I D C V E N D O R S P O T L I G H T
T h e N e w D a t a c e n t e r : H o w V i r t u a l i z a t i o n a n d
C o n v e r g e n c e A r e C h a n g i n g t h e C o r p o r a t e
N e t w o r k
October 2010
Adapted from Datacenter Network Customer Requirements, by Lucinda Borovick, IDC #217999
Sponsored by Force10
Economic conditions, combined with the fundamental way in which business is conducted in a networked global world, are forcing enterprises to rethink their IT infrastructure. Virtualization and convergence have become critical components in shifting the datacenter from a capital expense center to an operational investment in the business. This shift is requiring organizations to change how IT resources, both virtual and actual, are linked together. This Vendor Spotlight takes a look at how virtualization and convergence are affecting the datacenter’s networking requirements. The paper also looks at Force10 Networks, a supplier of high-performance networking solutions and its role in this strategically important market.
Introduction
While the datacenter has always been a strategic asset to enterprises, IT often has been considered a support organization. The new global economy is changing this mindset, however. Today, IT and the datacenter are critical drivers of business in a networked world.
But the recent global economic crisis has had a profound effect on organizations’ investments in IT and operational priorities. Some of the changes, such as delayed capital expenditures and staff reductions, are temporary, but others have accelerated long-developing, fundamental transformations in business practices and IT operations. Senior executives want to shift IT investments from capital expenditures to operational expenditures to make it easier for enterprises to react quickly to major changes in a networked business environment.
This shift to operations-based IT is becoming a priority for corporate and IT executives alike. Both recognize that the following two key elements in a successful IT transformation effort are:
Boost operational efficiency by:
• Continuing aggressive consolidation of servers, storage and network assets through the
use of technologies such as virtualization (server and storage) and unified networks (10 Gigabit Ethernet or 10 GbE)
• Improving utilization of installed IT assets with technologies such as thin provisioning and
data deduplication
• Reducing costs associated with system, data, and application maintenance and migration
through use of live migration and automated data movement
Boost responsiveness to business needs through:
• Faster, automated provisioning of IT assets
• Reduction of corporate risk from data loss or misuse
• Transitioning to a World of Virtualized IT
The key requirement for IT today is to find ways to gain greater use and value from the IT assets that have already been deployed. This is true across the IT spectrum including servers, storage, and networking. The area where most companies have seen the most dramatic and tangible utilization benefits is the use of server virtualization for x86 systems. As a result, it is not surprising that surveys show that server virtualization remains the top trend in IT provisioning.
Based on typical server replacement cycles and new application deployments, IDC believes that the majority of all installed server applications will be running as virtual machines. In fact, 2009 was the first year in which more new application instances were deployed as virtual machines on a virtualized server than were deployed on a dedicated physical server. This rapid virtualization of server assets is part of corporate efforts to reduce capital costs through aggressive consolidation while boosting asset utilization.
IDC finds, however, that expectations for further boosts in IT asset use and operational efficiency often fail to materialize. The shift to virtualized servers often leads to significant disruptions,
particularly in making servers, storage and networks work in harmony. These disruptions include the overloading of storage and data network facilities, the overprovisioning of storage capacity and associated increase in the need for storage administration and missed or incomplete data and
application recovery. In combination, these problems can quickly overwhelm any hopedfor benefits as the scope of virtual server deployment expands.
Creating a Converged IT Infrastructure
Critical to the shift toward operational-based IT is developing a long-term strategy for building a virtualized datacenter with a converged, network-based IT infrastructure. Typically, enterprises undertake a phased, two-step approach:
Movement to networked storage. This allows an organization to better take advantage of live
migration technologies to reduce planned downtime. This includes migration to a unified network (typically 10 GbE based) for both storage and data network traffic to reduce cabling and
reconfiguration costs while also providing greater network capacity for future growth.
Virtualization of storage. Using virtual volumes and thin provisioning allows companies to avoid
both storage administration overload and the massive over provisioning of storage for virtualized servers.
While these steps are critical for regaining control of fast-growing virtual environments, they are simply the start of a long-term, more systematic approach to deploying and managing IT assets in the highly virtualized datacenter of the future. Faced with a future where organizations need to deploy and effectively use hundreds, thousands, even tens of thousands of server (and/or desktop) application instances in a virtual environment, IT teams are rethinking the entire IT infrastructure process from purchasing, to configuration, to ongoing operations. The underlying goals of these efforts include the following:
Enabling more cost effective and predictable spending on servers, storage and network
equipment
Ensuring that all of purchased IT assets are used to their maximum level
Reducing the operational and management burdens associated with new application deployment,
The key foundation underlying many organizations' strategy for meeting these goals is to leverage emerging converged IT infrastructure solutions. A converged infrastructure is the development of a datacenter based on a set of networked standard elements that can be dynamically partitioned and automatically rebalanced to support a range of different business applications. The goal is no longer to deploy each element – server, storage, network, etc. – individually, but to build the optimal environment for creating an agile IT infrastructure.
