cost. Our systems utilize industry-standard components wherever possible and are compatible with lower-cost operating systems such as Linux and Windows, enabling our customers to achieve optimal performance in a cost-effective manner while minimizing vendor dependence associated with proprietary systems.

High capacity, cost-effective storage systems.  Similar to our compute servers, our storage servers utilize industry-standard components and lower-cost operating systems. We offer storage servers that allow our customers to run their own storage software or storage management applications. We also sell the RapidScale storage appliance that leverages our proprietary RapidScale clustered file system appliance software. If requested by our customers we also offer storage servers that include pre-installed storage software appliances that we source from third parties. Our storage platforms leverage many of the same design principles and component technologies as our compute server platforms. We believe we are well positioned to benefit from a convergence of server and storage technologies, and from the continuing improvements in storage system component technologies such as disk drives and RAID controller cards, as well as advances in storage protocol technologies such as iSCSI. We also source storage hardware technologies from third parties under both OEM and resale arrangements. We source our OmniStor line of external RAID arrays from a third party under an OEM arrangement. Over time, we expect to continue to expand our storage offerings through both internally developed and externally sourced products and technologies.

Customers

We have sold our products to more than 300 customers, including companies with large Internet businesses, and companies in vertical markets such as semiconductor design, enterprise software, federal government, entertainment, financial services, oil and gas and biotechnology and pharmaceuticals. Because we generally sell to customers with large-scale datacenter deployments who may procure servers hundreds or thousands of systems at a time, a single customer in any given quarter can account for a large portion of our sales. However, because our revenue has largely been generated in connection with these customers’ decisions to deploy large-scale server and storage farms, their capacity requirements can become fulfilled, whether temporarily or otherwise, and as a result they could purchase significantly fewer or no products from us in subsequent periods. For the fiscal year ended December 31, 2006, Microsoft and Yahoo! accounted for approximately 34% and 26% of our revenues, respectively. For the fiscal year ended December 31, 2005, Microsoft, Yahoo! and Amazon.com accounted for approximately 14%, 22% and 24% of our revenues, respectively. For the fiscal year ended December 31, 2004, Microsoft and Yahoo! accounted for approximately 36% and 23% of our revenues, respectively. Sales to Microsoft during 2004 include sales to Microsoft through a contractual intermediary.

Technology

We have a core technology focus on mechanical chassis and cabinet design to enable higher density, improved thermal management and better serviceability. We also have technologies focused on power management and power distribution, as well as hardware- and firmware-based remote management. In addition, as a result of our acquisition of Terrascale, we have our RapidScale clustered file system technology.

4

Table of Contents

Chassis and Cabinet Design.  We designed the “half-depth” 1U rack-mounted chassis and the mounting of this chassis back to back in a purpose-built cabinet. A “U” is an industry unit of measure equal to approximately 1.75 inches. We hold patents around the use of back-to-back mounting, the resulting air flow patterns and the front-facing I/O placement. We apply this innovation across a range of specific 1U, 2U, and 3U chassis types in our Foundation Series line, as well as with our Scale Out Series line.

Roamer Remote Management.  Our Roamer remote management technology is designed to provide a range of out-of-band and in-band remote management features to our customers. Our out-of-band features work even if

5

Table of Contents

the system is not functional, due to the failure of either hardware or the operating system. Out-of-band features include reboot/power cycling, stagger-start up, temperature monitoring, flashing of an LED, scripting messages to an optional LCD and changing basic input/output system, or BIOS, settings. In-band features are only accessible if both the hardware and the operating system that is running on the hardware are functional. In-band features include full serial console access to the operating system and the ability to write messages from within the operating system to the optional LCD. Roamer enables rapid identification and location of failed servers and results in a significant reduction in the need for onsite maintenance technicians. Because Roamer includes both in-band and out-of-band features, it can offer significant cost advantages versus alternative remote management solutions that may require both managed power strips and KVM (keyboard, video and mouse) and Ethernet based server remote management solutions. In addition to our serial-based Roamer remote management option, we now offer an Ethernet-based model. We also enable Intelligent Platform Management Interface, or IPMI, as supported by particular motherboard platforms, upon customer request.

We provide application programming interfaces to our remote management solution and offer a “shell mode” that enables our customers to create software scripts and integrate our remote management with their datacenter automation software platforms. For example, many of our customers utilize readily-available third-party software, such as grid, cluster and utility computing applications, in conjunction with our compute servers. Our remote management solution is based on serial or Ethernet access techniques and is compatible with any industry standard motherboard that supports a serial header and serial BIOS redirection. We are able to support serial console operating system access for Linux, Solaris, FreeBSD and current versions of Windows such as Windows Server 2003.

Power Management and Distribution.  We employ a range of techniques to minimize power consumption and enable more servers to be deployed per circuit or per power distribution unit, or PDU. We offer a DC-powered option for our Foundation Series and Scale Out Series lines, and have applied for patents associated with our technique to enable AC-DC conversion at the rack level and distribution of DC power inside the rack. Our power management techniques include:

6

Table of Contents

RapidScale Storage Appliance Software.  Leveraging our full-depth and back-to-back chassis designs, we offer RapidScale appliance configurations preloaded with our proprietary RapidScale clustered file system technology which has the following features:

Products

Our high-density compute and storage servers offer a broad range of component choices and are typically built-to-order, providing a high degree of flexibility to our customers. We believe we offer a broader range of product configurations than larger global x86 server market competitors. We continually qualify new configurations as new motherboard, processor, DRAM, disk drive and other component technologies become available. Additional information and sample current configurations follow for our Foundation Series, Scale Out Series and Storage Server lines:

Foundation Series Compute Server.  We introduced our Foundation Series compute server line in early 2000. The Foundation Series adheres to rack-mount standards, but increases density levels and improves serviceability when deployed in our proprietary, purpose-built cabinet. We currently offer Foundation Series compute servers for single or dual processor motherboards supporting AMD and Intel 32- and 32/64-bit processors such as the AMD Opteron, Intel Pentium 4 and the Intel Xeon. We also offer Foundation Series compute servers for two processor motherboards supporting the 64-bit Intel Itanium 2 processor, and for four processor motherboards supporting the 32/64-bit AMD Opteron 800 Series processors. Server configurations are typically also available with dual-core processors from both AMD and Intel, enabling even higher density. We also customize our systems to meet customer-specific requirements. Revenues from our Foundation Series compute server product line comprised 83%, 83%, and 93% of our total revenues in fiscal 2006, 2005 and 2004, respectively.

