P. Dasgupta and R. J. LeBlanc. The Clouds Distributed Operating System. IEEE Computer, 24(11):34--44, November 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....ping pong effect by allowing a processor to pin a page down for a ffi period of time. During such a ffi period, the page may not be invalidated or migrated to another processor. The value of ffi is defined at application startup and remains constant for the duration of the application. Clouds [DLAR91] uses application defined segments as the sharing memory unit. With segments, shared data may reside in memory units small enough so that false sharing can be eliminated. In addition, the ping pong effect that accompanies false sharing is eliminated as well. To reduce the ping pong effect that ....

Partha Dasgupta, Richard LeBlanc, Mustaque Ahamad, and Umakishore Ramachandran. The Clouds distributed operating system. In IEEE Computer, Apr 1991.

....complicated task of instrumenting the OS with the necessary hooks to do reconfiguration on a case by case basis. K42 is not the first operating system to use an object oriented design. Object oriented designs have helped with organization [27, 47] extensibility [11] reflection [54] persistence [15], and decentralization [1, 51] In addition, K42 s method of detecting a quiescent state is not unique. Sequent s NUMAQ used a similar mechanism for detecting quiescent state [37] and recently, SuSE Linux 7.3 has integrated a mechanism for detecting quiescence in kernel modules [36] 7.1 Open ....

P. Dasgupta, R. J. LeBlanc, Jr, M. Ahamad, and U. Ramachandran. The Clouds distributed operating system. IEEE Computer, 24(11):34--44, November 1991.

....Each step raised the level of abstraction and increased programmer productivity. Operating systems, on the other hand, remained largely unaffected by this process. Although there have been attempts to build object oriented or object based operating systems (Spring [27] Choices [10] Clouds [17]) and many operating systems internally use object oriented concepts, such as vnodes [31] there is a growing divergence between application programming and operating system programming. To close this semantic gap between the applications and the OS interface a large market of middleware ....

P. Dasgupta, R. J. LeBlanc, M. Ahamad, and U. Ramachandran. The Clouds distributed operating system. In IEEE Computer, 24(11), pp. 34-44, Nov. 1991.

....to continue even when the locus is blocked. Once this step is taken there is no need to maintain a time stamp for a locus. The only proviso is that a locus must always have a host container, and therefore invocation must be an atomic operation. This atomicity is similar to that used by Clouds [4]. Since invocation is the only communication mechanism provided by Grasshopper, once invocation becomes atomic, it no longer makes sense for a consistent cut to cross a message send. This further simplifies the notion of a consistent cut. A container s vector time is updated whenever it is ....

Dasgupta, P., LeBlanc, R. J. and Appelbe, W. F. "The Clouds Distributed Operating System", Proceedings, 8th International Conference on Distributed Computing Systems, 1988.

....data from accidental or malicious misuse. In persistent systems this is typically provided via the programming language type system [29] through data encapsulation [26] using capabilities [11] or by a combination of these techniques. To date, most persistent systems, with a few exceptions [6, 9, 34], have been constructed above conventional operating systems. Implementors of persistent languages are invariably forced to construct an abstract machine above the operating system, since the components of a persistent system are different in nature to the components of a conventional operating ....

Dasgupta, p., LeBlanc, R., Mustaque, A. and Umakishore, R. "The Clouds Distributed Operating System", Arizona State University, Technical Report 88/25, 1988.

....error prone. Over the last decade several object oriented operating systems have been developed. This might mean the kernel was written in an object oriented fashion (perhaps using C ) as with Choices [2] that the OS was tailoured speci cally to suit object oriented applications (e.g. Clouds [6]) or that the OS used an orthogonal object model in user and kernel space as in Spring [20] During the last few years, several component based operating systems have materialised [3, 11] These abandon the tradition of a kernel supporting processes for a nucleus managing many user components. ....

P. Dasgupta, R. LeBlanc, M. Ahamad, and U. Ramachandran. The clouds distributed operating system. IEEE Computer 24, November 1992.

....and protection at a fine grain. Because every invocation requires crossing a domain boundary, performance has been a problem, both in hardware [16, 11] and software based implementations [17] Clouds is a distributed operating system that is, like Lipto, based on the object thread model [5]. However, a CloutIs object is persistent and resides in its own protection domain. Consequently, Clouds objects are heavyweight and they do not support fine grained decomposi tion. Choices is an object oriented operating system [4] It has a fine grained, modular structure based on the ....

P. Dasgupta, R. J. LeBlanc Jr., M. Ahamad, and U. Ramachandran. The Clouds distributed operating system. IEEE Computer. To appear.

