HP NetServer LP 2000r 8-way Systems Enable Enterprise Applications - Page 2

FXVWRPHUýXVDJHýJXLGH +3ý 1HW6HUYHUý /;Uýåèíí A Closer Look Industry Trend: Server consolidation is an important trend that impacts the data center in several ways. Applications running on less powerful previous generation servers can be combined onto a single new server with several times more processing power. Consolidation of a server environment can occur in one of several ways: logical, physical or recentralization. Logical consolidation attempts to normalize or standardize aspects of the server environment, such as management or backup procedures, and in so doing, achieve a reduction in staff workloads. In a physical consolidation, the servers are re-located into a centralized data center to be "racked and stacked". The immediate benefits are improved physical security, lowered system management complexity across the servers and a more efficient sharing of peripherals. The third type of server consolidation, re-centralization, involves collapsing the processing loads from several different servers onto a single, larger server (or servers). This approach is used to reduce the number of dedicated servers or to respond to changing business flows. Re-centralization achieves the greatest return on investment (ROI) benefit and is most suitable for 8-way server configurations. All three methods require that companies employ a consolidation approach that encompasses technology, process and people in order to achieve the benefits (reduced staff workloads, improved manageability and floor space utilization) attributed to server consolidation. Data Center Needs The data center, as a repository of information assets, is an integral component of the enterprise infrastructure. It generally consists of mission critical, application-specific data stored in large databases and on file servers. Current technology trends such as network computing, the Internet and massive data growth are dramatically increasing the demands on this environment. Companies must pay attention to not only their hardware and software choices but also service and support options to dependably deliver the information critical to today's business needs. Selecting the appropriate server equipment that resides at the center of this information fabric is critical. Data centers have three core needs for this server equipment: performance (data processing capacity), high availability and scalability. Performance for the data center is the ability to process and transfer large amounts of data quickly within the various server sub-systems, also expressed as transactions completed per unit of time. Each of the sub-systems in a server must be optimized as well as balanced, so that the system delivers the maximum performance and no one component becomes a bottleneck. Performance also includes the ability of the server system to accommodate peak business periods and occasional spikes in the workload. The more a server is sized for growth, the more likely it will have sufficient headroom and be able to accommodate usage peaks. (See Scalability below.) Availability is a measure of the ability of the server system to deliver constant computing services to clients, also expressed as a percentage of time that the system is "available". Data center equipment must be highly available (approaching 100% uptime). If an application stops, revenue generation can come to a halt. The design features of reliability, component redundancy, pro-active manageability, multi-node fail-over clustering capabilities and the appropriate levels of service and support combine to produce highly available systems. 2