A. Kaplan and J. C. Wileden. Conch: Experimenting with enhanced name management for persistent object systems. In Proc. of the Intl. Workshop on Persistent Object Systems (POS), Workshops in Computing, Tarascon, Provence, September 1994. Springer-Verlag.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....it outlines how several name servers can coexist in a distributed system. The remainder of this paper, then, is focused to study alternative mapping techniques. 2 2. 1 Separation of the Name Server While there are only few approaches known so far to enhance name management in database systems [KW94] separated name servers are widely accepted in the area of operating systems [Wat81] Here, many different forms of named objects coexist (such as files, environment variables and hosts) and it is useful to handle them in a uniform way. Many database systems, however, integrate the name service ....

....avoid network communication; but, they can just as well run on different workstations. 2.2 Distributed Name Services In general, several name servers can coexist to make objects accessible in a distributed environment. Many name services (like the ones used in DNS [Moc87] NIS [Sun93] Conch [KW94] and SHORE [CDF 94] use a hierarchically structured name space. Spring uses a general naming graph in which name servers can be arbitrarily connected. Like coupling of name servers to clients and data servers described in the previous section, the distribution architecture is independent of ....

A. Kaplan and J. C. Wileden. Conch: Experimenting with enhanced name management for persistent object systems. In Proc. of the Intl. Workshop on Persistent Object Systems (POS), Workshops in Computing, Tarascon, Provence, September 1994. Springer-Verlag.

....and it outlines how several name servers can coexist in a distributed system. The remainder of this paper, then, is focused to study alternative mapping techniques. 2. 1 Separation of the Name Server While there are only few approaches known so far to enhance name management in database systems [KW94] separated name servers are widely accepted in the area of operating systems [Wat81] Here, many different forms of named objects coexist (such as files, environment variables and hosts) and it is useful to handle them in a uniform way. Many database systems, however, integrate the name service ....

....avoid network communication; but, they can just as well run on different workstations. 2.2 Distributed Name Services In general, several name servers can coexist to make objects accessible in a distributed environment. Many name services (like the ones used in DNS [Moc87] NIS [Sun93] Conch [KW94] and SHORE [CDF 94] use a hierarchically structured name space. Spring uses a general naming graph in which name servers can be arbitrarily connected. a) Server Name Data Server 6 6 Handle OID Data Handle Client b) 6 oe Name Server Data Handle Data Server Data OID Client c) ....

A. Kaplan and J. C. Wileden. Conch: Experimenting with enhanced name management for persistent object systems. In Proc. of the Intl. Workshop on Persistent Object Systems (POS), Workshops in Computing, pages 318--331, Tarascon, France, September 1994. Springer.

.... the scope of this paper, the name management mechanism in PolySPIN allows names to be assigned to objects in binding spaces (where binding spaces are collections of name object pairs) and names for objects to be resolved in contexts (where contexts are constructed from existing binding spaces) KW94] In addition, binding spaces may be assigned names, resulting in the ability to hierarchically organize the name space for objects (similar to directory structures found in almost all modern file systems) Coupled with the persistent store, this approach results in a name based persistence ....

Alan Kaplan and Jack Wileden. Conch: Experimenting with enhanced name management for persistent object systems. In Sixth International Workshop on Persistent Object Systems, Tarascon, Provence, France, September 1994.

....[6] was developed as a unifying conceptual foundation for integrating extended features in software systems. SPIN has previously been used as a basis for seamlessly integrating persistence, interoperability and naming capabilities in extended versions of the C and CLOS APIs of the Open OODB [8, 11]. The SPIN framework itself evolved out of our earlier work on persistence [16, 18, 15] interoperability [19] and name management [9, 10] all of which aimed at minimizing the impact of the extended capability on software developers or preexisting code. When extended with automated support for ....

....particular style of the Jspin persistence capability is similar to that provided for C and CLOS by the Open OODB. Enhanced name management: Jspin will provide a set of extended name management capabilities, based on the Piccolo model [10] and therefore suitable for use with Conch style tools [8]. These capabilities will be independent of (that is, orthogonal to) persistence. As a result, this enhanced approach to name management will be uniformly applicable to C , CLOS, and Java objects. Basis for polylingual interoperability among C , CLOS, and Java: The extensions provided by Jspin ....

A. Kaplan and J. Wileden. Conch: Experimenting with enhanced name management for persistent object systems. In M. Atkinson, D. Maier, and V. Banzaken, editors, Sixth International Workshop on Persistent Object Systems, Workshops in Computing, pages 318--331, Tarascon, Provence, France, Sept. 1994. Springer.