The immediate benefit of creating networked pools of converged infrastructure is an upfront reduction in spending on hardware assets. In part, this is because organizations can avoid wasteful over provisioning and the continual rebalancing of server, storage, and network resources as applications evolve. Equally important, though less glamorous, IT organizations hope to get rid of much of the cable cluttering up their datacenters, and this can bring surprisingly large savings. Converged infrastructures also help in scaling IT to meet the growing needs of both revenue generating and operational requirements.
But savings alone can’t justify the shift to converged IT hardware. Without advanced, unified, cross tier network management capabilities, enterprises simply have a bunch of bundled hardware that can’t do much dynamically as needs change. The real benefits come when IT teams can significantly improve the utilization rates for all of the IT assets in the datacenter.
In addition to bundling, organizations need to add unified orchestration functions that automate provisioning and enable policy-based resource management throughout the network for hardware and the applications running on it.
Ultimately, the success of a converged IT infrastructure depends on extending efficient operations and automated resource management across multiple assets in the datacenter and across multiple datacenters.
Without this extension of unified orchestration, IT organizations will just be creating new stovepipes, albeit at a different scale. They also will be unable to deliver effective and efficient DR/business continuity. Navigating the phases of converged IT infrastructure maturity also will be at the core of many organizations' efforts to implement a private cloud for their business applications.
Many private cloud efforts focus on the shift to a service-oriented application (SOA) architecture. At their core, however, they depend upon the use of a robust, scalable, and highly efficient IT
infrastructure with virtualized IT resources, live application and data migration, performance monitoring and metering and automated provisioning to enable rapid application deployment and expansion.
A major pain point in this move to a converged IT infrastructure is optimizing traffic over the network to take advantage of available bandwidth. If there is a reduction in telecommunication purchases, then upgrading expensive WAN links may no longer be included in the budget. WAN application delivery solutions can increase throughput in existing WAN links, thereby avoiding the need to upgrade and reducing opex. Using the right approaches, enterprises can get the most out of, and even improve performance over their existing WAN links without increasing their opex spending. To accomplish this, IT must look beyond traditional networking schemes to higher performance, rapidly configurable data networks.
Considering Force10 Networks
Force10 Networks is a San Jose, Calif.-based provider of high-performance networking solutions. The company solutions are used by organizations whose very business depends on high density networks, such as Salesforce.com, Yahoo!, e-Harmony, Alibaba, Novartis, NYSE Euronext, Dow Jones and others. Partnerships include VMware for virtualization, F5 /Citrix for application
acceleration, Dell/IBM/Oracle-Sun for servers and, BlueArc/Isilon/Panasas for storage solutions. Privately-held Force10 is backed by numerous leading investors and has filed for an Initial Public Offering.
Force10’s products are designed to meet the needs of high-end datacenters as they consolidate to increase port density and reduce operating costs. The company provides solutions that better enable customers to virtualize their computing environment and move to cloud computing. Force10 solutions help organizations further cut costs by automating the provisioning of high-speed networks.
The company offers the two major product lines. Force10’s high-speed Ethernet equipment provides an automatic upgrade path to 40 and 100 GbE speeds. Product families include the following:
ExaScale / TeraScale E-Series — Chassis-based switch/routers, launched in April 2009.
Optimized for core deployment in large datacenters as well as end of row in high-density copper server aggregation
C-Series — Midrange chassis-based switches, with very low latency across converged networks.
Can be used for end-of-row deployment or in the core of small-to-midsize datacenters
S-Series — Small form factor fixed Layer 2 switches, can be deployed as line rate Gigabit
Ethernet and 10 GbE top-of-rack switches for datacenter, storage or compute facility
Underlying these devices is a Force10 Operating System (FTOS). This Unix-like NetBSD kernel gives a consistent operating environment across the product line.
FTOS uses a non-blocking architecture to maximize network utilization and minimize congestion under heavy traffic conditions. Force10 Networks also plans to launch an FCoE/DCB-enabled top-of-rack 10 GbE / 40 GbE solution in the second half of 2010.