Scale Out Series Compute Server.  Our Scale Out Series compute server was introduced in August 2004 and utilizes a proprietary server form factor that is approximately half the width, half the depth and twice the height of a traditional 1U server. We mount the servers back-to-back and side-by-side in our proprietary, purpose-built cabinet. This design offers a number of advantages for large-scale compute server deployments including higher density levels, improved thermal management, enhanced serviceability and internalized cable management. We are able to offer essentially any compute server configuration that is supported in our C1000 series, with the additional benefits and feature sets offered by the Scale Out Series, summarized below. Revenues from our Scale Out Series compute server product line comprised 7%, 7% and less than 1% of our total revenues in fiscal 2006, 2005, and 2004, respectively.

7

Table of Contents

RapidScale Storage Appliance. Our RapidScale storage appliances were introduced in November 2006. It is available as a 3U, half-depth DC power model, the SA2150, and as a 3U, full-depth redundant AC model, the SA3100. The hardware configurations leverage our own chassis and cabinet technology. The software is our own proprietary clustered file system appliance software. Our appliances are based on serial ATA II drive technology. We will likely offer new versions of the appliances based on new disk drive technologies such as SAS, as they become available. We intend to continue to invest in the RapidScale software and add additional features such as snapshot, data replication and block access support. In addition to selling complete appliance-based storage solutions, we may consider selling software licenses to customers and/or OEMs. Revenues from our RapidScale storage appliances was $0.7 million in fiscal 2006.

Foundation Series Storage Server.  Our Foundation Series storage server was introduced in 2000. Our storage servers are available in a number of 3U half-depth and full-depth configurations supporting from six to 16 disk drives. Our storage servers are largely based on serial ATA and serial ATA II drive technology, but we have also qualified various parallel ATA and SCSI configurations. We intend to track high volume commodity disk drive technology and will likely offer new storage products based on new technologies, such as SAS, as they become available. Many of our customers run their own software applications on our storage servers, but we may also qualify specific third-party software applications that we pre-install on our hardware. We also source storage hardware technologies from third parties under both OEM and resale arrangements. We source our OmniStor line of external RAID arrays from a third party under an OEM arrangement. Over time, we expect to continue to expand our storage offerings through both internally developed and externally sourced products and technologies. Revenues from our Foundation Series storage segment comprised 10%, 9% and 7% of our total revenues in fiscal 2006, 2005 and 2004, respectively.

Service and Support

We offer a range of service and support packages, primarily associated with hardware break-fix support and onsite hardware service. We offer several standard levels of warranty support that vary depending on specific services, response times, coverage hours and duration. In addition to our standard offerings, we also customize our warranty and support agreements to meet the specific needs of customers. In addition, we often contract with third parties to provide support services to our customers, particularly in regions where we have limited customer concentrations and it is more cost-effective for us to do so.

We offer a range of professional services such as onsite cabling assistance and installation of third-party software such as grid, clustering and utility computing applications. We intend to expand the range of professional services that we offer, and we may do so through growing our internal staff or partnering with or acquiring other companies.

While we desire to expand our support and services revenues, we also employ a number of techniques that mitigate the cost associated with deployment and maintenance. Our compute and storage server products are designed to be shipped in a turnkey fashion, pre-racked and pre-cabled in our cabinets. We use specifically designed shipping crates to reduce the potential for shipping damage. While we also ship smaller installations in boxes, the majority of our shipments are delivered in this turnkey fashion. This enables us to minimize the time and effort associated with onsite deployment work.

On occasion, at the request of a customer that has informed us that one of our products is defective, we will ship a replacement product to that customer prior to the customer returning the defective product to us. This enables the customer to use the replacement product until the defective product is repaired and returned to the customer. Generally, we do not charge for this service unless the customer fails to return the defective product to us, in which case we will charge the customer for the replacement product.

Revenues from our service and support packages comprised 1% of our total revenues in fiscal year ended December 31, 2006, and less than 1% of our total revenues in the fiscal years ended December 31, 2005 and 2004.

8

Table of Contents

Sales and Marketing

We focus our sales and marketing activity on large enterprises with the need to purchase hundreds of servers per year. We have historically targeted accounts primarily within the United States, and we intend to replicate our current sales and operations model internationally, focusing initially on Europe and China.

Our products are sold primarily through our direct sales team, which allows us to maintain close client contact and feedback throughout the entire sales process. Our sales process begins with leads generated through targeted teleprospecting, tradeshows and referrals, which are then logged, qualified and assigned to an account executive. After an initial lead qualification, our sales executives and sales engineers collect information regarding the customer’s datacenter environment and application requirements. We then collaborate with the customer’s technical point of contact and our own internal technical resources to agree upon a particular system configuration for the customer. For larger customers, we allow evaluation of one or more hardware configurations to enable the customer to conduct their own benchmarking analyses. While we tailor solutions for most customers, there is substantial commonality in underlying components, so we are able to mitigate the impact on our manufacturing and sourcing operations.

In addition to augmenting our current sales team, our long-term plan for growing sales includes expanding distribution channels, such as value added resellers, systems integrators and independent software vendors. We also plan to examine the potential of working with large-scale OEMs both domestically and internationally.

