Special Issues of IEEE Transactions on Systems, Man, and Cybernetics, Part A

On the Theme:

Self-organization in Distributed Systems Engineering




Franco Zambonelli

Dip. di Scienze e Metodi dell’Ingegneria

Università di Modena e Reggio Emilia

Via Allegri 13, 42100 Reggio Emilia, ITALY

E-mail: franco.zambonelli@unimore.it

Home Page: www.dismi.unimore.it


Omer F. Rana

School of Computer Science

Cardiff University

5 The Parade, Cardiff CF24 3XF, Wales, UK

E-mail: o.f.rana@cs.cardiff.ac.uk

Home Page: www.cs.cf.ac.uk




Submissions of Papers:          June 1, 2004

Notifications:                         September 1, 2004

Revised Submission:              October 1, 2004

Final Notification:                  November 1, 2004

Publication Date:                    March 2005



A variety of novel distributed computing scenarios are emerging, enabled by recent advances on microelectronics, communication and information technologies. At one extreme, a variety of networked micro-scale computer systems (e.g., sensor networks and distributed MEMS) will soon populate and pervade our everyday environments. At the other extreme, worldwide distributed computing (e.g., global P2P networks and Grid computing) is eventually becoming a reality. In between these two extremes, various scenarios of use are arising, including personal-area networks, MANETs, and networked furniture.

Despite their macro dissimilarities, all of these networked systems shares similar characteristics and introduce very similar issues with regard to programming and management, challenging traditional approaches to distributed systems engineering. First, all of these networks are intrinsically dynamic with regard to both their topology and the presence of ephemeral nodes. Second, the dynamics of the network make impossible for application components to rely on strong a priori information about their execution context (i.e., the other application components with which to interact, their spatial location, the characteristics of the local environment in which they are dived). Third, the intrinsic decentralization of all these networks and the possibly high-number of distributed components make very difficult for engineers to enforce a strict micro-level control over all application components.

The above characteristics call for novel approaches to distributed systems engineering, in which applications and systems must be engineered so as to: (i) have them adaptively self-configure their execution parameters depending on the current characteristics of the operational environment; (ii) have them survive the unpredictable dynamics of the operational environment by preserving specific structural properties and quality levels. Said concisely: in future scenarios, software systems shall be able to self-organize their activities, leaving humans mostly out-of-the-loop. The role of software engineers will mainly be that of ensuring that a designed systems will correctly evolve according to specific applications goals, and possibly that of enforcing some sort of decentralized dynamic control to direct/correct its spontaneous evolution.

It is worth emphasizing that industry too has realized the importance of self-organization and decentralized management approaches (cfr., e.g., the “Autonomic Computing” project at IBM). In particular, it has been realized that the increasing complexity and dynamics of distributed systems scenarios makes traditional approaches – in which engineers are called to directly control and understand each component of a systems and the effects of their interactions – both technically hard and economically unbearable.



In the context of the above scenario, this special issue welcomes original paper that focus on aspects of self-organization that may have some impact (or give useful insights) on the activities of engineering (i.e., designing, developing, maintaining and controlling) next generation distributed systems. Papers that discuss self-organization in very general and purely speculative terms, or that simply report on observation/simulation of peculiar self-organization phenomena in large distributed systems without discussing the potential impact for distributed systems engineering, or that proposes new approaches without supporting them with appropriate validation (simulations and/or performance results), are of limited interest.



Topics of interest (non exhaustive list):

·       Direct engineering of robust adaptive behaviors based on self-organization.

·       Reverse engineering of self-organization in biological, physical, and social systems, and its application to distributed systems engineering.

·       Programming paradigms and middleware for self-organization.

·       Modeling, validation, and testing, of self-organizing distributed systems.

·       Comparative studies (e.g., quantitative comparison of non-self-organization vs. self-organization approaches, comparison among different types of self-organization approaches).

·       Control (not simply observation!) of complex emergent behaviors in distributed systems.

·       Innovative applications based on self-organization (e.g., autonomous distributed control system, sensor networks, ambient intelligence, multiagent systems, computational economies, computational self-assembly, intelligent fabrics, etc.)

·       Migrating from existing approaches to self-organization principles


Format of Submissions

Perspective authors should submit their papers in PDF format by sending them as E-mail attachment to both the guest editors. Each paper should include the full name and affiliation of each of the authors and of the contact author, an abstract, and list of keywords. Papers should not exceed 25 double-spaced pages in length, and the main text should be 11 pt minimum for fonts, on A4 or letter page size with reasonable margins (2 cm).


Papers submitted to the issue must not have been published before and must not be under consideration in another journal or conference. Papers previously appeared in conference or workshop proceedings may be acceptable, provided that they notably extend upon the previous version.

Please ensure that papers do not contain embedded fonts (such as embedded Asian fonts). These papers can be difficult to read and view with many PDF viewers not equipped to display certain character sets.



The guest editors will strictly overview the process and will ensure its fairness and scientific rigour. Each paper will be reviewed by three independent experts. Only papers having been judged of very high quality will be selected for the second phase. In the second phase, authors will be asked to revise their papers according to the comments received, and the reviewers will be asked to check if the required revisions have been correctly implemented. The editors will reserve the right to reject papers even in this second phase.