Force10’s transport/access solutions include the Traverse Multiservice Transport Switch, TraverseEdge Multiservice Multiplexers, TransAccess and Wide Bank TDM Edge Multiplexers, MASTERseries Cell Site Access Aggregators, Axxius 800 Cell Site Access Aggregators, Adit 600 Converged Services Access Gateway and Broadmore Platform.
Force10 has retained positive mind share and reasonable success in Tier-one datacenter customers among portals, hosting and Web 2.0 companies. According to the company, its high
price/performance ratio is based on the high-density and high-performance architecture of their products, enabling them to deliver superior energy efficiency and a low total cost of ownership. The product lines also offer a broad framework of network automation and virtualization capabilities, demonstrating a commitment to the datacenter market. Also, the company is a strong participant in standards bodies, especially for 40 and 100 GbE.
Challenges
Force10 Networks does face challenges, however. First, it is a smaller company with relatively low mind share in the overall enterprise market. Like its larger, more established competitors, Force10 must compete for limited IT dollars and must work to be recognized as a leading player.
Similarly, the company's solutions are aimed directly at firmly entrenched networking approaches. Force10 must show that its solutions can better help enterprise IT achieve its goal of a converged infrastructure. To that end, the company must emphasize not only its technical differences, but also its strong customer base and product performance in mission critical environments.
Conclusion and Recomendations
The transition to converged IT infrastructure will play a vital role in helping IT meet the fast evolving business needs of today’s global enterprises. It will also be critical in efforts to reduce both the capital and operational costs of running datacenters and the applications/information residing in them. To meet these objectives, IT suppliers will need to deliver solutions that more tightly integrate hardware elements, provide an open operating environment, and support full orchestration of resources across the entire datacenter.
But merely delivering more capable solutions is not sufficient; IT must adjust existing product selection and management practices to take advantage of converged IT infrastructure. This requires taking a look at all assets – servers, storage and especially networking – and determining how consolidation and virtualization can be used to help the organization. Today more than ever, datacenters need a network architecture that can fluidly meet the needs for adapting to business requirements. For organizations looking to support the future datacenter network, IDC recommends the following:
Enable increased utilization of existing IT investments across server, storage, and network
Continue investments in 10 GbE as it will be the foundation of the future datacenter network
Support new dense computing platforms including blade servers.
Enable virtualization mobility across the datacenter and between datacenters is critical
Reduce the total cost of ownership of existing and future products
Critical to successful change is for IT to work with the finance department, because a shift to a converged IT infrastructure typically requires a rethinking of IT budgeting and cost allocation. In today’s economic climate, executive management plays a more direct role in the decision-making process, so it is important not to let organizational/institutional barriers stand in the way of this important datacenter effort.
A big challenge for organizations will be finding the right solutions and networking partners to help the migration to a converged infrastructure, especially as they need to link virtualized environments together, often across many locations. These partners will help a company define characteristics of hardware asset pools; integrate converged IT systems with existing IT environments and
management systems; and help retrain IT staff to support the more unified approach to provisioning and managing converged assets and services.
Enterprises will need to look for appropriate product portfolios, but most important, find vendor partners with the expertise to help them move from a static IT environment to a converged,
virtualized, dynamic IT resource. No one supplier will be able to deliver every element of a converged IT infrastructure with equal effectiveness. As such, organizations should not automatically exclude suppliers that specialize in one segment. These suppliers are often early movers of converged IT infrastructure in their own area and can – typically do – partner with others. They are, almost by definition, not beholden to traditional technology (i.e., approaches that can be a roadblock to migrating to a converged datacenter).
To the extent Force10 can successfully address the challenges described in this paper, the company should be considered by organizations looking to converge their datacenter infrastructure through virtualization.
A B O U T T H I S P U B L I C A T I O N
This publication was produced by IDC Go-to-Market Services. The opinion, analysis, and research results presented herein are drawn from more detailed research and analysis independently conducted and published by IDC, unless specific vendor sponsorship is noted. IDC Go-to-Market Services makes IDC content available in a wide range of formats for distribution by various companies. A license to distribute IDC content does not imply endorsement of or opinion about the licensee. C O P Y R I G H T A N D R E S T R I C T I O N S
Any IDC information or reference to IDC that is to be used in advertising, press releases, or promotional materials requires prior written approval from IDC. For permission requests contact the GMS information line at 508-988-7610 or [email protected]. Translation and/or localization of this document requires an additional license from IDC.
For more information on IDC visit www.idc.com. For more information on IDC GMS visit www.idc.com/gms.