An Approach to Self-adaptive Software Based on Supervisory Control

Abstract

Abstract:Self-adaptive software systems use observations of their own behavior, and that of their environment, to select and enact adaptations in accordance with some objective(s). This adaptation is a higher-level system function that performs optimizations, manages faults, or otherwise supports achieving an objective via changes in the running system. In this paper, we show how this capability can be realized using techniques found in hierarchical control systems, and we discuss interrelated issues of stability, assurance, and implementation.

Siblingrivalry

“…We compare starting from a random configuration ("w/o offline") to configurations found through offline training on the same and a different architecture. In the dynamic autotuning space, there have been a number of systems developed [7,9,12,22,23,25,30] that focus on creating applications that can monitor and automatically tune themselves to optimize a particular objective. Many of these systems employ a control systems based autotuner that operates on a linear model of the application being tuned.…”

“…In response to some of these challenges, there is a growing body of work [7,9,12,22,23,25] focused on creating applications that can monitor and automatically tune themselves to optimize a particular objective (e.g. meeting response time goals by trading quality of service (QoS) for increased performance or lower power usage).…”

A Quality-Driven Approach to Enable Decision-Making in Self-Adaptive Software

“…Compressing data [110] Saving bandwidth by transceiving compressed data Tuning [93] Adjusting parameters to meet some adaptation goals (i.e., buffer size and delay time)…”

“…Although closed-loop is the most widely used model for control-based self-adaptive software, adaptive and reconfigurable models are also recommended for several reasons, including large-range dynamic disturbances [101]. On the other hand, considering the discrete nature of software systems, one of the appropriate control-based approaches for self-adaptive software is supervisory control of discrete event system (DES) [168]; see for example [208] and [93].…”

Intelligent Adaptive Monitoring for Cardiac Surveillance

“…In addition, we have proposed a supervised method to learn such temporal patterns automatically from annotated temporal data. In [11], a self-adaptive software is introduced to modify the processing chain according to some predefined events such as sensor loss. But this approach does not deal with high-level goals as temporal reasoning which is crucial for monitoring evolving systems.…”