Innovations in Hardware Utilization

For most businesspeople the most important issues are what the hardware enables, how it is advancing, and how rapidly it is advancing. In many industries, exploiting computer hardware is a key to achieving competitive advantage. Successful hardware exploitation comes from thoughtful consideration of the following questions:

• How do organizations keep up with the rapid price and performance advancements in hardware? For example, how often should an organization upgrade its computers and storage systems? Will upgrades increase personal and organizational productivity? How can organizations measure such increases?

• How should organizations determine the need for the new hardware infrastructures, such as server farms, virtualization, grid computing, and utility computing?

• Portable computers and advanced communications technologies have enabled employees to work from home or from anywhere. Will these new work styles benefit employees and the organization? How do organizations manage such new work styles?

To fully understand hardware, we should have an idea of current innovations in hardware. In this article I discuss how companies are using their hardware resources in innovative ways, including server farms, virtualization, grid computing, utility computing, cloud computing, edge computing, autonomic computing, and nanotechnology.

Many companies are finding that they do not have enough computer processing power to meet their needs. In particular, they are experiencing an increasing shortage of facilities needed to manage, transmit, and store the data flowing from Web-based applications. To address this problem they are building massive data centers called server farms, which contain hundreds of thousands of networked computer servers. The huge number of servers in a server farm provides redundancy and fault tolerance in case one or more servers fail. Server farms require massive amounts of electrical power, air conditioning, backup generators, security, and money. They also need to be located fairly closely to fiber optic communications links.

Virtualization

Utilization rates on servers range from 5–10 percent. That is, most of the time, organizations are using only a small percentage of their total computing capacity. One reason for this low rate is that most organizations buy a new server every time they implement a new application. CIOs tolerate this inefficiency in order to make certain that they can supply enough computing resources to users when they are needed. Also, server prices have dropped more than 80 percent in the last decade, making it easier and cheaper to buy another server than to increase the utilization of the servers the company already has. However, virtualization has changed this situation. Virtualization means that servers no longer have to be dedicated to a particular task. Server virtualization uses software-based partitions to create multiple virtual servers (called virtual machines) on a single physical server. Therefore, multiple applications can run on a single physical server, with each application running within its own software environment. Many benefits accrue to organizations using virtualization, including the following:

• A lower number of physical servers leads to cost savings in equipment, energy, space in the data center, cooling, personnel, and maintenance;

• Enhanced organizational agility, as virtualization enables organizations to quickly modify their systems to respond to changing demands;

• The focus of the information technology department can shift from the technology itself to the services that the technology can provide.

Grid Computing

Grid computing applies the unused processing resources of many geographically dispersed computers in a network to form a virtual supercomputer. Target problems are usually scientific or technical in nature and require a great number of computer processing cycles or access to large amounts of data.

Utility Computing

In utility computing, a service provider makes computing resources and infrastructure management available to a customer as needed. The provider then charges the customer for specific usage rather than a flat rate. Utility computing is also called subscription computing and on-demand computing. Utility computing enables companies to efficiently meet fluctuating demands for computing power by lowering the cost of owning hardware infrastructure. Utility computing also provides fault tolerance, redundancy, and scalability. That is, if one server fails, another takes it place. Scalability means that if an application requires additional servers, they can easily be added as they are needed.

Cloud Computing

Every year, companies spend billions of dollars on information technology infrastructure and expert staffs to build and maintain complex information systems. Software licensing, hardware integration, power and cooling, and staff training and salaries add up to a large amount of money for an infrastructure that may or may not be used to its full capacity. Enter cloud computing. In cloud computing, tasks are performed by computers physically removed from the user and accessed over a network, in particular the Internet. The cloud is composed of the computers, the software on those computers, and the network connections among those computers. The computers in the cloud are typically located in data centers, or server farms, which can be located anywhere in the world and accessed from anywhere in the world. Although some experts differentiate among the concepts, cloud computing simply incorporates the characteristics of grid computing and utility computing on a global basis. That is, the cloud supplies as many computers as are needed for a particular task (grid computing) and users pay for the amount of actual processing and storage used (utility computing). The advantages of cloud computing include much lower infrastructure costs, and the disadvantages consist of privacy, security, and reliability concerns.

Tweet