In-Depth
Top 10 Server Trends for 2010
Understanding these ten server trends can help CIOs succeed despite an unsteady economy.
By Jim Ganthier
As organizations worked to squeeze the most out of their existing IT infrastructure and budgets in 2009, many looked to technology to help them do more with less.
Research shows that more than 90 percent of senior business decision makers believe business cycles will continue to be unpredictable in the next few years. As a result, 80 percent recognize a need to be far more flexible in their approaches to business and technology (see Note 1). This requires IT to build elastic infrastructures that can easily be scaled up or down as business needs change.
In addition to flexibility, 84 percent of senior business leaders in the same studybelieve innovation will be critical to their organization's success, and 71 percent would sanction more technology investments if they could see how those investments met their organization's time-to-market and business opportunity needs.(1)
CIOs need technology capable of helping them tackle the challenges of today's unsteady business environment while laying an infrastructure foundation that can easily scale to support future growth. The innovations built into many of the latest generations of servers help CIOs increase system performance while slashing energy bills. Such a platform will be instrumental in enabling an organization to achieve a faster return-on-investment, delivering more value from every server dollar spent.
Marking 20 years of x86 server experience this year, HP has identified the top ten server trends for the New Year that will help CIOs succeed despite an unsteady economy:
1. "Urge to Converge" -- Convergence of Computing, Networking, and Storage
A converged infrastructure unifies software, servers, storage and networking to improve business, application and infrastructure functions. A converged infrastructure enables simpler management of complex technology environments. Further, organizations are able to use the individual components of their data centers in a synergistic manner, reducing costs and accelerating business growth.
2. "Ultimate Virtual" -- Virtualization Beyond the Server to the Infrastructure
Although virtualization originally gained the most traction with individual servers, we now see virtualization being applied more broadly so that servers, storage, and networking resources can be pooled and utilized across the data center. You'll see companies continue to apply the principles of virtualization throughout the infrastructure in ways that could only be imagined a few years ago. Client virtualization will continue to grow as organizations look to meet their employees' mobility needs and take advantage of declining acquisition costs.
With centralized, mission-critical security and reliability, client virtualization offers a breakthrough workstation computing experience that allows individuals or teams collaborate, remotely, and securely.
3. "Scale Up" -- Scale-out/Skinless
As organizations in Web 2.0, cloud, and high-performance computing (HPC) wrestle with rapidly growing mega data centers hosting thousands of compute nodes, these organizations are looking to leverage technology to maintain a competitive advantage in a crowded marketplace. A "skinless" server design strips out unused features to drive down costs and energy requirements, enabling new business models. Skinless servers help customers scale without limits and improve data center efficiency with a lightweight high-performance, flexible, and easy-to-deploy solution.
4. "Be Cool" -- Power and Cooling/On-board Sensors/Power Management
Heat dissipation is a key issue for servers today due to increased server density and workload demands that force IT administrators to make the most of available power. Smart sensors designed into a server can greatly improve their thermal performance while reducing energy loss, thereby lowering power and cooling costs in the data center. Expect to see an upswing in sensors that automatically control server thermals to optimize power usage and cooling.
Increased focus will be placed on power management because it enables companies to reclaim trapped power and cooling capacity by safely limiting server power consumption. In fact, organizations have the potential to more than triple the capacity of an existing data center through the combination of these intelligent server technologies and more effective management tools.
5. "Keep Up with Demands" -- Virtual Server-to-network Connections
As companies roll out blade-based virtualization deployments in order to gain the benefits of agility and reduced costs, many find that they have limited choices for network input/output (I/O). Although virtualization has led to server and equipment consolidation, the technology has had the opposite effect on networks by actually driving up bandwidth requirements.
Virtual machines do help bring more applications onto each server, but increased server utilization requires more network bandwidth per server. This leads to increased complexity and network costs. Expect more companies to adapt interconnect technology designed to simplify the connections of servers to data center networks to maximize the benefits of virtualization.
6. "Orchestrate the Environment" -- Orchestration and Management of Service Delivery
Traditional data centers are built upon aging, siloed architectures, limiting IT's ability to improve efficiencies and service levels. Lifecycle management capabilities are typically limited and often driven through manual, labor-intensive processes. Coupled with the rise of virtualization where virtual machines are easily added and stale machines persist unchecked, management complexities increase further. A next-generation data center allows CIOs to focus on business innovation and meeting dynamic market demands. To do this requires an infrastructure operating environment that automates provisioning, modification, and management of applications.
Companies will look to adopt a shared-services engine that can provision and adapt infrastructure on-the-fly, ensuring the technology can instantly respond to changing business demands. In this way, customers can simultaneously control and optimize all elements of an infrastructure that are needed to deliver a technology service -- including servers, storage, network connections, and facility resources. With advanced management capabilities, customers can provision and adapt complex infrastructures in minutes instead of weeks or even months.
7. "Stack the Rack" -- Growth of Rack-mount Servers
Rack mount servers offer increased flexibility over tower options because they can be slotted into an equipment rack. This allows equipment of all sorts to be mounted on top of each other, using less space than if they were standing side by side. Enterprises have long adopted rack-mount servers, and they are becoming the standard for growing small and midsize businesses. SMBs can particularly benefit from rack mount cabinets if they do not have standalone data centers.
Once an organization begins to grow to the point where it has a few small servers dotted around offices with loud fans running next to people working, a rack-mount server configuration can reduce noise. Look for the advent of racks of all shapes and sizes to come to market with removable doors and side panels.
8. "PODnomena" -- Advent of the Mobile Data Center
As organizations confront increasing infrastructure workloads, they must figure out how to reconcile growing demands with their traditional brick-and-mortar data centers. Enter "containerized" data centers that integrate servers, storage, and networking in a mobile, "pay-as-you-grow" data center alternative, delivering the right amount of capacity when an organization needs it.
Data center containers provide a more flexible architecture, faster time to deployment, and significantly reduced overall power consumption. This allows customers to tailor their data center strategy to deliver IT resources exactly when and where they need more capacity, density, and power. These containerized data center options will continue to populate the market in 2010, as customers explore alternative facility strategies that are energy efficient, portable, and flexible.
9. "Proud of the Cloud" -- Advent of Private Cloud/Shared Services
Cloud computing reached a fever pitch in 2009 and innovation will continue in 2010. Cloud computing has evolved to encompass a mix of many technologies and methods that have existed for years, even decades, as well as more recent developments that enable computing to be hosted remotely. Although more vendors crowd the market touting their cloud capabilities, customers will quietly be rolling out internal shared-services environments or "private clouds" that deliver similar benefits in speed and cost without the security risks of the "public cloud."
10. "Rock the Box" -- Movement Toward Blade Servers and All-in-one Units
Blade servers are part of a computer configuration in which power, cooling, storage, and connectivity are largely provided by an outer housing or chassis. Blade servers -- each one making up a complete computer or service device -- contain vital processing and storage elements.
With the slowdown of the economy in 2009 inflicting a deteriorating effect on technology sales, blade servers emerged as the bright spot because they outperformed the rest of the market in sales and revenues. Accelerated growth and adoption of blade technologies will continue over the next year as customers seek power-efficient, small-footprint, computing options.
Organizations can expect to see continued innovation in server performance, energy efficiency, virtualization capabilities, and management in 2010.
Note 1: HP Research: Thrive in Unpredictability, Coleman Parkes Research Ltd, October 2009
Jim Ganthier is vice president of marketing, Industry Standard Servers, at HP. You can contact the author at [email protected]