K. Marzullo, R. Cooper, M. D. Wood, and K. P. Birman. Tools for Distributed Application Management. IEEE Computer, 24:42-51, August 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....based on flow B, representing the access to the context sensors , and flow C representing the information flowing from the sensors to the application. Provision of a flow of control has not been widely exploited by mobile adaptive systems. Some distributed systems platforms, such as ISIS META [19], do provide such a flow, but have not been adapted for mobile environments or context aware application development. One more recent and notable exception is the Puppeteer system [5] which provides support for mobile environments through control of non adaptive applications by the platform. ....

....triggered to reduce the use of bandwidth in order to allow the e mail client to download a large e mail within the timeliness requirements specified by the user. 6. Influential Work In this section we briefly discuss related systems that have been directly influential on our work. ISIS META [19] provides an infrastructure for reactive management of distributed systems. It introduces the concepts of sensors and actuators that provide information about the operation of applications and allow control by external entities. The functionality of these elements can by viewed as similar to the ....

K. Marzullo, R. Cooper, M. Wood, and K. Birman. Tools for distributed application management. IEEE Computer, 24(8), 1991.

....or device provides the contextual information to the application and the application is responsible for adapting to the change of the context. According to our knowledge the only middleware platform that provides a flow of control from the platform to the application, i.e. flow D, is ISIS META [17] though it should be noted that only a small amount of research has been undertaken into making ISIS operate in a mobile environment [4] In ISIS META the platform is able to explicitly request a specific behaviour from the application through the use of call back functions called actuators. Flows ....

Marzullo K., R. Cooper, M. Wood and K. Birman. "Tools for Distributed Application Management". IEEE Computer, 24(8), pp 42-51, 1991.

....to the application. According to our knowledge no platform supporting context aware adaptation provides a flow of control from the platform to the application. Indeed, although examples of systems providing this type of flow can be found in the distributed systems community, e.g. ISIS META [15], it should be noted that these systems consider only network triggered adaptation. 4 Architectural Requirements The previous sections have described the limitations of current approaches for supporting adaptive mobile applications. In particular, these approaches do not provide appropriate ....

Marzullo K., R. Cooper, M. Wood, K. Birman.: Tools for Distributed Application Management. In: IEEE Computer, 24(8) (1991) 42-51

....to accommodate varying runtime circumstances impacting on real time performance. In CHAOS [10] real time systems are adapted with the use of an entity relation database modeling system structure. Control systems ideas have been used in distributed application management as well. Meta [16] is an architecture and a tool that uses a non hierarchical control system to optimize performance in fault tolerant distributed systems using Isis. Distributed application management (e.g. 2] 21] employs services supporting the dynamic management of distributed applications. Network ....

K. Marzullo, R. Cooper, M. D. Wood, and K. P. Birman, "Tools for Distributed Application Management", IEEE Computer, August 1991, pp. 42-51.

....analyze the global effects of local filtering policies is an important area of future research. Such tools would help administrators understand the global effects of MLF partitioning, for example. Other policy specification languages include those for the management of distributed applications [29] and for distributed trust management [30] The former allows an application manager to specify reaction rules for distributed real time applications. These rules are compiled into enforcement code that is then distributed to the application components. Such distribution allows the components to ....

K. Marzullo, R. Cooper, M. Wood, and K. Birman, "Tools for Distributed Application Management," IEEE Computer, August, 1991, pp. 42--51.

.... HiFi active monitoring enables the agents to react spontaneously (e.g. corrective actions) which improves the management operations response time compared with humanin the loop model [21] A number of monitoring approaches and systems for monitoring distributed systems have been proposed (e.g. [5, 11, 15, 17, 21, 22, 23]) Although some of these systems provide mechanisms for modifying the monitoring requests dynamically, these mechanisms are man2 ual (i.e. they require human intervention) and insufficient to support a programmable or self directed monitoring tasks (actions) as described in this paper. In ....

....The ERS is a HiFi library linked with the monitored application during compilation to facilitate event reporting. The main function of the instrumentation component is to facilitate the process of inserting the monitoring instruction (sensors) inside the program code. In many monitoring systems [17, 20], programmers write a considerable size of code for each generated event. This makes the instrumentation task tedious and error prone. In HiFi, users only insert the user sense that specifies just the event name (e.g. ReportEvent(WarningEvent) The instrumentation component then pre processes ....

