The final set of controls associated with the avoidance of systems failures is related to technology. That is, organizations often make additional investments in their IT infrastructure with the explicit goal of avoiding or at least minimizing the consequences of information system failure.
There are numerous technical controls associated with information security. In this section, we introduce a range of infrastructure investments intended to improve the overall reliability of infrastructure and consequently, the reliability of organizational IT-enabled information services. These include:
- Redundant critical components (equipment, communications, etc.)
- Power conditioning and backup power
- System backup capabilities
- Network and system monitoring tools
Organization requiring high levels of availability will often find themselves needing to buy redundant hardware. In simplest terms, redundancy occurs when you buy to pieces of hardware to perform a function when one device is capable of performing the function adequately. Normally this is considered a bad thing to do. However, if the possibility exists of that component failing and disrupting a critical information service, then purchasing and installing the second device may be quite reasonable. The basic philosophy of operational redundancy is best illustrated with some simple probability calculations.
For example, assume an important device, say a hard disk used for storing data, has I device has a 1% probability of failing and disrupting service in any given year. The organization must determine if it is willing to accept the possibility of data loss (the data added since the last time the disk was backed up) and service disruption. If the 1% probability seems too high, the organization can purchase a second drive which we also assume has a 1% annual probability of failing. Now, what is the probability of the organization experiencing the data loss and service failure? Assuming drive failures are independent events, the probability of both drives failing simultaneously is: .01% or 1 in a thousand. This represents a considerable improvement over the 1 in 100 chance of experiencing a single drive failure. In complex environments, the probability calculations can become quite complex but the general logic remains the same. Investing in redundancy decreases the probability of systems failure.
Management must consider all potential causes of failure in prioritizing its investments in technical controls. In areas where power stability is a problem, power conditioning and the provisioning of on-site generators to provide backup power are critically important. Computer equipment is not just susceptible to power outages. Voltage fluctuations, drops as well as surges, can damage expensive equipment. Many organizations find it prudent to purchase uninterruptible power supplies (UPSs), essentially large batteries with regulators, to condition the power to avoid such damage. Where prolonged power outages occur, or where even short disruptions of service cannot be tolerated, organizations will find it necessary to provide backup generators to ensure that at least the most critical systems remain operating. As with all technical investments, the cost of systems varies greatly.
We have already discussed the importance of backing up system data and will not elaborate much beyond that here. With respect to technical investments, organizational managers must recognize that purchasing some type of backup system is probably essential. However, there is a wide range of backup systems. As indicated above, managers will want to assess the consequences of data loss and service disruption in determining how sophisticated, really meaning how expensive, a backup system is required.
Likewise with systems and network management applications, we have addressed the usage of these tools in the discussion of systems monitoring ans incident response. A wide range of tools exist and the technology in this area is continuously improving. Early versions of these tools were largely limited the reporting of component failures. Later generations of system management tools started to provide more information about the health and status of the IS components they were designed to monitor. The latest generation of tools has begun to more directly monitor the health and performance of actual information services rather than monitoring the technical components which comprise the services. By monitoring services rather than technical components, these tools monitor what organizations are really interested in, service quality. The most advanced systems and network applications are even capable of conducting rudimentary forms of causal analysis such that when service outages or performance degradations are detected, the applications are able to combine service and component analysis to determine the specific causes of the observed problem.
It is unrealistic to expect that organizational managers to be involved in making the technical choices required in the acquisition of technical controls. However, management does need to understand that such technical choices are required if the organization's IT infrastructure is going to support the quality of service delivery that the organization requires. Better managers will be able to engage their IT staff in a conversation in which the relationship between recommended technical controls and service quality can be explained.
- 1485 reads