We focus our marketing efforts on increasing brand awareness, communicating product advantages and generating qualified leads for our sales force. We rely on a variety of marketing vehicles, including our website, trade shows, advertising, public relations, industry research and collaborative relationships with suppliers and technology vendors to gain wider market access.

See Note 16 of our “Notes to Consolidated Financial Statements” included under Part II, Item 8. “Financial Statements and Supplementary Data” for additional geographic and segment information, which is incorporated herein by reference, including long-lived assets by geographic area as well as revenue from external customers attributable to segments and geographic areas during the last three years.

Product Development

Our engineering staff is responsible for the design, development, quality, documentation and release of our products. Our product development efforts focus on addressing the needs of customers deploying large, scale out computing infrastructure. These needs include density, performance, power efficiency and thermal management. We have developed cooperative working relationships with many of the world’s most advanced technology companies, such as AMD and Intel, to leverage their research and development capabilities. Additionally, we work closely with our customers to develop product innovations and incorporate these into subsequent product design. This cooperative approach allows us to develop products that meet our customers’ needs in a cost-effective manner. We monitor new technology developments, new component availability and the impact of evolving standards through customer and supplier collaboration.

Our total research and development expense was $13.6 million, $2.2 million and $0.8 million for the years ended December 31, 2006, 2005 and 2004, respectively.

Manufacturing and Operations

Our operational strategy is to provide our customers with built-to-order, high-quality turnkey solutions. Our operations are based on a hybrid manufacturing model that maintains control of key processes, including procurement, product configuration validation, quality assurance and testing, while leveraging the labor

9

Table of Contents

resources of our contract manufacturers. As a result, we are able to maintain the flexibility to scale operations as needed while minimizing capital expenditures and fixed operating costs.

Personnel in our California facilities currently manage procurement, validation of product configuration, assembly and testing. In order to control and monitor quality and inventory levels, we use a just-in-time procurement strategy. We have established relationships with key technology vendors and distributors for the supply of our core components and sub-assemblies, including chassis metal fabrication, memory, processors, hard disk drives, printed circuit boards, fans and power supplies. We generally use multiple vendors on commodity products in order to obtain competitive pricing.

As part of our hybrid manufacturing model, we maintain in-house capabilities that we typically use to fulfill urgent orders, smaller orders and prototype system production. In 2004, Sanmina-SCI manufactured and assembled the majority of our finished products, measured by revenue, at their facility in Santa Clara, California. In order to reduce our reliance on a single vendor, we established relationships with three other contract manufacturers, E-Cycle, Synnex and Realtime Technologies. During 2006, Synnex and E-Cycle manufactured and assembled the majority of our products. During 2005, Synnex manufactured and assembled the majority of our products. We continue to evaluate additional contract manufacturers in order to expand capacity and address new geographic markets. None of our contract manufacturers is the sole manufacturer of any one of our products. We maintain staff at each of the manufacturing facilities to ensure that we have adequate control over the manufacturing process and quality control.

We perform final testing, cabling and packaging onsite at our contract manufacturers’ facilities, as well as in our facilities. We deliver our cabinet-level products in specially designed shipping crates, and generally ship them fully racked, cabled and tested. This simplifies the delivery and installation process, allowing field delivery staff to roll the cabinets out of the crate and quickly connect them to the datacenter power grid and network environment.

We generally only manufacture products after receipt of orders from our customers. However, customers may generally cancel or reschedule orders without penalty, and delivery schedules requested by customers in these orders frequently vary based upon each customer’s particular needs. For these reasons, orders may not constitute a firm backlog and may not be a meaningful indicator of future revenues.

Competition

The market for our products is highly competitive, rapidly evolving and subject to changing technology, customer needs and new product introductions. In the compute server market, we compete primarily with large vendors of x86 servers based in the United States, such as Dell Inc., Hewlett-Packard Company, International Business Machines Corporation and Sun Microsystems, Inc. In addition, we compete with a number of smaller companies, some of which may become significant competitors in the future. In the storage server market, we compete primarily with EMC Corporation, Hewlett-Packard Company, Hitachi Data Systems, Ltd. and Network Appliance, Inc. As we enter international markets, we anticipate facing additional competition from foreign vendors. We currently compete principally on the basis of product features and performance, initial pricing, total cost of ownership, customer service, configurability and manageability.

Intellectual Property

We rely on a combination of patent, trademark, copyright and trade secret laws and disclosure restrictions to protect our intellectual property rights. We also enter into confidentiality and proprietary rights agreements with our employees, consultants and other third parties and control access to our proprietary information. We have nine issued patents and a number of patent applications pending in the United States and a number of patent applications pending internationally. Our most important patent relates to back-to-back positioning, front facing I/O and multi-directional air flow cooling, and expires in January 2020. If a claim is asserted that we have

10

Table of Contents

infringed the intellectual property of a third party, we may be required to seek licenses to that technology. In addition, we have a number of non-exclusive licenses from third party hardware and software vendors that allow us to resell certain hardware products and incorporate their computer software products in our product offerings.

The steps we have taken to protect our intellectual property rights may not be adequate. Third parties may infringe or misappropriate our proprietary rights. In order to protect our intellectual property rights, we may need to take legal action against those we believe infringe our rights. For example, in September 2005, we filed a patent infringement action against Super Micro Computer, Inc, and this case is still pending. In addition, the past we have brought patent infringement suits against King Star Computer, Inc. in one action, and Einux, Inc., Einux Systems, L.L.C., Einux Network Solutions, HPC Systems, Inc. and Micro-Technology Concepts, Inc. in another action. Although we were successful in the suits we brought prior to the Super Micro matter, there is no guarantee that we will be successful in protecting our intellectual property in the future. Competitors may also independently develop technologies that are substantially equivalent or superior to the technologies we employ in our products. If we fail to protect our proprietary rights adequately, our competitors could offer similar services, potentially significantly harming our competitive position and decreasing our revenues.

