Towards Real-Time Fault-Tolerant CORBA Middleware

Gokhālé, Anirüddhā; Natarajan, Balachandran; Schmidt, Douglas C.; Cross, Joseph K.

doi:10.1023/b:clus.0000039493.73008.13

Cited by 32 publications

(19 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Receiving the right answer too late becomes the wrong answer (Gokhale 2004). Run time error and exception handling in the development phase; typically require a different set of tasks to be initiated when an error occurs (Westley 2004).…”

Section: Modelling Agent Real-time Constraintsmentioning

confidence: 99%

Towards Modelling Real Time Constraints

Ashamalla¹,

Beydoun²,

Low³

et al. 2012

Proceedings of the 7th International Conference on Software Paradigm Trends

View full text Add to dashboard Cite

Software agents are highly autonomous, situated and interactive software components. They autonomously sense their environment and respond accordingly. Agents behaviours are often constrained by by real time constraints such as the time in which the agent is expected to respond .i.e. time needed for a task to complete. Failing to meet such a constraint can result in a task being not achieved. This possibly causes an agent or a system to fail, depending on how critical the task is to the agent or system as a whole. Our research aims at identifying and modelling real time constraints in the early phase of analysis which helps in creating a more reliable and robust system. Abstract:Software agents are highly autonomous, situated and interactive software components. They autonomously sense their environment and respond accordingly. Agents behaviours are often constrained by by real time constraints such as the time in which the agent is expected to respond .i.e. time needed for a task to complete. Failing to meet such a constraint can result in a task being not achieved. This possibly causes an agent or a system to fail, depending on how critical the task is to the agent or system as a whole. Our research aims at identifying and modelling real time constraints in the early phase of analysis which helps in creating a more reliable and robust system.

show abstract

Section: Modelling Agent Real-time Constraintsmentioning

confidence: 99%

Towards Modelling Real Time Constraints

Ashamalla¹,

Beydoun²,

Low³

et al. 2012

Proceedings of the 7th International Conference on Software Paradigm Trends

View full text Add to dashboard Cite

show abstract

“…Similar to the COLD PASSIVE replication style used in FaultTolerant CORBA (FT-CORBA) [15], all subtasks are assumed to be stateless (except the connections between subsequent subtasks which are maintained by the middleware) so that the overhead of active state synchronization is avoided. The migration mechanism works as follows: Each subtask can have a primary instance and a few backup instances on the processors where it has the required resource.…”

Section: Middleware Implementationmentioning

confidence: 99%

Towards Controllable Distributed Real-Time Systems with Feasible Utilization Control

Wang

Chen

et al. 2009

IEEE Trans. Comput.

View full text Add to dashboard Cite

Abstract-Feedback control techniques have recently been applied to a variety of real-time systems. However, a fundamental issue that was left out is guaranteeing system controllability and the feasibility of applying feedback control to such systems. No control algorithms can effectively control a system which itself is uncontrollable or infeasible. In this paper, we use the multiprocessor utilization control problem as a representative example to study the controllability and feasibility of distributed real-time systems. We prove that controllability and feasibility of a system depend crucially on end-to-end task allocations. We then present algorithms for deploying end-to-end tasks to ensure that the system is controllable and utilization control is feasible for the system. Furthermore, we develop runtime algorithms to maintain controllability and feasibility by reallocating tasks dynamically in response to workload variations, such as task terminations and migrations caused by processor failures. We implement our algorithms in a robust real-time middleware system and report empirical results on an experimental test-bed. We also evaluate the performance of our approach in large systems using numerical experiments. Our results demonstrate that the proposed task allocation algorithms improve the robustness of feedback control in distributed real-time systems.

show abstract

“…We look at the two important characteristics of distributed stream processing applications, namely the fact that they are composite and strict, to understand their availability requirements that differentiate them from other distributed applications, and to guide our placement decisions. In many distributed applications such as distributed storage [2,4,17,30] or client/server computing [6,12,14,18,20,25,36] that focus on the availability of individual objects, such as files or processes, an increase in the number of replicas usually implies a similar increase in the availability of the application. While we expect that increasing the number of component replicas will also increase the availability of a distributed stream processing application composed of them, we find that this increase greatly depends on the relative placement of the individual component replicas.…”

Section: Maximizing Application Availabilitymentioning

confidence: 99%

“…Replication for high availability and fault tolerance has been studied from different perspectives in a variety of domains, including distributed databases [13,22,24,34], distributed object systems [11,12,14,18,21,25,36], and web services [6,20]. While many aspects of replication have been studied extensively, and solutions such as active [31] and passive [9] replication are widely accepted, in this work we focus on replica placement for maximizing the availability of distributed stream processing applications.…”

Section: Introductionmentioning

confidence: 99%

“…Furthermore, the scale of distributed stream processing applications, both in terms of data volume and rate, as well as in terms of numbers of components, affects significantly the placement decisions. For example, not all primary replicas can be hosted by the same server, as might be the case with object-, component-, or service-based architectures, such as CORBA [11,12,14,21], Enterprise JavaBeans [36], or multi-tier architectures [13,22]. Our placement mechanisms however can be applied to such systems, if their scale requires the primary replicas of a composite application's components to be distributed and significant amounts of data transferred between them make the relative placement of components important.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Replica placement for high availability in distributed stream processing systems

Repantis

Kalogeraki

2008

Proceedings of the Second International Conference on Distributed Event-Based Systems

View full text Add to dashboard Cite

A significant number of emerging on-line data analysis applications require the processing of data streams, large amounts of data that get updated continuously, to generate outputs of interest or to identify meaningful events. Example domains include network traffic management, stock price monitoring, customized e-commerce websites, and analysis of sensor data. In this paper we look at the problem of high availability in such a distributed stream processing system. By taking into account the particular characteristics of stream processing applications we first identify design principles for a replica placement algorithm for high availability. We incorporate these principles in a decentralized replica placement protocol that aims to maximize availability, while respecting resource constraints, and making performance-aware placement decisions. We have integrated our replica placement protocol in Synergy, our distributed stream processing middleware. Our experimental comparison over PlanetLab with the current state of the art corroborates our claims that our techniques maximize availability while sustaining good performance.

show abstract

Towards Real-Time Fault-Tolerant CORBA Middleware

Abstract: An increasing number of applications are being developed using distributed object computing (DOC)

Cited by 32 publications

References 44 publications

Towards Modelling Real Time Constraints

Towards Modelling Real Time Constraints

Towards Controllable Distributed Real-Time Systems with Feasible Utilization Control

Replica placement for high availability in distributed stream processing systems

Contact Info

Product

Resources

About