[Article contains additional citation context not shown here]

K. Marzullo, R. Cooper, M. D. Wood, and K. P. Birman. Tools for distributed Application Management. IEEE Computer, 24(8):42--51, August 1991.

....computing such as distributed synchronization, resilient computation, logging and recovery. With the various tools in ISIS and the well ordered environment that ISIS provides, writing dynamic and fault tolerant distributed programs can be as simple as writing programs for a single central machine [MWC90]. Note that ISIS is basically a big machine technology. It runs well on UNIX workstations and mainframes, whereas PC s are essentially ignored in ISIS. The ISIS system addresses a broader and more complete distributed environment with job management and monitoring than other solutions such as ....

K. Marzullo, M. Wood, and R. Cooper. Tools for distributed application management. Technical Report 90-1136, Cornell University, Department of Computer Sciences, July 1990.

....the reliable and efficient execution of distributed programs, and the availability of essential services or processes in spite of node or communication failures. Some work has been done in implementing specific tools for monitoring and controlling distributed applications, such as the Meta Toolkit[12], the Megascope tool[15] within the Project Pilgrim[14] and the tools developed by Huang and Kintala[7] However, until now, we have not seen any system supporting the availability management of DCE based applications. Sampa, which stands for System for Availability Management of Process based ....

....(panel) for collecting and querying cell specific monitored data. Huang and Kintala [7] have also implemented a nice set of tools for controlling availability and checkpointing, but their work is not concerned with describing global availability specifications. Another tool the Meta Toolkit [12], which is based on the ISIS system [3] It provides means for instrumenting distributed application processes with sensors and actuators, which are used for monitoring and controlling the execution of the application processes from a control layer, where monitoring configuration is specified in ....

K. Marzullo, R. Cooper, M.D. Wood, and K.P. Birman. Tools for Distributed Application Management. IEEE Computer, 24(8):42--51, August 1991.

....specifications of the actions to be taken when such events occur. Most of the programming environments use periodic heartbeat messages for status monitoring. The group membership service in Isis supports facilities to monitor group membership and site failures and to trigger recoveries. Isis Meta [10] uses rule based specifications for configuration management. PVM and Express also support detection of process crashes, but do not support explicit mechanisms for reconfiguration. Both V system and Mach provide kernel level support for exception handling. There are no mechanisms available to the ....

Keith Marzullo, Robert Cooper, Mark Wood, and Kenneth Birman. Tools for distributed application management. IEEE Computer, pages 42--51, August 1991.

....environments, not only must such systems perform correctly, but they must also operate with high performance. Much of the previous work in distributed computing has addressed the design of large scale systems that function correctly, in spite of correctness faults of individual components [18, 49, 82, 86]. However, there has been little development of techniques to tolerate performance faults unexpected performance fluctuations from the components that comprise the system. Due to this shortcoming, many systems are overly sensitive to performance variations, in that global performance is high if ....

....account how global performance characteristics will be altered under component performance variations. Fortunately, much of the previous work in the field of distributed computing has addressed the design of large scale systems that can tolerate such correctness faults in individual components [18, 24, 49, 61, 82, 86, 107, 114]. The practical notion behind such work is that distributed and parallel systems consist of both hardware and software components that will periodically fail; a system that works continuously on top of such unreliable components must be designed to operate in spite of such failures. A good example ....

Keith Marzullo, Robert Cooper, Mark Wood, and Kenneth Birman. Tools for Distributed Application Management. IEEE Computer, pages 42--51, August 1991.

.... has triggered actions which are similar to our obligation policies but these databases are not distributed, and the triggered actions are defined statically and cannot be changed without re compilation [12] Lomita is a rule based language for programming the management layer in the Meta system [13]. Lomita rules are of the form on condition do action , which is also similar to Jonscher s triggered actions, but there is no explicit subject or target they are defined implicitly. This paper has presented a notation which can be used to express management policies for distributed systems ....

Marzullo, K., Cooper, R., Wood, M., Birman, K. (1991) Tools for Distributed Application Management, IEEE Computer 24 (8): 42--51.

