Martina Maggio

Controlling software applications via resource allocation within the heartbeats framework

A formalism was recently introduced to instrument, monitor and control computer applications based on the rate of heartbeats they emit, thereby quantitatively signaling their progress toward goals. To date, the idea was however used essentially in an heuristic manner. This work first shows that a very simple dynamic heartbeat rate model can be devised, an that said model allows to address the corresponding control problems in a methodologically grounded way. A general solution is then devised, that can be realized through different actuation mechanisms, depending on which type of resource—CPU, memory, bandwidth, etc.—is constraining the application performance in the particular situation at hand. Experiments prove the efficacy of the proposed extension to the heartbeats framework, both with applications that fit the proposed model and with more complex test cases, for which said model is just a coarse approximation.

Decision making in autonomic computing systems: comparison of approaches and techniques

Autonomic computing systems adapt themselves thousands of times a second, to accomplish their goal despite changing environmental conditions and demands. The literature reports many decision mechanisms, but in most realizations a single one is applied. This paper compares some state-of-the-art decision making approaches, applied to a self-optimizing autonomic system that allocates resources to a software application providing performance feedback at run-time, via the Application Heartbeat framework. The investigated decision mechanisms range from heuristics to control theory and machine learning: results are compared by means of case studies using standard benchmarks.

Finite-precision implementation issues in narrowband active control

The domain of active control of narrowband disturbances, of either vibrational or acoustic noise source, poses several methodological and implementational issues. Besides all the problems that normally arise in digital control, some specific issues are related to the attenuation of harmonic or quasi-harmonic signals, which require the usage of control blocks approaching, or even touching, the stability limit, such as notch filters or dynamic oscillators. The finite-precision implementation of these blocks can evidence several problems related to nonlinear effects such as signal quantization, numerical representation, and approximate computation. This paper focuses on the effects related to finite-precision computations and floating point numerical representation, and, after a brief literature review, shows the necessity of extending conventional methodologies for the analysis of finite-precision controller implementation. Some illustrative examples are also provided to illustrate the implementation problems discussed.

Model-Based PI(D) Autotuning

This chapter deals with Model-Based AutoTuning (MBAT). Besides describing the main flow and phases of such a tuning procedure, the main advantages of the underlying methodological approach are evidenced, namely the availability of a parametric process model to simulate the loop and assess its behaviour and characteristics such as performance and robustness, the possibility of stipulating the control specifications with respect to the said model, and tuning parameters’ interpretability. In the light of such an organisation, some MBAT methods are presented, and a minimal taxonomy of them is suggested. The chapter also highlights the major problems connected with MBAT, i.e., the interplay between the model parameterisation procedure and the tuning results—for which technological limitations typically prevent the use of results from the identification for control domain—and the limitations posed by the parametric approach, essentially to the representation of model error and uncertainty in view of a robustness analysis. Some possible future developments and perspectives are finally outlined.

Function Inlining in Modelica Models

The equation-based Modelica language allows the modeller to specify custom functions. The body of a function is an algorithm that contains procedural code to be executed when the function is called. This language feature is useful for many applications; however, the insertion of a function often prevent model optimizations that require the model to be formulated in purely declarative form by equations only. This paper discusses several non-trivial cases in which the function call and the corresponding algorithmic code can be transformed into an equivalent purely equation-based model, thus allowing further optimization. The inlining algorithms presented in the paper go well beyond the state of the art in commercial and open-source Modelica tools. (Less)

Control and Design of Computing Systems: What to Model and How

The application of feedback control to computing systems is a promising research area, but has to date been hindered by the almost unanimously perceived complexity in creating control-oriented system models. Computing systems are in fact considered very hard to describe with dynamic models allowing for simple and powerful control design tools, so that complex ones need bringing in to the detriment of efficiency and result assessment. In this work a novel approach to the modelling of computing systems is proposed, in a view to explain and partially avoid such complexity, by capturing only their relevant dynamics with the simplest possible models. The approach is shown to work at least on two relevant case studies, so that a significant generality can be inferred from it, being able to reproduce the relevant parts of the systems behaviour and paving the way to control design and synthesis. (Less)

Teaching to write control code

Abstract Regulators are conceived as dynamic systems, but almost invariantly implemented as computer code. In the opinion of the authors the relationships between those two worlds are frequently overlooked in control education, although an insufficient knowledge of the involved concepts can hinder the correct operation of the realised systems. This manuscript presents a didactic activity that aims at closing the mentioned gap. The scope is for now limited modulating control, although extensions can be envisaged.

Enhancing feedback process scheduling via a predictive control approach

Abstract In some recent papers it was shown that preemptive process schedulers in multitasking operating systems can be viewed, and above all designed , as discrete-time feedback controls with very simple (I- or PI-type) regulators, yielding significant advantages over classical scheduling policies as for time complexity and parameter interpretability. In this work, the same problem is tackled with a predictive control approach. Doing so allows to release some simplification hypotheses of the mentioned papers and to improve the achieved solutions’ flexibility, widening the application possibilities at an affordable additional cost.

A PI-based control structure as an operating system scheduler

Abstract Many functions of operating systems are keen to be realised as feedback controllers. Doing so has a non negligible design impact, but also a significant payoff in terms of simplicity and generality. This paper presents a complete operating system scheduler, at present implemented in a microcontroller kernel, entirely composed of a PI-based control structure. The proposed scheduler is experimented with in several load conditions. In all of them, it performs in a comparable manner with respect to the classical (i.e., not control-based) policy optimised for that condition, as long as design assumptions such as schedulability are met. In addition, if some off-design situation is encountered, the proposed control-based scheduler definitely outperforms those not conceived as controllers.

A byte of systems theory

This paper provides the barely essential system-theoretical foundations. Basically, its goal is to introduce the notion of dynamic system and one of its simplest specialisations, namely the discrete-time linear time-invariant case. To overcome such a problem, the choice was made to provide just the most important ideas in a manner deliberately abstracted from their utilisation, with just some proof sketches when relevant, and having the sole purpose of maintaining a consistent treatise.