B. Koch et al. Cache coherency and storage management in a persistent object system. In S.Zdonik A.Dearle, G.Shaw, editor, The Fourth International Workshop on Persistent Object Systems, 1990.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....programming in BETA. This section briefly discusses a general proxy based implementation of distributed BETA. The implementation of distributed BETA is based on the notion of proxy objects [2] One possible alternative would be to use a single large (e.g. 64 bit) distributed shared memory [6, 9]. However, this possibility is ruled out by the CSCW applications needing support for heterogeneous wide area networks. In the proxy approach, remote references are, as illustrated in figure 9, handled by actually referencing a local representative of the remote object, the proxy. It is then the ....

B. Koch et al. Cache coherency and storage management in a persistent object system. In S.Zdonik A.Dearle, G.Shaw, editor, The Fourth International Workshop on Persistent Object Systems, 1990.

....level. In addition programmers could also control the placement of computations for performance reasons. Others have attempted to implement transparent distribution at the persistent store level [LJGS90, GADV92, Gru92, LDS92] or even to build a distributed persistent operating system itself [KSD 90, DdBF 94] Transparent distribution is a step towards more scalable architectures and offers a simple programming model. However, transparent distribution cannot be made to scale indefinitely [MdSAB96] The main reason is that the many inter store references needed to support transparent ....

.... based on remote references to the object being shared [DC93] is highly dependent on the availability of the store in which the object resides, and thus precludes autonomous stores, availability and scalability [MdSAB96] Either cache coherency of copies is maintained requiring update propagation [KSD 90] or remote references denote objects requiring all operations to refer to the original store [Wai88] We propose instead the concept of persistent spaces in which publishers release objects that can later be fetched by any number of subscribers. The idea is better described using an example (see ....

B. Koch, T. Schunke, A. Dearle, F. Vaughan, C. Marlin, R. Fazakerley, and C. Barter. Cache coherency and storage management in a persistent object system. In Dearle et al. [DSZ90].

....have found very few systems extending DSM with persistence in the literature. One of them is Casper [121] It takes advantage of the external pager mechanism [65] of the Mach operating system [1] to provide transparent access to data in secondary storage. Given that its GC specification is sketchy [70], it is not clear how PBR could be supported. Feeley et. al [49] describe a transactional DSM that supports fine grained distributed data sharing in cooperative applications. However, there is no support for PBR. 3.3 Larchant Model The main goal of this description of the Larchant model is to ....

Bett Koch, Tracy Schunke, Alan Dearle, Francis Vaughan, Chris Marlin, Ruth Fazakerley, and Chris Barter. Cache coherency and storage management in a persistent object system. In Proceedings of the Fourth International Workshop on Persistent Object Systems, pages 99--109, Martha's Vineyard, MA (USA), September 1990.

....the demands of these new domains. OODBs support new application domains with extensible type systems that allow new abstractions to be added easily. All OODBs use some distinct service that provides stable storage for objects or pages, referred to as object servers and page servers respectively [11, 5, 19, 13, 1]. Whenever an OODB attempts to model a domain for which specialized secondary storage structures exist, the object store causes a problem. Concurrency control, recovery and other modules need to be changed in order to accommodate the new storage structure. An extensible object store would ....

B. Koch et al. Cache coherency and storage management in a persistent object system. In The Fourth International Workshop on Persistent Object Systems, 1990.

....stress existing database technology to the limit [1] Object Oriented Databases (OODBs) were developed to meet the demands of these new domains. All OODBs use some distinct service that provides stable storage for objects or pages, referred to as object servers and page servers respectively [9, 4, 12, 6, 7]. OODBs support new application domains with extensible type systems that allow new abstractions to be added easily. This causes problems whenever an OODB attempts to model a domain for which specialized secondary storage structures exist. Incorporating a new storage structure into an existing ....

B. Koch et al. Cache coherency and storage management in a persistent object system. In The Fourth International Workshop on Persistent Object Systems, 1990. 22

....and Morrison [3, 17] in the early 1980 s. PBR related collection is considered for instance in O Toole and Nettles [18] where the collector scans a possibly old copy of the data. We have found two distributed shared memories with PBR in the literature. The specification of the Casper collector [13] is sketchy and seems incapable of collecting a persistent object that has become garbage. EOS [12] has a tracing and copying GC that takes into account user placement hints to improve locality. However, their GC is quite complex and has not been implemented. Larchant RDOSS is a simplified version ....

Bett Koch, Tracy Schunke, Alan Dearle, Francis Vaughan, Chris Marlin, Ruth Fazakerley, and Chris Barter. Cache coherency and storage management in a persistent object system. In Proceedings of the Fourth International Workshop on Persistent Object Systems, pages 99--109, Martha's Vineyard, MA (USA), September 1990.

....and efficient. Another collector for DSM was developed by Kordale[11] His design is very complex and relies on a large amount of auxiliary information in the form of tables. These tables are used basically to control inter node references and the algorithm is a variation of mark and sweep. Casper [10] is a system that supports a single persistent address space shared by several clients in a network. Objects are cached by clients that are served by a central server where persistent objects are stored. Each client creates its objects within a separate area of the persistent address space, called ....

