Workload in Cloud Computing
Here use the term workload to refer to the utilization of IT resources on which an application is hosted. Workload is the consequence of users accessing the application or jobs that need to be handled automatically. Workload becomes imminent in different forms, depending on the type of IT resource for which it is measured: servers may experience processing load, storage offerings may be assigned larger or smaller amounts of data to store or may have to handle queries on that data. Communication IT resources, such as networking hardware or messaging systems may experience different data or message traffic. In scope of the abstract workload patterns, we merely assume this utilization to be measurable in some form [14].
Static Workload: IT resources with an equal utilization over time experience static workload.Static workloads are characterized by a more-or-less flat utilization profile over time within certain boundaries. This means that there is normally no explicit necessity to add or remove processing power, memory or bandwidth for change in workload reasons. When provisioning for such a workflow the necessary IT resources can be provisioned for this static load plus a certain overprovisioning rate to deal with the minimal variances in the workload. There is a relatively low cost overhead for this minimal overprovisioning.
Periodic Workload: IT resources with a peaking utilization at reoccurring time intervals experience periodic workload.In our real-lives periodic tasks and routines are very common. For example, monthly paychecks, monthly telephone bills, yearly car checkups, weekly status reports, or the daily use of public transport during rush-hour, all these tasks and routines occur in well-defined intervals. They are also characterized by the fact that a lot of people perform them at the same intervals. As a lot of the business processes supporting these tasks and routines are supported by IT systems today, there is a lot of periodic utilization that occurs on these supporting IT systems.
Once-in-a-Lifetime Workload:IT resources with an equal utilization over time disturbed by a strong peak occurring only once experience once-in-a-lifetime workload.As a special case of periodic workload, the peaks of periodic utilization can occur only once in a very long timeframe. Often, this peak is known in advance as it correlates to a certain event or task. Even though this means that the challenge to acquire the needed resources does not arise frequently, it can be even more severe. The discrepancy between the regularly required number of IT resources and those required during the rare peak is commonly greater than for periodic workloads. This discrepancy makes long term investments in IT resources to handle this onetime peak very inefficient. However, due to the severe difference between the regularly required IT resources and those required for the one-time peak, the demand can often not be handled at all without increasing IT resources.
Unpredictable Workload: IT resources with a random and unforeseeable utilization over time experience unpredictable workload.Random workloads are a generalization of periodic workloads as they require elasticity but are not predictable. Such workloads occur quite often in the real world. For example, sudden increases of Website accesses due to weather phenomena or shopping-sprees when new products gain an unforeseen attention and public interest. The resulting occurrence of peaks or at least their height and duration often cannot be foreseen in advance under these conditions [14].
Read More