....implemented using logging [11] or some shadowing technique [26] If persistent systems are to be anything other than research vehicles, they must be both stable and resilient. The persistent systems which exhibit these properties and that have been constructed to date, with a few exceptions [19, 33], have been constructed on top of traditional operating systems. Existing operating systems do not provide an ideal platform for the development of persistent systems. This is not surprising since this was never part of their design goals. Indeed, most operating systems have files as their only ....

Dasgupta, P., LeBlanc, R. J. and Appelbe, W. F. "The Clouds Distributed Operating System", Proceedings, 8th International Conference on Distributed Computing Systems, 1988.

....data from accidental or malicious misuse. In persistent systems this is typically provided via the programming language type system [27] through data encapsulation [25] using capabilities [12] or by a combination of these techniques. To date, most persistent systems, with a few exceptions [7, 10, 32], have been constructed above conventional operating systems. Implementors of persistent languages are invariably forced to construct an abstract machine above the operating system, since the components of a persistent system are different in nature to the components of a conventional operating ....

Dasgupta, p., LeBlanc, R., Mustaque, A. and Umakishore, R. "The Clouds Distributed Operating System", Technical Report Thesis, 88/25, 1988.

....currently existing dialogues with the object involved will be broken with a migration exception. Users performing invocations over these dialogues will thus detect that a nontransparent migration has occurred. 7. Related Work Many existing systems address the issues of object mobility [5, 8, 11]. A representative example is the Hermes system [3] which uses a variety of techniques to improve the efficiency of object invocation in the presence of location changes. A base technique is the use of forwarding addresses, which allow for transparent rerouting of invocation requests. Such ....

P. Dasgupta, R.J. LeBlanc, M. Ahamad, and U. Ramachandran, "The Clouds Distributed Operating System " IEEE Computer, Vol.24 (11) , pp. 34-44 (November, 1991).

....through the distributed shared memory. A parallel program running on a parallel machine can be easily converted into a distributed application running over networked workstations with distributed shared memory. Notable systems that built software distributed shared memory include IVY[45] Clouds [19], Munin [14, 9] Midway [10] TreadMarks [1] 93 and Quarks [42] High cost of distributed synchronization and lack of fault tolerant support are the disadvantage of these systems. A system with similar DSM concepts but with a di#erent approach is Linda[13] It provides a globally shared space ....

....or without failures. Other type of systems address fault tolerance separately and provide as an 97 add on feature. Fault tolerant techniques includes check pointing, replication, and migration. Systems that provides fault tolerance features with these techniques are CIRCUS[17] LOCUS[47] Clouds[19], Fail safe PVM [44] PLinda [32] and FT Linda [5] These systems often provide fault tolerance features independent to other system functions and require user intervention when failures are present. Calypso [6] Chime [51] and our system, belong to a di#erent group in which load balancing and ....

P. Dasgupta, R. J. LeBlanc Jr., M. Ahamad, and U. Ramachandran. The Clouds distributed operating system. IEEE Computer, 1990.

....[24] Mirage uses the time window coherence protocol to throttle thrashing that is caused when multiple nodes concurrently access the same page. It keeps pages resident on nodes for some minimumamount of time, ensuring that nodes accomplish some work before the page is transferred. Clouds [18] and Orca [5] both provide an object based distributed shared memory. Because they manage data in user level abstractions rather than OS pages, opportunities for optimizations are available. For example, the system could send all of the object, which might span multiple OS pages, to a node when it ....

Partha Dasgupta, Richard J. LeBlanc Jr., Mustaque Ahmad, and Umakishore Ramachandran. The Clouds distributed operating system. Computer, 24(11):34--44, November 1991.

....tuple space accesses. 2.8.2 Programming Languages and Systems Supporting Transactions There have been research efforts to develop programming languages and systems to use transactions as the foundation for constructing distributed applications. They are Argus[47] Avalon[26] Camelot[26] Clouds[21] and TABS[27] Argus is a programming language and system to support the implementation and execution of distributed applications such as mail systems and inventory control systems[47] The principal mechanism of Argus is guardians which are a special kind of abstract objects. Guardians ....

....sequence of separate transactions. Thus, processes can be easily designed to communicate with each other. Implementation of the runtime system is also relatively simpler because of the flat transaction model. Checkpoint protected tuple space makes transaction commits efficient. Camelot[26] Clouds[21] and TABS[27] provide distributed transaction facilities as a set of user libraries or operating system features. Like Avalon and Argus, they are aimed at distributed applications and have similar drawbacks as Avalon and Argus, when they are used for parallel computation. 2.9 Summary In this ....

[Article contains additional citation context not shown here]

P. Dasgupta, R.J. LeBlanc, M. Ahamad, and U. Ramachandran. The Clouds distributed operating system. IEEE Computer, pages 34--44, November 1991.

