物件導向軟體之架構(Object-Oriented Software Construction)探討

[서평]「Component Software」 - Beyond Object-Oriented Programming -

A new class of models, formalisms and mechanisms has recently evolved for describing concurrent and distributed computations based on the concept of ``coordination''''. The purpose of a coordination model and associated language is to provide a means of integrating a number of possibly heterogeneous components together, by interfacing with each component in such a way that the collective set forms a single application that can execute on and take advantage of parallel and distributed systems. In this chapter we initially define and present in sufficient detail the fundamental concepts of what constitutes a coordination model or language. We then go on to classify these models and languages as either ``data-driven'''' or ``control-driven'''' (also called ``process-'''' or ``task-oriented''''). Next, the main existing coordination models and languages are described in sufficient detail to let the reader appreciate their features and put them into perspective with respect to each other. The chapter ends with a discussion comparing the various models and some conclusions.

/pdf/coordination-models-and-languages-4p4lzj3oar.pdf

Coordination models and languages

A super-peer is a node in a peer-to-peer network that operates both as a server to a set of clients, and as an equal in a network of super-peers. Super-peer networks strike a balance between the efficiency of centralized search, and the autonomy, load balancing and robustness to attacks provided by distributed search. Furthermore, they take advantage of the heterogeneity of capabilities (e.g., bandwidth, processing power) across peers, which recent studies have shown to be enormous. Hence, new and old P2P systems like KaZaA and Gnutella are adopting super-peers in their design. Despite their growing popularity, the behavior of super-peer networks is not well understood. For example, what are the potential drawbacks of super-peer networks? How can super-peers be made more reliable? How many clients should a super-peer take on to maximize efficiency? we examine super-peer networks in detail, gaming an understanding of their fundamental characteristics and performance tradeoffs. We also present practical guidelines and a general procedure for the design of an efficient super-peer network.

Designing an Super-Peer Network

All one needs to know about fog computing and related edge computing paradigms: A complete survey

We present a matchmaking system that exploits ontology-based (OWL-S) service descriptions to discover service compositions capable of satisfying a client request. Efficiency is achieved by pre-computing off-line a (hyper)graph that represents the functional dependencies among different (sub)services. The notion of Semantic Field [1] is employed to cross different ontologies.

/pdf/automated-discovery-of-compositions-of-services-described-3e4eyhnki2.pdf

Automated discovery of compositions of services described with separate ontologies

Compatibility of Linda-based Component Interfaces

The OASIS TOSCA specification aims at enhancing the portability of cloud-based applications by defining a language to describe and manage service orchestrations across heterogeneous clouds. A service template is defined as an orchestration of typed nodes, which can be instantiated by matching other service templates. In this paper, after defining the notion of exact matching between TOSCA service templates and node types, we define three other types of matching (plug-in, flexible and white-box), each permitting to ignore larger sets of non-relevant syntactic differences when type-checking service templates with respect to node types. We also describe how service templates that plug-in, flexibly or white-box match node types can be suitably adapted so as to exactly match them.

/pdf/matching-cloud-services-with-tosca-3oc8g6ubwb.pdf

Matching Cloud Services with TOSCA

Since deployment automation technologies are heterogeneous regarding their supported features and modeling languages, selecting a concrete technology is difficult and can result in a lock-in. Therefore, we presented the Essential Deployment Metamodel (EDMM) in previous work that abstracts from concrete technologies and provides a normalized metamodel for creating technology-independent deployment models. In this demonstration, we present tool support for EDMM in the form of the EDMM Modeling and Transformation System, which enables (i) creating EDMM models graphically and (ii) automatically transforming them into models supported by concrete deployment automation technologies.

The EDMM Modeling and Transformation System

Some basic notions of object-oriented programming such as objects, messages and inheritance are provided of a clean definition according to the logic programming paradigm. Objects are represented by logic theories, while inheritance mechanisms are expressed through meta-level axioms. The sending of a message is interpreted as a request for the proof of a formula. The main contribution of the paper is to provide a clean semantic characterization for these notions defining both a proof-theory and a compositional model-theory.

Antonio Brogi

Papers

Automated discovery of compositions of services described with separate ontologies

Compatibility of Linda-based Component Interfaces

Matching Cloud Services with TOSCA

The EDMM Modeling and Transformation System

Objects in a Logic Programming Framework.