A New Dimension in IT Infrastructure Management: Integrated KVM and Serial Console Control Systems

-- "White paper published by Raritan International”

Companies are shifting to Web-centric computing to support their extended enterprises. This shift has led to the phenomenon of data center sprawl-- the proliferation of large, multiplatform, heterogeneous IT infrastructures with geographically distributed applications and human resources. IDC believes this sprawl will grow in a number of dimensions:

In 2003, 5.2 million servers shipped worldwide; in 2008, server unit shipments are expected to increase to more than 9 million!

Much of this growth is due to the introduction of blades into the marketplace. In 2003, 184,694 blades were shipped; in 2008, unit shipments of blades are expected to increase to more than 3 million. !

Along with the increase in the number of server shipments, there will be a corresponding increase in the shipment of peripheral devices such as networking switches and routers. In 2003, 103 million LAN switch ports were shipped; in 2008, shipments of LAN switch ports are expected to increase to more than 170 million. Managing all of these servers and peripherals has become even more of a challenge as issues of security, reliability, and serviceability have grown in importance for IT managers and CIOs. These issues will only increase over time.

ENTERPRISE NETWORK COSTS AND COMPLEXITY

As an enterprise's IT infrastructure increases in size, so does its complexity, even to the point where there are negative impacts on economies of scale. Despite this situation, CIOs and IT managers are expected to improve efficiencies and productivity with a budget that is the same as, or less than, the previous year's. IT executives are faced with reducing staff; operation, administration, and maintenance (OA&M) costs; physical space allocations for servers and peripheral devices; electricity costs; HVAC costs; and cabling requirements. Achieving all of this with limited resources, while making sure the network is always up, robust, and secure, is a significant challenge. The average cost associated with server hardware (average sales price) has decreased over the past decade, due to competition and increased use of industrystandard components, but overall IT costs have not declined. In fact, system administration costs as a percentage of total data center costs have continued to rise .

IDC's research into total cost of ownership (TCO) shows that much of the after purchase costs that are associated with server platforms are driven by inefficient use of management resources, downtime and its associated costs, and diminished productivity of management staff and end users (see Figure 2).

Not only has data center and management complexity increased, but the cost of system downtime has also increased as companies depend more and more on their IT equipment to run their businesses. Airlines lose tens of thousands of dollars in new reservation revenue for every minute their reservation systems are down. Companies that host Internet auctions lose millions of dollars every day when transactions cannot occur due to a failed server farm or network. For these companies and others, revenue depends on servers being up and running at maximum efficiency. Many CIOs now consider email to be a mission-critical computing workload due to lost employee productivity during email outages.

Reductions in easily identified data center costs, as well as cost reductions associated with better management and reduced downtime that improve productivity and efficiency, fall directly to a firm's bottom line. Because of these factors, reducing mean time to repair (MTTR) is an important component in data center management. The ability to monitor alarms, troubleshoot, and repair in real time is of utmost importance.

Read the full white paper here.