....and mobility. The type conformity rules adopted for Guide, as well as the overall design of the run time structures for method selection, are similar to those of Emerald. However, Emerald does not provide support for persistent objects, which is a major goal of Guide. The main goal of Clouds [10] is to develop a distributed, fault tolerant computing environment (operating system and applications) which appears as a uniform, integrated computing resource to the users. The system is based on the notions of passive objects, location independent invocations, and nested actions. All objects ....

....separate objects from execution structures, i.e. define passive objects executed by independently defined processes. We did not find strong arguments in favor of either solution. Both have been adopted in existing object based systems (e.g. active objects in Emerald [4] passive objects in Clouds [10], Amoeba [12] and SOS [16] The two solutions are dual: both can offer the same functionality to the user, but in different ways. The choice is mostly influenced by considerations of efficiency and adequacy to the hardware and to the applications. Our system is intended to be a collection of ....

Leblanc, R.J. and Appelbe, W.F., "The Clouds Distributed Operating System," Proc. 8th Int. Conf. on Distributed Computing Systems, pp. 2-9, San Jose, Calif., June 1988.

....a comprehensive review is well beyond the scope of this paper. We are interested in a system in which there is a single, network wide, object name space spanning machine boundaries. We may identify two main approaches for object support. In the first approach (used, for example by Clouds [Dasgupta 91] Orca [Bal 90] Emerald [Jul 88] individual objects are directly supported by the operating system (as in Clouds) or the run time system (as in Orca or Emerald) The system provides a location services that allows to find an object when a call is made to one of its methods. The second ....

P. Dasgupta, R. J. LeBlanc Jr, M. Ahamad, and U. Ramachandran, The Clouds distributed operating system, IEEE Computer, 24,11 (nov. 1991), pp 34-44 5

....Clouds to Sun 4, Sun 386i and Sequent multiprocessor architectures and to design and implement more system objects. Contact: P. Dasgupta, School of Information and Computer Science, Georgia Institute of Technology, Atlanta GA. 30332 References: 106] 107] 108] 109] 110] 111] 112] [113], 114] 115] 116] 117] 118] 2.13 Cosmos Main Goal Cosmos has been designed at the University of Lancaster, U. K. Cosmos gives integral support for both distribution and programming environment functions. Location and access transparency, and eventually replication and synchronization ....

P. Dasgupta, R.J. LeBlanc, and W.F. Appelbe, "The Clouds Distributed Operating System", In Proc. 8th Int. Conf. on Distributed Computing Systems, pages 1--9, San Jose, Ca., June 1988.

....file and database and web services (notably from Sun, Tandem, IBM and Microsoft) The promise of distributed computing for generalpurpose computations was a major thrust in the development of Distributed Operating Systems in the mid 1980s. Operating systems such as Amoeba [27] Mach [21] Clouds [7], Chorus [22] and so on failed to bring the power of distributed general purpose computing to the desktop. In retrospect, the shortcoming of these systems resulted from the application development barrier. The application development barrier is the void of applications for a new platform. If we ....

Dasgupta, P., LeBlanc Jr., R. J., Ahamad, M. and Ramachandran, The Clouds Distributed Operating System. In IEEE Computer, Nov. 1991.

No context found.

P. Dasgupta and R. J. LeBlanc. The Clouds Distributed Operating System. IEEE Computer, 24(11):34--44, November 1991.

No context found.

P. Dasgupta, R. LeBlanc, M. Ahamad, and U. Ramachandran. The Clouds Distributed Operating System. IEEE Computer 24, November 1992.

No context found.

P. Dasgupta, R. J. LeBlanc, Jr., and W. F. Appelbe. The Clouds Distributed Operating System. In The 8th International Conference on Distributed Computer Systems. IEEE Computer Society Press, 1988.

No context found.

P. Dasgupta, R. J. LeBlanc, Jr, M. Ahamad, and U. Ramachandran. The Clouds distributed operating system. IEEE Computer, 24(11):34--44, November 1991.

No context found.

P. Dasgupta, R.J. LeBlanc, M. Ahamed, and U. Ramachandran. The Clouds Distributed Operating System. IEEE Computer, 24(11):34--44, November 1991.

No context found.

P. Dasgupta, R. J. LeBlanc, M. Ahamad and U. Ramachandran, `The Clouds distributed operating system', IEEE Computer, 24, (11) 34--44 (1991).

No context found.

P. Dasgupta, R.J. LeBlanc, M. Ahamad y U. Ramachandran. "The Clouds Distributed Operating System". IEEE Computer, 24(11). 1991. Pg. 34-44.

No context found.

Dasgupta P., Le Blanc R. J. Jr., Appelbe W. F., Ramachandran U.: the Clouds Distributed Operating System, IEEE Computer, November 1991

First 50 documents  Next 50

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