Introduction to Control Theory And Its Application to Computing Systems

Abstract

Feedback control is central to managing computing systems and data networks. Unfortunately, computing practitioners typically approach the design of feedback control in an ad hoc manner. Control theory provides a systematic approach to designing feedback loops that are stable in that they avoid wild oscillations, accurate in that they achieve objectives such as target response times for service level management, and settle quickly to their steady state values. This paper provides an introduction to control theory for computing practitioners with an emphasis on applications in the areas of database systems, real-time systems, virtualized servers, and power management.

A control theoretical view of cloud elasticity: taxonomy, survey and challenges

“…A recent development in this regard is the performance evaluation framework proposed in [112]. However, this framework is generic and there is a stronger need for specific control-theoretical benchmarks that have the ability to facilitate analysis of control solution specific characteristics, e.g., SASO (Stability, Accuracy, Short settling and Overshoot) properties as discussed in [2]. (viii) Computational overhead analysis most of the control methodologies are adaptive in nature as they have the ability to dynamically adapt to the changing environments.…”

“…In such cases, we found the following two possibilities: (1) in the case of horizontal elasticity, the common objective is to maintain the CPU utilisation level of overall cluster (e.g., [50][51][52]). (2) Whereas, in the case of vertical elasticity, the objective depends on the nature of reconfigurable resources. For example, there are control solutions, where the control objective is the readjustment of the memory allocation (such as [53,54]) or the readjustment of the CPU allocation (such as [49,[55][56][57][58]).…”

“…Such feedback controllers are designed to be stable in order to avoid oscillation and settle quickly to the steady state by appropriately responding to disturbances. They are better for achieving service level objectives, such as response time or throughput [2]. The use of control theory is not limited to the advent of cloud computing or elasticity.…”

A convex optimization-based approach to control of uncertain execution platforms

“…This has successfully been done by e.g. Tarek Abdelzaher, who treats it in several publications, including [1]. This work is relevant for the problem studied in this paper, but more focused on computer server farms than the embedded space.…”

Towards the deep, knowledge-based interoperability of applications

“…Examples of usage of the internal modeling approach (quality management, risk management, business process management) (Moen et al 2006;Brache 2002) (Mareels et al 1996) and computing systems (Abdelzaher et al 2008).…”