Extreme makeover, the new data center edition

11 Jun Extreme makeover, the new data center edition

Many companies are looking at ways to cut IT costs while maintaining or improving service levels to their end-users. One not-so-obvious area to look is data center network technologies.
Centralized data centers are in vogue again, with many of the technologies migrating back to a single environment. File servers, print servers, and mission-critical applications are all being evaluated as potential consolidation efforts. This mass migration of technologies is causing a new set of challenges for the data center manager – namely, how best to virtualize and optimize available resources.
Common threads
One way to do so is to use the common thread between every data center, branch office, and remote office as a tool to gain a competitive advantage. By looking at the network proactively, organizations can gain productivity and have access to resources like never before. Adding more bandwidth, incorporating optimization technologies, and creating service-oriented network architecture can help transform the data center for better, more cost-effective performance.
Data center switching is no longer just basic routing and switching. Modular, top-of-rack and blade switches are available to support security, load balancing, and application acceleration via blades or line cards — all integrated into the chassis. This architecture provides scalability, resilience and operational management throughout your data center networking environment. Your switch solution should support low latency, line-rate throughput, and advanced security features.
Storage Area Networking is a cost-effective way to store and manage the business critical information across your enterprise. Intelligent Fabric-based storage switches and directors typically have support for advanced storage services like virtualization, server-less backup, data replication, and continuous data protection to allow for enhanced business continuance and data migration. Within this architecture, one can have IP-based blades creating a unique mixture of real and virtualized storage direction. Storage services modules can also be inserted into data center-class switch chassis for space and power concerns.
Information Security in a virtualized data center is more important than ever. Data centers house the most critical and sensitive resources of any organization and consequently, the opportunities to target these critical data center networks, servers, and databases have increased exponentially. New and emerging regulatory requirements, such a HIPAA, Sarbanes-Oxley, Basel II, and PCI, place a special emphasis on protecting the access to, transmission, and storage of sensitive information, such as the personal and financial information of customers and employees.
Just as IP telephony virtualized the need for PBXs to be situated in every physical office location, the next generation network components is enabling companies to view their data center equipment as true virtualized resources.
Application network services
Consider consolidating branch office server and storage resources to the data center to lower your overall costs and increase data security while providing LAN speed application access and performance. Further, with the use of these tools, you can optimize WAN bandwidth usage and distribute requests over multiple devices to help scale applications and to ensure uptime for those critical applications. Bandwidth optimization tools should be added where appropriate. These services and solutions can be obtained within a blade/line card in a chassis-type design or in some cases via an appliance.
Data center virtualization
Data center managers can build their own utility computing data center out of industry-standard servers, storage, applications, and management tools. Virtualization can help reduce IT expenses and provide advantages such as leveraging shared pools of resources, virtualized server and I/O infrastructure and the ability to respond to new business demands by swiftly redeploying servers. Furthermore, it provides the ability to add automated fail-over from a generic pool of resources and create unified 10 Gbps fabric for IP, storage area network, and server-to-server communications. This all adds up to leveraging existing investments with the appropriate tools to increase your data center ROI.
High performance computing applications
These solve complex, computationally intensive problems and are widely deployed within enterprises, as they deliver significant business benefits. A key enabler for the broad adoption of HPC applications is the practice of clustering multiple industry-standard servers using a high-speed network to provide supercomputer performance at a fraction of the cost of traditional supercomputers. This can be done with technologies such as InfiniBand Compute Fabric Switches and Gateways as well as Gigabit Ethernet. Coupling these technologies within the data center ties together the Management, I/O and Inter-Process Communications (IPC) networks.
Tying it All Together
The network can and should be looked upon as the new operating system inside the data center walls. Add to that IP-based storage and server switching technologies and you now have a data center that is better equipped to draw on resources when you need them no matter if they are in the next aisle of servers or in the data center across the country.
Previous Forsythe articles
• James Geis/Steven Harris: The details, dollars, and sense of virtual data centers
• When reacting to data loss, communication is the key
Related articles
• CDW Berbee “goes green” with new data center
• New data center opens for business in Eau Claire
• Data center sticker shock? Companies may be in for cost surprise
• Pitching virtualization: Benefits go far beyond cost cutting