Objects in large distributed applications (OLDA-II)

The OLDA-II workshop was the successor to the OLDA workshop held as part of OOPSLA '91 in Phoenix, Arizona. The primary goal of both of these workshops was to bring together academics, industrial researchers, and developers interested in large-scale distributed object-based applications and the support they require. In contrast to many other workshops, which focus on a particular topic in some detail, the OLDA workshop deliberately provides a vertical slice through the problem domain. That is, one of the principal goals of the workshop is to bring together developers and researchers from a variety of communities: application builders: software engineering methodologists; systems builders; and researchers investigating very low-level issues. In addition, this year one focus of the workshop was on language issues, with a panel of language developers. The reason for doing this is twofold , firstly in distributed systems low-level architectural issues can profoundly influence application design and high-level issues such as fault-tolerance. Secondly, there is a clear need for integration of the many aspects of systems into coherent wholes: much current research is on single issues, yet the solutions proposed often have consequences for related problems and solutions. The workshop attracted 45 participants in all, from ten different countries, representing a range of universities, research laboratories and industrial Languages: the Next Generation, joined us for half of the lunchtime panel session, described below. In addition to the five panelists, fifteen presentations were given, all but one of which were ten minutes long. With hindsight, ten minutes was not quite long enough, the fifteen minute slots of the first OLDA workshop provided more scope for detail to be presented; however the brief presentations and substantial allowance of time for discussion (almost half of each session) generated a lively and informal atmosphere. In the next two sections the topics presented and discussed are briefly summarised. I apologise if I have misquoted or misrepresented anybody, in this brief overview of a long and tiring day! The morning and afternoon were both divided into two sessions, with each of the four following a very loose theme. The two morning sessions were dedicated to papers on individual systems issues: basic problems being addressed in the first session; resource allocation and load balancing being the focus of the second. Obviously, whole workshops can be, and are, dedicated to the topics addressed, so the papers selected were chosen to highlight a single interesting feature, or …