A Formal Model for Reasoning about Adaptive QoS-Enabled Middleware Nalini Venkatasubramanian 1 , Carolyn Talcott 2 , and Gul Agha 3 1 Univ. of California Irvine, Irvine, CA 92697-3425, USA, nalini@ics.uci.edu 2 Stanford University, Stanford, CA 94305, USA, clt@cs.stanford.edu 3 University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA, agha@cs.uiuc.edu Abstract. Systems that provide QoS-enabled services such as mul- timedia are subject to constant evolution - customizable middleware is required to effectively manage this change. Middleware services for resource management such as scheduling, protocols providing security and reliability, load balancing and stream synchronization, execute concurrently with each other and with application activities and can therefore potentially interfere with each other. To ensure cost-effective QoS in distributed systems, safe composability of resource management services is essential. In this paper we present a meta-architectural framework for customizable QoS-based middleware based on the actor model of concurrent active objects. Using TLAM, a semantic model for specifying and reasoning about components of open distributed systems, we show how a QoS brokerage service can be used to coordi- nate multimedia resource management services in a safe, flexible and efficient manner. In particular, we show that a system in which the multimedia actor behaviors satisfy the specified requirements, provides the required multimedia service. The behavior specification leaves open the possibility of a variety of algorithms for resource management as well as adding additional resource management activities by providing constraints to ensure their non-interference. Keywords: meta-object models, distributed systems, theoretical foun- dations, object-oriented applications, multimedia 1 Introduction In the coming years, QoS-enabled distributed servers will be deployed to deliver a variety of interactive services. Applications such as telemedicine, distance learn- ing and electronic commerce exhibit varying requirements such as timeliness, security, reliability and availability. The set of servers, clients, user requirements, network and system conditions, in a wide area infrastructure are changing con- tinuously. Future applications will require dynamic invocation and revocation of services distributed in the network without violating QoS constraints of ongoing J.N. Oliveira and P. Zave (Eds.): FME 2001, LNCS 2021, pp. 197–221, 2001. c  Springer-Verlag Berlin Heidelberg 2001