Bett Koch, Tracy Schunke, Alan Dearle, Francis Vaughan, Chris Marlin, Ruth Fazakerley, and Chris Barter. Cache coherency and storage management in a persistent object system. In Proceedings of the Fourth International Workshop on Persistent Object Systems, pages 99--109, Martha's Vineyard, MA (USA), September 1990.

....at the language level. In addition programmers could also control the placement of computations for performance reasons. Others have attempted to implement transparent distribution at the persistent store level [LJGS90, GADV92] or even to build a distributed persistent operating system itself [KSD 90, DdBF 94] Transparent distribution is a step towards more scalable architectures and offers a simple programming model. However, transparent distribution cannot be made to scale indefinitely [MdSAB96] The main reason is that the many inter store references needed to support transparent ....

....the target store. The traditional solution for sharing objects between stores is based on remote references to the object being shared [DC93] for example, by remote references requiring all operations to refer to the original store [Wai88] or cache coherency of copies requiring update propagation [KSD 90]. However, remote references force sharing to be highly dependent on the availability of the store in which the object resides, and thus precludes autonomous stores, availability and scalability [MdSAB96] We propose the concept of persistent spaces in which publishers release objects that can ....

B. Koch, T. Schunke, A. Dearle, F. Vaughan, C. Marlin, R. Fazakerley, and C. Barter. Cache coherency and storage management in a persistent object system. In Dearle et al. [DSZ90].

....with multiprocessors [2, 13] or with a small scale DSM [18] These authors make strong coherence assumptions. The concept of PBR was first proposed by Atkinson and Morrison [3, 20] in the early 1980 s. We have found two PBR DSM systems in the literature. The specification of the Casper collector [16] is sketchy and seems incapable of collecting a persistent object that has become garbage. EOS [14] has a tracing andcopyingGC that takes into account user placement hints to improve locality. However, their GC design is quite complex and has not been implemented. 6 Implementation of Larchant BMX ....

Bett Koch, Tracy Schunke, Alan Dearle, Francis Vaughan, Chris Marlin, Ruth Fazakerley, and Chris Barter. Cache coherency and storage management in a persistent object system. In Proceedings of the Fourth International Workshop on Persistent Object Systems, pages 99--109, Martha's Vineyard, MA (USA), September 1990.

....an asynchronous communication protocol with causal delivery is necessary to ensure GC safety. 5 Related work The concept of PBR was first proposed by Atkinson and Morrison [3, 15] in the early 1980 s. We have found two PBR DSM systems in the literature. The specification of the Casper collector [11] is sketchy and seems incapable of collecting a persistent object that has become garbage. EOS [10] has a tracing GC that takes into account user placement hints to improve locality. However, their GC is quite complex and has not been implemented. Much previous work in distributed garbage ....

Bett Koch, Tracy Schunke, Alan Dearle, Francis Vaughan, Chris Marlin, Ruth Fazakerley, and Chris Barter. Cache coherency and storage management in a persistent object system. In Proceedings of the Fourth International Workshop on Persistent Object Systems, pages 99--109, Martha's Vineyard, MA (USA), September 1990.

....system of Rosenblaum and Outerhout [24] all data, persistent and transient, is kept in logs. Recently there has been a trend in stock operating systems to provide limited access to the paging mechanisms influencing recovery management implementations such as page based logging and shadow paging [5, 15, 19]. Each of these systems alters the relative cost of crash recovery. To illustrate the MaStA model, we chose four specific recovery mechanisms: object logging and page logging , both with deferred updates [11] after image shadow paging (AISP) 9, 16] and before image shadow paging (BISP) 3, ....

Koch, B., Schunke, T., Dearle, A., Vaughan, F., Marlin, C., Fazakerley, R. & Barter C. "Cache Coherency and Storage Management in a Persistent Object System". in Dearle, Shaw, Zdonik (eds.), Implementing Persistent Object Bases, Principles and Practice, Morgan Kaufmann, 1991 pp 103-113.

....and Morrison [3, 16] in the early 1980 s. PBR related collection is considered for instance in O Toole and Nettles [17] where the collector scans a possibly old copy of the data. We have found two distributed shared memories with PBR in the literature. The specification of the Casper collector [12] is sketchy and seems incapable of collecting a persistent object that has become garbage. EOS [11] has a tracing andcopyingGC that takes into account user placement hints to improve locality. However, their GC is quite complex and has not been implemented. Larchant RDOSS is a simplified version ....

Bett Koch, Tracy Schunke, Alan Dearle, Francis Vaughan, Chris Marlin, Ruth Fazakerley, and Chris Barter. Cache coherency and storage management in a persistent object system. In Proceedings of the Fourth International Workshop on Persistent Object Systems, pages 99--109, Martha's Vineyard, MA (USA), September 1990.

No context found.

Koch, B., Schunke, T., Dearle, A., Vaughan, F., Marlin, C., Fazakerley, R. & Barter C. "Cache Coherency and Storage Management in a Persistent Object System". in Dearle, Shaw, Zdonik (eds.), Implementing Persistent Object Bases, Principles and Practice, Morgan Kaufmann, 1991 pp 103-113.

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