Employees

As of December 31, 2006, we had 286 full-time employees, consisting of 71 in manufacturing, 55 in service and support, 73 in sales and marketing, 34 in general and administrative functions, and 53 in research and development within the organization. We consider our relationships with our employees to be good.

Thomas K. Barton joined Rackable Systems in December 2002 as our President, Chief Executive Officer and a member of our board of directors. From June 2002 to December 2002, Mr. Barton provided consulting services to us as a consultant at Callero Partners. From September 2000 to May 2002, Mr. Barton was a Venture Partner and Entrepreneur in Residence at Lightspeed Venture Partners. From November 1999 to May 2000, Mr. Barton was Senior Vice President of Client Services at Red Hat, Inc., a software and services company. Mr. Barton joined Red Hat via its acquisition of Cygnus Solutions, a software development tools company. From August 1996 to its acquisition by Red Hat in November 1999, Mr. Barton held many positions at Cygnus Solutions, including interim CEO. Mr. Barton holds an A.B. in History, a B.S. in Industrial Engineering and an M.B.A. from Stanford University.

Giovanni Coglitore co-founded Rackable Systems in 1999 and has been our Chief Technology Officer since December 2002 and was a member of our board of directors from our formation until January 2006. From September 1999 to December 2002, Mr. Coglitore served as Chief Executive Officer. From April 1992 to September 1999, Mr. Coglitore served as co-founder and General Partner of International Computer Systems, a server and storage company. Mr. Coglitore holds a B.A. in History from the University of California at Los Angeles.

Todd R. Ford joined Rackable Systems in December 2002 as our Chief Financial Officer and he became our Executive Vice President of Operations in December 2003 and our President in April 2006. From June 2002 to

11

Table of Contents

December 2002, Mr. Ford provided consulting services to us as a consultant at Callero Partners. From June 2000 to March 2002, Mr. Ford served as Chief Financial Officer at Noosh, Inc., a print management software company, and served as Director of Finance at Noosh from March 1999 to June 2000. Previously, Mr. Ford worked at Ernst & Young and Arthur Andersen. Mr. Ford holds a B.S. in Accounting from Santa Clara University and is a Certified Public Accountant.

William P. Garvey joined Rackable Systems in January 2005 as our General Counsel, Vice President of Corporate Development and Secretary. From September 1997 to December 2004, Mr. Garvey served as General Counsel, Vice President of Corporate Development and Secretary for Actuate Corporation, a publicly traded software company. Mr. Garvey holds a B.S. in Applied Sciences and Engineering from the United States Military Academy and a J.D. from Stanford Law School.

Madhu Ranganathan joined Rackable Systems in December 2005 as our Vice President of Finance and she became our Chief Financial Officer in April 2006. From August 2002 to November 2005, Ms. Ranganathan served as the Vice President and Corporate Controller at Redback Networks, a publicly traded company providing telecommunication networking equipment. From January 2000 to August 2002, Ms. Ranganathan served as Vice President of Finance at Jamcracker, Inc., an application services provider. From November 1996 to December 1999, Ms. Ranganathan served as Director of Finance for BackWeb Technologies, an Internet software company. Ms. Ranganathan holds a B.S. in Finance from the University of Madras, India and an M.B.A. from the University of Massachusetts and is a Certified Public Accountant in California and a Chartered Accountant from India.

Item 1A. Risk Factors

Risks Related To Our Business and Industry

Our quarterly operating results have fluctuated significantly in the past and will continue to fluctuate in the future, which could cause our stock price to decline.

Our quarterly operating results have fluctuated significantly in the past, and we believe that they will continue to fluctuate in the future, due to a number of factors, many of which are beyond our control. We expect our revenues, gross margin and earnings per share will fluctuate on a quarterly basis in future periods. If in future periods our operating results do not meet the expectations of investors or analysts who choose to follow our company, our stock price may fall. Factors that may affect our quarterly operating results include the following:

12

Table of Contents

We are substantially dependent on a concentrated number of customers, specifically Internet companies that purchase in large quantities. If we are unable to maintain or replace our relationships with customers and diversify our customer base, our revenues may fluctuate and our growth may be limited.

Historically, a significant portion of our revenues has come from a limited number of customers. There can be no guarantee that we will be able to sustain our revenue levels from these customers because our revenues have largely been generated in connection with these customers’ decisions to deploy large-scale server and storage farms and their capacity requirements may become fulfilled. For example, our largest customer in 2005 accounted for 24% of our revenues, but did not account for any of our revenues in 2004. In 2005 our three largest customers accounted for approximately 60% of our revenues. In 2006, our two largest customers accounted for 60% of our revenues. Moreover, the proportion of our revenues derived from a limited number of customers may be even higher in any future quarter. If we cannot maintain or replace the customers that purchase large amounts of our products, or if they do not purchase products at the levels or at the times that we anticipate, our ability to maintain or grow our revenues will be adversely affected. In addition, customer concentration may expose us to credit risk, as a large portion of our accounts receivable may be from a small number of customers. For example, at December 31, 2006, 62% of our accounts receivable were owed to us from two customers.

In 2005 and 2006, a substantial majority of our revenues were generated from companies that compete in the internet services market. A significant part of our growth has been due to the fact that these companies are doing well and in many cases expanding, and, as a result, have been purchasing large quantities of products from us. If economic factors change for these Internet companies, as they have before, our ability to maintain or grow our revenues will be adversely affected.

We face intense competition from the leading enterprise computing companies in the world as well as from emerging companies. If we are unable to compete effectively, we might not be able to achieve sufficient market penetration, revenue growth or profitability.

