Response times in N-user replicated, centralized, and proximity-based hybrid collaboration architectures

Abstract

We evaluate response times, in N-user collaborations, of the popular centralized (client-server) and replicated (peer-to-peer) architectures, and a hybrid architecture in which each replica serves a cluster of nearby clients. Our work consists of definitions of aspects of these architectures that have previously been unspecified but must be resolved for the analysis, a formal evaluation model, and a set of experiments. The experiments are used to define the parameters of and validate the formal analysis. In addition, they compare the performances, under the three architectures, of existing data-centric, logic-centric, and stateless shared components. We show that under realistic conditions, a small number of users, high intra-cluster network delays, and large output processing and transmission costs favor the replicated architecture, large input size favors the centralized architecture, high inter-cluster network delays favor the hybrid architecture, and high input processing and transmission costs, low think times, asymmetric processing powers, and logic-intensive applications favor both the centralized and hybrid architectures. We use our validated formal model to make useful predictions about the performance of the three kinds of architectures under realistic scenarios we could not create in lab experiments.

Securing the Deployment of Cloud-Hosted Services for Guaranteeing Multitenancy Isolation

“…The architectures or cloud patterns used to deploy cloud-hosted services to the cloud are of great importance to software architects because they determine whether or not the system's essential quality attributes (e.g., performance) will be exhibited [1,8,9].…”

“…Architectural and design patterns have long been used to provide known solutions to many common problems a distributed system face [1,10]. A system/ application architecture decides whether or not it will show its necessary quality attributes (e.g., performance, availability, and security) [1,8].…”

“…Our definition of cloud deployment pattern is similar to the concept of design patterns [10], (architectural) deployment patterns [1], collaboration architectures [8], cloud computing patterns [11], cloud architecture patterns [12], and cloud design patterns [13].…”

Beyond the lan

“…These surveys introduce other techniques such as dead-reckoning that have been used in games and distributed simulations, and also mention some of the network techniques seen here (aggregation and compression, although the details are different to those of the game libraries). Finally, a core of groupware researchers is also looking at performance issues, including performance aspects of groupware distribution architectures [14,16], scalability [13], QoS [8,9], generalized compression [10], and interface techniques for hiding delay [11,12]. This work offers a valuable complement to the present study; although the performance models are often based on abstractions of the real network environment, and thus do not have sufficient detail to capture the issues addressed in our study, prior research provides a high-level context of toolkits and distribution architectures in which our techniques can be used.…”

Research on Late Join in Collaborative Pattern Design

“…By using the replication architecture [4,5], the system has multiple copies of application data at various distribution sites, which enables the system have high reliability and scalability. However, the biggest problem which is brought about by using this structure is that the mechanisms adopted to maintain and ensure the consistency of the distributed data will be more complicated.…”