....1992) has triggered actions which are similar to our obligation policies but these databases are not distributed, and the triggered actions are defined statically and cannot be changed without re compilation. Lomita is a rule based language for programming the managementlayer in the Meta system (Marzullo, Cooper, Wood and Birman, 1991). Lomita rules are of the form on condition do action , whichisalso similar to Jonscher s triggered actions, but there is no explicit subject or target they are defined implicitly. This paper has presented a notation which can be used to express management policies for distributed systems as ....

Marzullo, K., Cooper, R., Wood, M. and Birman, K. (1991). Tools for distributed application management, IEEE Computer 24(8): 42--51.

.... HiFi active monitoring enables the agents react spontaneously (e.g. corrective actions) which improves the management operations response time compared with human in the loop model [13] A number of monitoring approaches and systems for monitoring distributed systems have been proposed (e.g. [5, 9 11, 13 15]) Although some of these systems provide mechanisms for modifying the monitoring requests dynamically, these mechanisms are manual and insufficient to support a programmable or self directed monitoring tasks (actions) as described in this paper. In addition, they do not support a scalable and ....

Marzullo, K., Cooper, R., Wood, M.D., Birman, K.P.: Tools for distributed Application Management. IEEE Computer, Vol. 24. 8 (1991) 42--51

....tasks. Increased flexibility is achieved through a declarative and comprehensible monitoring language, simple code instrumentation process, and automated monitoring agents administration. A number of monitoring approaches and systems for monitoring distributed systems have been proposed (e.g. [5, 6, 7, 8, 10, 11, 12, 13, 14]) Although some of these systems have various design goals and objectives, they are insufficient to support a scalable and fine grain event filtering mechanism for monitoring large scale distributed systems such as Internet based applications. In addition, the monitoring filtering tasks in such ....

K. Marzullo, R. Cooper, M. D. Wood, and K. P. Birman. Tools for distributed Application Management. IEEE Computer, 24(8):42--51, August 1991.

....without interfering with the running application, and specifying programmable management tasks that can automatically initiate new monitoring operations based on the events detection. A number of monitoring approaches and systems for monitoring distributed systems have been proposed (e.g. [4, 10, 11, 13, 16, 17, 18, 19, 21, 22, 24]. Although some of these systems have various design goals and objectives, they are insufficient to support a scalable and fine grain event filtering mechanism for monitoring large scale distributed systems such as Internet based applications. In addition, the monitoring filtering tasks in such ....

....The ERC is then used by the Event Reporting Stub (ERS) for constructing and reporting the event notification. The main function of the instrumentation component is to facilitate the process of inserting the monitoring instruction or sensors inside the program code. In many monitoring systems [17, 19], programmers write a considerable size of code for each generated event. This makes the instrumentation task tedious and error prone. In HiFi, users only inserts the user sense that just specifies the event name (e.g. ReportEvent(WarnEvent) The instrumentation component then pre processes the ....

K. Marzullo, R. Cooper, M. D. Wood, and K. P. Birman. Tools for distributed Application Management. IEEE Computer, 24(8):42--51, August 1991.

....services, and for analysis over different time scales. The management system also has to provide automatic reactions of various kinds to maintain services and service quality. Compared to network and system management, there has been little research on distributed application management (e.g. [2, 3, 4, 32]) The distinguishing difficulty of this level of management is that applications are, by their nature, more specialised, more distinctive, and more heterogeneous, than network devices or file services. Management tends to be developed specifically for one application or one domain. ....

....configuration management, fault management and performance management. 4.3 Monitoring and Control of Distributed Applications Some work has been done in implementing specific tools for monitoring and controlling distributed applications. Some of these tools are described below. ffl Meta Toolkit [32] This toolkit is a system for managing distributed applications developed using the Isis distributed programming toolkit [5] ffl Huang and Kintala Tools [17] This set of tools provides services for detecting whether a process is alive or dead; specifying and checkpointing critical data; ....

K. Marzullo, R. Cooper, M. D. Wood, and K. P. Birman. Tools for Distributed Application Management. IEEE Computer, 25(8):42--51, August 1991.

.... application and thereby minimizes the monitoring overhead (intrusiveness) A survey of monitoring and event filtering related work can be found in [1, 2] We classified the monitoring distributed systems related work into three classes: hardware monitoring such as [9] software monitoring such as [5, 11], and hybrid monitoring [8] An example of event filtering mechanisms in communication protocols is [3] in distributed systems is [4] and in active databases is [6] The primary motivations for our work are (1) the necessity of employing an efficient monitoring technique for LSD systems to ....