....in Open OODB 1.0) make available for C and CLOS. Enhanced name management: Independent of (that is, orthogonally to) persistence, JavaSPIN will provide a set of extended name management capabilities, based on the Piccolo model [6] and therefore suitable for use with Conch style tools [4]. As a result, this enhanced approach to name management will be uniformly applicable to C , CLOS, and Java objects. A Basis for polylingual interoperabilty among C , CLOS, and Java: The extensions provided by JavaSPIN will transparently incorporate the necessary information into Java objects ....

A. Kaplan and J. Wileden. Conch: Experimenting with enhanced name management for persistent object systems. In M. Atkinson, D. Maier, and V. Banzaken, editors, Sixth International Workshop on Persistent Object Systems, Workshops in Computing, pages 318--331, Tarascon, Provence, France, Sept. 1994. Springer.

....mechanisms. Our experimental platform is Open OODB, an open, extensible object oriented database management system [14] that presently supports both C and CLOS. Our work to date includes the development (using C ) of a context controlling shell implemented as a user interface to Open OODB [7]. In particular, the approach facilitates the sharing of context information among objects (both persistent and transient) while at the same time permitting individual objects to define their own context formation requirements. Furthermore, since context data can be associated with objects via an ....

A. Kaplan and J. C. Wileden. Conch: Experimenting with enhanced name management for persistent object systems. In Sixth International Workshop on Persistent Object Systems, Tarascon, France, Sep 1994.

.... the scope of this paper, the name management mechanism in PolySPIN allows names to be assigned to objects in binding spaces (where binding spaces are collections of name object pairs) and names for objects to be resolved in contexts (where contexts are constructed from existing binding spaces) [3]. In addition, binding spaces may be assigned 1 Name management is the means by which a computing system allows names to be established for objects, permits objects to accessed using names, and controls the meaning and availability of names at any point in time in a particular computation. ....

A. Kaplan and J. Wileden. Conch: Experimenting with enhanced name management for persistent object systems. In Sixth International Workshop on Persistent Object Systems, Tarascon, Provence, France, Sept. 1994.

....among, multiple and separate name management mechanisms in their underlying domains to be yet another aspect of the impedance mismatch problem in object oriented databases. Here we mention only three representative problems. A more extensive examination of these issues can be found in [17]. One of the primary goals of OODBs is to make persistence an orthogonal property to type, thus obviating the need, at least from the programmer s perspective, for traditional file systems and (relational) databases. A flexible and powerful approach to name management is crucial, therefore, since ....

....the model, we can avoid ad hoc solutions, reason about properties of a mechanism before implementing it, and achieve uniformity in the functioning of the mechanism. We have, in fact, used Piccolo as a basis for implementing a prototype user interface shell for an object oriented database system [17]. Being based on the Piccolo concepts, this shell provides much more general and powerful name management capabilities to users of the OODB than did its original interface, or than are found in most such systems. As a simple illustration of the Piccolo model s potential value to software ....

[Article contains additional citation context not shown here]

Kaplan, A., and Wileden, J. Conch: Experimenting with enhanced name management for persistent object systems. In Sixth International Workshop on Persistent Object Systems (Tarascon, Provence, France, Sept. 1994).

....[8] was developed as a unifying conceptual foundation for integrating extended features in software systems. SPIN has previously been used as a basis for seamlessly integrating persistence, interoperability and naming capabilities in extended versions of the C and CLOS APIs of the Open OODB [10, 13]. The SPIN framework itself evolved out of our earlier work on persistence [20, 23, 19] interoperability [24] and name management [11, 12] all of which aimed at minimizing the impact of the extended capability on software developers or pre existing code. When extended with automated support for ....

....particular style of the JSPIN persistence capability is similar to that provided for C and CLOS by the Open OODB. Enhanced name management: JSPIN will provide a set of extended name management capabilities, based on the Piccolo model [12] and therefore suitable for use with Conch style tools [10]. These capabilities will be independent of (that is, orthogonal to) persistence. As a result, this enhanced approach to name management will be uniformly applicable to C , CLOS, and Java objects. Basis for polylingual interoperability among C , CLOS, and Java: The extensions provided by JSPIN ....

A. Kaplan and J. Wileden. Conch: Experimenting with enhanced name management for persistent object systems. In M. Atkinson, D. Maier, and V. Banzaken, editors, Sixth International Workshop on Persistent Object Systems, Workshops in Computing, pages 318--331, Tarascon, Provence, France, Sept. 1994. Springer.

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