The markets for compute server products and storage products are highly competitive. Our most significant competitors, Dell, Inc., Hewlett-Packard Company, International Business Machines Corporation and Sun Microsystems, Inc. in the compute server market and EMC Corporation, Hewlett-Packard Company, Hitachi Data Systems, Ltd. and Network Appliance, Inc. in the storage market, represent some of the most established companies in the computer industry. These competitors have the following advantages over us:

Because these competitors have greater financial strength than we do and are able to offer a more diversified bundle of products and services, they may have the ability to severely undercut the pricing of our products, which would make us less competitive or force us to reduce our average selling prices, negatively impacting our margins. In the fourth quarter of 2006, we had a large transaction where a competitor undercut our prices causing us to reduce our price, which negatively impacted our gross margin on that transaction and our overall gross margin for the quarter. In addition, we have on occasion lost sales opportunities due to a competitor undercutting the pricing of our products or maintaining superior brand recognition. These competitors may be able to develop products that are superior to the commercially available components that we incorporate into our products, or

13

Table of Contents

may be able to offer products that provide significant price advantages over those we offer. For instance, a competitor could use its resources to develop proprietary motherboards with specifications and performance that are superior in comparison with the platforms that are currently available to the marketplace, which could give that competitor a distinct technological advantage. In addition, if our competitors’ products become more accepted than our products, our competitive position will be impaired.

As the enterprise computing industry evolves, we expect to encounter additional competitors, including companies in adjacent technology businesses such as storage and networking infrastructure and management, companies providing technology that is complementary to ours in functionality, such as datacenter management software, contract manufacturers, and other emerging companies that may announce server product offerings. Moreover, our current and potential competitors, including companies with whom we currently have strategic alliances, may establish cooperative relationships among themselves or with other third parties. If this occurs, new competitors or alliances may emerge that could negatively impact our competitive position.

Our products incorporate open standard, commoditized components and materials that we obtain in spot markets, and, as a result, our cost structure and our ability to respond in a timely manner to customer demand are sensitive to volatility of the market prices for these components and materials.

A significant portion of our cost of goods sold is directly related to the pricing of commoditized materials and components utilized in the manufacture of our products, such as memory, hard drives and central processing units, or CPUs. As part of our procurement model, we do not enter into long-term supply contracts for these materials and components, but instead purchase these materials and components in a competitive bid purchase order environment with suppliers or on the open market at spot prices. As a result, our cost structure is affected by the availability and price volatility in the marketplace for these components and materials, especially for dynamic random access memory, or DRAM, and new versions of hard drives and CPUs that are introduced by our suppliers. In the fourth quarter of 2006, we experienced volatility in DRAM prices that caused us to procure DRAM at prices higher than anticipated, which negatively impacted our overall gross margin for the quarter. This volatility makes it difficult to predict expense levels and operating results and may cause them to fluctuate significantly. In addition, if we are successful in growing our business, we may not be able to continue to procure components solely on the spot market, which would require us to enter into contracts with component suppliers to obtain these components. This could increase our costs and decrease our gross margins.

In addition, because our procurement model involves our ability to maintain a low inventory and to acquire materials and components as needed, and because we do not enter into long-term supply contracts for these materials and components, we may be in a position in which our ability to effectively and efficiently respond to customer orders may be constrained by the then-current availability or the terms and pricing of these materials and components. Our industry has experienced component shortages and delivery delays in the past, and in the future we may experience shortages or delays of critical components as a result of strong demand in the industry or other factors. As one example, DRAM can represent a significant portion of our cost of revenues, and both the price and availability of various kinds of DRAM are subject to substantial volatility in the spot market. In the past, we have encountered situations where we were forced to pay higher prices than we anticipated for DRAM, which occurred in the fourth quarter of 2006, and we have encountered situations where DRAM was in tight supply and we were unable to deliver customer orders on their anticipated delivery dates. As another example, the industry in the past has experienced a shortage of selected Intel chips, which caused some of our motherboard suppliers to reduce or suspend shipments of motherboards using these chips. This impacted our ability to ship selected configurations to some of our customers, and in some cases accelerated a transition to other platforms. If shortages or delays arise, the prices of these components may increase or the components may not be available at all. We may not be able to secure enough components at reasonable prices or of acceptable quality to build new products to meet customer demand, which could adversely affect our business and financial results.

14

Table of Contents

We intend to expand our operations and increase our expenditures in an effort to grow our business. If we are not able to manage this growth and expansion, or if our business does not grow as we expect, our operating results may suffer.

We intend to continue to grow our business by entering new markets, developing new product and service offerings and pursuing new customers. In connection with this growth, we expect that our annual operating expenses will increase over the next several years as we expand our sales and marketing, research and development, manufacturing and production infrastructure, and our customer service and support efforts. Our failure to timely or efficiently expand operational and financial systems and to implement or maintain effective internal controls and procedures could result in additional operating inefficiencies that could increase our costs and expenses more than we had planned and might cause us to lose the ability to take advantage of market opportunities, enhance existing products, develop new products, satisfy customer requirements, respond to competitive pressures or otherwise execute our business plan. Additionally, if we do increase our operating expenses in anticipation of the growth of our business and this growth does not meet our expectations, our financial results will be negatively impacted.

If we acquire or invest in other companies, assets or technologies and we are not able to integrate them with our business, or we do not realize the anticipated financial and strategic goals for any of these transactions, our financial performance may be impaired.

If appropriate opportunities present themselves, we may consider acquiring or making investments in companies, assets or technologies that we believe are strategic. We only have limited experience in doing so, and if we do succeed in acquiring or investing in a company, asset or technology we will be exposed to a number of risks, including:

In addition, from time to time we may enter into negotiations for acquisitions or investments that are not ultimately consummated. These negotiations could result in significant diversion of management time, as well as substantial out-of-pocket costs. If we were to proceed with one or more significant acquisitions or investments in which the consideration included cash, we could be required to use a substantial portion of our available cash. To the extent we issue shares of capital stock or other rights to purchase capital stock, including options and warrants, existing stockholders might be diluted and earnings per share might decrease. In addition, acquisitions and investments may result in the incurrence of debt, large one-time write-offs, such as of acquired in-process research and development costs, and restructuring charges.

15

Table of Contents

We have begun to expand our sales into international markets and intend to further expand our sales into additional international markets, which may be more difficult than we expect, and if we are unable to do so successfully, our revenues and operating results may be adversely impacted.

One component of our growth strategy is to expand into international markets, which includes our establishment of subsidiaries in Ireland during September 2005 and in Hong Kong in September 2006. However, we have limited experience in selling our systems overseas, and we may encounter unexpected difficulties in doing so. If we are not able to successfully expand into international markets, our ability to grow our business will be adversely affected. Some of the factors that may impact our ability to initiate and maintain sales in foreign markets include:

Our business depends on decisions by potential customers to adopt our modular, open standard-based products and to replace their legacy server systems with our products, and they may be reluctant to do so, which would limit our growth.

Our business depends on companies moving away from large proprietary RISC/UNIX servers, to servers based on modular, open standard-based architecture, including servers that run on Linux and Microsoft Windows operating systems and that utilize commercially available x86 processor architectures. If enterprises do not continue to adopt this open standard-based approach, the market for our products may not grow as we anticipate and our revenues would be adversely affected. Excluding sales to Microsoft Corporation, we believe that a majority of the server systems that we sold in 2004, 2005 and 2006 ran on the Linux operating system, and are subject to the GNU General Public License. Pending litigation involving the SCO Group’s claim against IBM that Linux is an unauthorized derivative work of the UNIX ^® operating system, if it results in a ruling that users of Linux must pay royalties to SCO or others, could impede broader Linux adoption and could materially harm our ability to sell our products based on the Linux operating system. In addition, the GNU General Public License is a subject of litigation, and it is possible that a court could hold these licenses to be unenforceable in that litigation or that a party could prove a claim for proprietary rights in the Linux operating system or other programs developed and distributed under these licenses. Any ruling by a court that these licenses are not enforceable, or that the Linux operating system or significant portions of it may not be copied, modified or distributed subject only to the minimal restrictions contained in these licenses, could also impede broader Linux adoption and limit our ability to sell our products based on the Linux operating system. Further, because potential customers have often invested significant capital and other resources in existing systems, many of which run mission-critical applications, customers may be hesitant to make dramatic changes to their datacenter systems. The failure of our customers and potential customers to replace their legacy server systems and adopt open standard-based modular technologies could have a material adverse impact on our ability to maintain or generate additional revenues.

We rely primarily on our direct sales force to generate revenues, and may be unable to hire additional qualified sales personnel in a timely manner or retain our existing sales representatives.

To date, we have relied primarily on our direct sales force to sell our products in the United States. Because we are looking to expand our customer base and grow our sales to existing customers, we will need to continue to hire qualified sales personnel if we are to achieve our anticipated revenue growth. The competition for qualified

16

Table of Contents

sales personnel in our industry, and particularly in Silicon Valley, is very intense. If we are unable to hire, train, deploy and manage qualified sales personnel in a timely manner, our ability to grow our business will be impaired. For example, in the past, it has taken us up to six months to hire a qualified sales executive and it may take a newly- hired sales executive up to nine months after hiring to become productive, resulting in aggregate lag time between the commencement of the search to productivity in excess of one year. In addition, if we are unable to retain our existing sales personnel, our ability to maintain or grow our current level of revenues will be adversely affected. In addition, a large percentage of our revenue is generated from a small number of customers. If we are unable to retain the sales personnel who are responsible for those customer accounts, our business could be negatively impacted.

We are developing a channel strategy to generate additional sales and revenues, and the failure to implement this strategy successfully might affect our ability to sustain revenue growth and may harm our business and operations.

An increasing portion of our sales strategy is to develop our sales efforts through the use of resellers and other third parties to sell our systems. We may not be successful in building or expanding relationships with these third parties. Further, even if we do develop and expand these relationships, they may conflict with our direct sales efforts in some territories. Ineffective marketing of our products by our resellers or disruptions in our distribution channels could lead to decreased sales or slower than expected growth in revenues and might harm our business and operations.

Our sales cycle requires us to expend a significant amount of resources, and could have an adverse effect on the amount, timing and predictability of future revenues.

The sales cycle of our products, beginning from our first customer contact to closing of the sale, often ranges from three to six months. We may expend significant resources during the sales cycle and ultimately fail to close the sale. The success of our product sales process is subject to factors over which we have little or no control, including:

We expend substantial time, effort and money educating our current and prospective customers as to the value of our products. Even if we are successful in persuading lower-level decision makers within our customers’ organizations of the benefits of our products, senior management might nonetheless elect not to buy our products after months of sales efforts by our employees or resellers. If we were unsuccessful in closing sales after expending significant resources, our revenues and operating expenses will be adversely affected.

If we are unable to protect our intellectual property adequately, we may not be able to compete effectively.

Our intellectual property is critical to our success and our ability to compete. If we fail to protect our intellectual property rights adequately, our competitors might gain access to our technology. Unauthorized parties may attempt to copy or otherwise obtain and use our proprietary technology despite our efforts to protect our intellectual property. In September 2005, we filed a patent infringement action against one defendant in response to which the defendant counterclaimed that two of our patents are invalid and not infringed. In addition, we previously initiated two patent infringement actions against six defendants, seeking to enforce our intellectual property rights. We obtained consent judgments and permanent injunctions against five of the defendants, and reached a confidential settlement with the sixth.

17

Table of Contents

However, litigation is inherently uncertain, and there is no assurance that any litigation we initiate will have a successful outcome. Monitoring unauthorized use of our technology is difficult, and we cannot be certain that the steps we have taken will prevent unauthorized use of our technology, particularly in foreign countries where we have two patent applications pending but do not have any issued patents, and the laws may not protect our proprietary rights as fully as the laws of the United States. Any claims or litigation that we have initiated or that we may initiate in the future to protect our proprietary technology could be time consuming and expensive and divert the attention of our technical and management resources whether or not the claims or litigation are decided in our favor. We currently have a small number of patents issued in the United States and a number of utility patent applications pending. Patents may not be issued from these patent applications, and even if patents are issued, they may not benefit us or give us adequate protection from competing products. For example, issued patents might be circumvented or challenged, as is the case in the current patent infringement action, and could be declared invalid or unenforceable. Moreover, if other companies develop unpatented proprietary technology similar to ours or competing technologies, our competitive position will be weakened.

If we are found to have violated the intellectual property rights of others, we could be required to indemnify our customers, resellers or suppliers, redesign our products, pay significant royalties and enter into license agreements with third parties.

Our industry is characterized by a large number of patents, copyrights, trade secrets and trademarks and by frequent litigation based on allegations of infringement or other violation of intellectual property rights. As we continue our business, expand our product lines and our product functionality, and expand into new jurisdictions around the world, third parties may assert that our technology or products violate their intellectual property rights. Although we have no knowledge of being in infringement, we cannot assure you that we are not in infringement of third party patents. Any claim, regardless of its merits, could be expensive and time consuming to defend against and would divert the attention of our technical and management resources. Successful intellectual property claims against us could result in significant financial liability or prevent us from operating our business or portions of our business. In addition, resolution of claims may require us to redesign our technology, to obtain licenses to use intellectual property belonging to third parties, which we may not be able to obtain on reasonable terms, to cease using the technology covered by those rights, and to indemnify our customers, resellers or suppliers. Any of these events could materially harm our business, financial condition and results of operations.

If we lose the services of one or more members of our current executive management team or other key employees, or if we are unable to attract additional executives or key employees, we may not be able to execute on our business strategy.

Our future success depends in large part upon the continued service of our executive management team and other key employees. In particular, Thomas K. Barton, our chief executive officer, and Todd R. Ford, our president, are critical to the overall management of our company as well as to the development of our culture and our strategic direction. Giovanni Coglitore, our chief technology officer and one of our founders, plays a key role in our future technology and product development and the service and support of our existing customers. All of our employees are “at will” and their employment can be terminated by us or them at any time. The loss of services of any of these executives or of one or more other members of our executive management or sales team or other key employees could seriously harm our business. Competition for qualified executives is intense, particularly in Silicon Valley, and if we are unable to continue expanding our management team or successfully integrate new additions to our management team in a manner that enables us to scale our business and operations effectively, our ability to operate effectively and efficiently could be limited.

18

Table of Contents

If we fail to maintain or expand our relationships with our suppliers, we may not have adequate access to new or key technology necessary for our products, and, as a result, our ability to deliver leading-edge products may be impaired.

In addition to the technologies we develop, our suppliers develop product innovations at our direction that are requested by our customers. In many cases, we retain the ownership of the intellectual property developed by these suppliers. In addition, we rely heavily on our component suppliers, such as Intel and Advanced Micro Devices, to provide us with leading-edge components. If we are not able to maintain or expand our relationships with our suppliers or continue to leverage their research and development capabilities to develop new technologies desired by our customers, our ability to deliver leading-edge products may be impaired and we could be required to incur additional research and development expenses.

We depend on our Foundation Series servers for the majority of our revenues. If the market acceptance of our Foundation Series servers does not continue, we may not be able to achieve or sustain our anticipated growth.

We sold our first Foundation Series servers in February 2000. For fiscal 2006 and 2005, sales of our Foundation Series product line accounted for approximately 93% and 92%, respectively, of our revenues. If our Foundation Series servers fail to maintain market acceptance, or if we are unsuccessful in developing improved products or products to replace or supplement our current product line, we may not grow our business and revenues as we expect. Further, because our customers are engaged in large-scale data center implementations, if customers believe that new generations of our products will become available in the near future, this perception may cause customers to delay or cancel existing orders, which would affect our ability to generate revenues in accordance with forecasted levels.

We rely on contract manufacturers to manufacture our products, and our failure to successfully manage our relationships with these contract manufacturers could impair our ability to deliver our systems in a manner consistent with required volumes or delivery schedules, which could damage our relationships with our customers and decrease our revenues.

We rely on a very small number of contract manufacturers to assemble and test a majority of our products. None of these third-party contract manufacturers are obligated to perform services or supply products to us for any specific period, or in any specific quantities, except as may be provided in a particular purchase order. Moreover, none of our contract manufacturers has provided contractual assurances to us that adequate capacity will be available to us to meet future demand for our products. If our contract manufacturers are not able to maintain our high standards of quality, are not able to increase capacity as needed, or are forced to shut down a factory, our ability to deliver quality products to our customers on a timely basis may decline, which would damage our relationships with customers, decrease our revenues and negatively impact our growth.

Our customers require a high degree of reliability in our products and services, and if we cannot meet their expectations our relationships with our customers could be damaged and demand for our products and services will decline.

Because our customers rely on our products and services for their enterprise or mission critical applications, any failure to provide high quality products and reliable services, whether caused by our own failure or failures by our suppliers or contract manufacturers, could damage our reputation and reduce demand for our products and services. In addition, delays in our ability to fill product orders as a result of quality control issues, such as an increase in failure rates or the rate of product returns, may negatively impact our relationships with our customers and harm our revenues and growth.

19

Table of Contents

If we are not able to recover sales and use taxes from past sales to several of our customers, we may have to pay those taxes out of our funds and record an additional charge to our operating results.

In December 2005, we identified potential state sales and use tax liabilities relating to certain of our product sales to customers outside of California. We estimated the gross tax liability to the states identified to be $1.2 million which was recorded in accrued sales and use tax as of December 31, 2005. We also quantified the exposure relating to interest and estimated collection exposures and, accordingly, recorded a $0.4 million charge to general and administrative expense in the fourth quarter of 2005 with an offsetting increase to other accrued expenses as of December 31, 2005. For the year ended December 31, 2006, we increased our estimated gross tax liability by $5.3 million to $6.5 million which was recorded in accrued sales and use tax. The increase in the estimated gross tax liability was a result of additional information received during the year relating to customers and their self assessment process. We also recorded a $0.3 million charge to general and administrative expense in 2006 with an offsetting increase to other accrued expenses as of December 31, 2006, increasing the balance included in other accrued expenses to $0.7 million. Although we believe that we will recover the sales tax from these customers, and accordingly have recorded a receivable for $6.5 million (included in other current assets), if we are not able to recover these sales taxes from these customers, we will record an additional charge to our operating results and need to pay these taxes out of our funds.

Maintaining and improving our financial controls and the requirements of being a public company may strain our resources, divert management’s attention and affect our ability to attract and retain qualified board members.

We are subject to the reporting requirements of the Securities Exchange Act of 1934 or the Exchange Act, the Sarbanes-Oxley Act of 2002 and The NASDAQ Stock Market Rules. The requirements of these rules and regulations will increase our legal and financial compliance costs, make some activities more difficult, time-consuming or costly and may also place undue strain on our personnel, systems and resources. The Exchange Act requires, among other things, that we file annual, quarterly and current reports with respect to our business and financial condition.

The Sarbanes-Oxley Act requires, among other things, that we maintain effective disclosure controls and procedures and internal control over financial reporting. This can be difficult to do. For example, in connection with our audit for the years ended December 31, 2003 and 2004, our independent registered public accounting firm noted a reportable condition with respect to the method of accounting that we applied to the embedded derivatives related to our mandatorily redeemable preferred stock, the preferred stock issued to two executives in exchange for promissory notes and certain stock options. As a result, audit adjusting entries were required for the years ended December 31, 2003 and 2004. Due to the magnitude of the adjustments, this reportable condition was determined to be a material weakness. The accounting for equity and derivative instruments is complex, and the relevant accounting implications must be closely monitored, researched and evaluated. In order to maintain and improve the effectiveness of our disclosure controls and procedures and internal control over financial reporting, significant resources and management oversight are required. As a result of this and similar activities, management’s attention may be diverted from other business concerns, which could have a material adverse effect on our business, financial condition and results of operations. In addition, if we are unable to continue to meet these requirements, we may not be able to remain listed on the NASDAQ Global Select Market.

In connection with these efforts, we implemented a new Enterprise Resource Planning, or ERP system in 2006. If we are unable to effectively use the new ERP system and fail to implement Sarbanes-Oxley compliance-related activities in a timely and efficient manner, management’s attention may be diverted from other business concerns and we may encounter difficulties in accurately reporting on our financial performance on a timely basis, either of which could have a material adverse effect on our business, financial condition and results of operations.

Under the Sarbanes-Oxley Act and NASDAQ rules, we are required to maintain an independent board. If we are unable to maintain adequate directors’ and officers’ insurance, our ability to recruit and retain qualified

20

Table of Contents

directors, especially those directors who may be deemed independent for purposes of NASDAQ rules, and officers will be significantly curtailed.

Risks Related to Owning Our Stock

Our stock price in the past has been volatile, and may continue to be volatile or may decline regardless of our operating performance, and investors may not be able to resell shares at or above the price at which they purchased the shares.

Our stock has been publicly traded for a relatively short period of time, having first begun trading in June 2005. During that time our stock price has fluctuated significantly, from a low of approximately $11 per share to a high of approximately $56 per share. At times the stock price has changed very quickly. For example, at the end July 2006 our stock price dropped from approximately $35 per share to approximately $21 per share. If investors purchase shares of our common stock, they may not be able to resell those shares at or above the price at which they purchased them. The market price of our common stock may fluctuate significantly in response to numerous factors, many of which are beyond our control, including:

In addition, the stock markets, and in particular the Nasdaq Market, have experienced extreme price and volume fluctuations that have affected and continue to affect the market prices of equity securities of many technology companies. Stock prices of many technology companies have fluctuated in a manner unrelated or disproportionate to the operating performance of those companies. In the past, stockholders have instituted

21

Table of Contents

securities class action litigation following periods of market volatility. If we were involved in securities litigation, it could have substantial costs and divert resources and the attention of management from our business.

Due to these factors, sales of a substantial number of shares of our common stock in the public market could occur at any time. These sales, or the perception in the market that the holders of a large number of shares intend to sell shares, could reduce the market price of our common stock.

Some provisions in our certificate of incorporation and bylaws may deter third parties from acquiring us.

Our certificate of incorporation and bylaws contain provisions that may make the acquisition of our company more difficult without the approval of our board of directors, including the following:

These anti-takeover defenses could discourage, delay or prevent a transaction involving a change in control of our company. These provisions could also discourage proxy contests and make it more difficult for stockholders to elect directors of their choosing and cause us to take other corporate actions they desire.

We do not expect to pay any cash dividends for the foreseeable future.

We do not anticipate that we will pay any cash dividends to holders of our common stock in the foreseeable future and our credit facility with HSBC prevents us from doing so. Accordingly, investors must rely on sales of their common stock after price appreciation, which may never occur, as the only way to realize any future gains on their investment. Investors seeking cash dividends in the foreseeable future should not purchase our common stock.

Item 1B. Unresolved Staff Comments

Not applicable.