....LSD systems. In other words, as an implication of our abstraction model, we monitor LSD application through detecting events (primitive or composite) occur during the execution which encapsulate the program run time behavior. This model has the following advantages over the querybased (time driven [11]) monitoring approach: 1) it enables the consumers (e.g. managers) to control granularity of the monitoring process by limiting the monitoring operations on observing specific system components and or interesting events, 2) it provides a faster response to applications problems such as ....

K. Marzullo, R. Cooper, M. D. Wood, and K. P. Birman. Tools for distributed Application Management. IEEE Computer, 24(8):42--51, August 1991.

....distributed as it is in the current implementation of Darwin, then global event sequencing is necessary to provide consistent views of the state of the executing processes. One promising necessary to provide consistent views of the state of the executing processes. One promising approach is Meta [MAR90] which provides for sensors and actuators to be embedded in an approach is Meta [MAR90] which provides for sensors and actuators to be embedded in an application program for monitoring and control by a control program. For consistency of application program for monitoring and control by a ....

....is necessary to provide consistent views of the state of the executing processes. One promising necessary to provide consistent views of the state of the executing processes. One promising approach is Meta [MAR90] which provides for sensors and actuators to be embedded in an approach is Meta [MAR90] which provides for sensors and actuators to be embedded in an application program for monitoring and control by a control program. For consistency of application program for monitoring and control by a control program. For consistency of view, Meta relies on the virtual synchrony facilities ....

Marzullo,K., Cooper,R., Wood,M. and Birman,K. (1990) "Tools for Distributed Application Management", Report TR 90-1136, Dept. of Computer Science, Cornell University, Ithica NY 14853.

....In other words, as an implication of our abstraction model, we monitor LDM application through detecting events (primitive or composite) that occur during the execution which encapsulate the program run time behavior. This model has the following advantages over the query based or time driven (Marzullo et al. 1991) monitoring approach: 1) it enables the consumers (e.g. managers) to control granularity of the mon Filter : FILTER = Filter Body Filter Body : Event Expr ] Filter Expr ] Action ] Filter Name . Event Expr : Event Name Event Rel Event Expr ) j Event Name ....

Marzullo, K., Cooper, R., Wood, M. and Birman, K. (1991) Tools for distributed Application Management. IEEE Computer, Vol. 24, number 8, pp. 42-51, August 1991.

....a graphical interface to the Meta system. This type of work is needed in order to make the Meta system more useful to people concerned with application management. Along those same lines, work is under way here at Cornell to develop a high level language for specifying application behavior [MCWB91] Rules expressed in this language would be translated into Meta guarded commands. In addition to building tools and applications on top of Meta, much remains to be done within the toolkit itself. We identify in this section a number of areas requiring additional research. The Meta toolkit ....

Keith Marzullo, Robert Cooper, Mark Wood, and Kenneth P. Birman. Tools for distributed application management. Computer, 24(8):42--51, August 1991.

....aggregate into a set of local conditions that can be evaluated close to the monitored objects, where polling is comparatively inexpensive and can be done frequently. For purposes of this text we will not cover their work in detail, but interested readers will find discussion of these topics in [Woo91, Mar90, MCWB91, BM93]. 23.3.2 Actuators An actuator is the converse of a sensor: the management system assigns a value to it, and this causes some action to be taken. Actuators may be physical (for example a controller for a robot arm) logical (a parameter of a software system) or may be connection to abstract ....

Keith Marzullo, Robert Cooper, Mark Wood and Kenneth P. Birman. Tools for Distributed Application Management. IEEE Computer. August 1991.

No context found.

K. Marzullo, R. Cooper, M. D. Wood, and K. P. Birman. Tools for Distributed Application Management. IEEE Computer, 24:42-51, August 1991.

No context found.

M. Wood, K. Marzullo, Tools for Distributed Application Management, Proceedings of the Spring 1991.

No context found.

K. Marzullo, R. Cooper, M. D. Wood, and K. P. Birman, Tools for Distributed Application Management, IEEE Computer, (August 1991), 42-51.

No context found.

K. Marzullo, R. Cooper, M. Wood, and K. Birman, "Tools for distributed Application Management, " IEEE Computer, Vol. 24, pp. 42-51, August 1991.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC