S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, and R. H. Badani. ECA Rule Integration into an OODBMS: Architecture and Implementation. In Philip S. Yu and Arbee L. P. Chen, editors, Proc. of ICDE, pages 341--348, Taipei, Taiwan, March 1995. IEEE Computer Society.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....a communication request is addressed to the user. Each group can be associated to recurring situations, like when the user is on holidays or away on business: only one rule group can be active (i.e. evaluated for incoming calls) at any given time. Rules are modeled according to the ECA paradigm [2], i.e. Event Conditions Actions: An event is an external stimulus addressed to the user. In our architecture, events model requests for communication, and activate rules evaluation. Conditions represent constraints on rules scope. In our architecture, conditions represent situations set by ....

S. Chakravarthy, et al., ECA Rule Integration into an OODBMS: Architecture and Implementation, Technical Report UF-CIS-TR-94-023, University of Florida, May 1994. 12

....the point in time when rule execution starts, scheduling of multiple rules, etc. Active database research has initially focussed on the integration of active behavior into relational DBMSs (Starburst [74] Ariel [48] POSTGRES [65] A second generation of projects (including Ode [38] Sentinel [17] and SAMOS, later followed by REACH [6, 7] ACOOD [4] NAOS [22] TriGS [50] Chimera [15] and [51] investigated object oriented ADBMS. Reactive behavior as offered by ADBMS can be beneficially used by numerous application areas, such as financial applications [20] network management [3] ....

....manager, the object manager, etc. Does this imply that a from scratch implementation is necessary, or can a passive system be extended towards an active one We addressed these questions as follows: A powerful event definition language. Together with the languages proposed in the Sentinel [17] and in the Ode project [38] the SAMOS language was among the first expressive event definition languages. Our event language concentrates on the definition of complex events and offers a small number of event constructors that allow expressive event definitions in combination with the concept of ....

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, R.H. Badani. ECA Rule Integration into an OODBMS: Architecture and Implementation. Proc. 11 th Int'l Conf. on Data Engineering, Taipei, Taiwan, March 1995.

....et al. 1988a#. ADBMSs have received great attention lately, and several prototypes of object oriented ADBMSs are already available #e.g. ACOOD #Berndtsson, 1991#, NAOS #Collet et al., 1994#, Ode #Agrawal and Gehani, 1989#, REACH #Buchmann et al. 1995#, SAMOS #Gatziu et al. 1994#, SENTINEL #Chakravarthy et al. 1994##. We are currently in a position to evaluate the performance of ADBMSs by concentrating on # the performance requirements of di#erent architectural approaches; i.e. integrated versus layered, # di#erent techniques used for standard tasks of an ADBMS; i.e. rule maintenance, event detection, ....

Chakravarthy S., Tamizuddin Z., Krishnaprasad V., and Badani R. H., ECA Rule Integration into an OODBMS: Architecture and Implementation, Technical Report UF-CIS-94-023, University of Florida, Florida, 1994.

....in federated database systems is required respecting as much as possible the autonomy of the local database systems. The approach we describe in this paper is based on mechanisms as they are used for integrity enforcement in active database systems (e.g. SAMOS [GD94b] REACH [BZBW95] Sentinel [CKTB95] From our point of view, active rules (ECA rules) are a promising principle not only for integrity enforcement in single, centralized database systems, but also for federated heterogeneous systems. This is due to the fact that the active rule paradigm is in principle system independent. The ....

....[GD94b] for example, provides active functionality on top of the object oriented database system ObjectStore. In contrast, in the built in approach all active capabilities are integrated into the kernel of a database system. This approach is used, for example, by the prototype systems Sentinel [CKTB95] and REACH [BZBW95] which integrate active mechanisms into the object oriented database system OpenOODB. So far, we have presented the essential notions and issues of active database systems needed to understand how active mechanisms can be used in a federated database environment for integrity ....

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, and R. H. Badani. ECA Rule Integration into an OODBMS: Architecture and Implementation. In P. S. Yu and A. L. P. Chen, editors, Proc. of the 11th IEEE Int. Conf. on Data Engineering (ICDE'95), Taipei, Taiwan, pages 341--348. IEEE Computer Society Press, March 1995.

....events which was followed by TriGS. It has to be mentioned that it was not the main goal of TriGS to establish yet another composite event definition language. Rather, the approach is based on the work of existing active systems, such as Ode [Lieu96] REACH [Buch95] Samos [Gatz95] and Snoop [Chak95]. Furthermore, The application of TriGS to a non standard application area as mentioned above made it necessary, to put special emphasize on the following design and implementation goals: Powerful Representation of Composite Events. The representation of composite events has to be powerful ....

Chakravarthy, S., Krishnaprasad, V., Tamizuddin, Z., Badani, R.H., ECA Rule Integration into an OODBMS: Architecture and Implementation, in P. S. Yu, A. L. P. Chen (eds.), Proceedings of the 11th International Conference on Data Engineering (ICDE '95), IEEE Computer Society Press, Taipei, Taiwan, 1995.

....Active mechanisms are a promising approach for consistency maintenance in federated systems. Although the usage of the active object paradigm was already proposed in the context of the DOM project [2] existing prototypes of active OODBMS (e.g. REACH [4] SAMOS [12] ODE [13] NAOS [10] Sentinel [9], Chimera [7] simply enhance homogeneous database systems with active mechanisms. Analyzing the limitations of existing active DBMSs [3, 25] it has to be explored to what extent active features can be applied in federated systems [21] Due to the coexistence of several data models and the local ....

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, R.H. Baddani; ECA Rule Integration into an OODBMS: Architecture and Implementation, in: Proceedings of the 11th Internat. Conference on Data Engineering, Taipeh, 1995.

....to detect a conflict, the lock manager must check if all owners of LRBs with a locking mode incompatible with the request are predecessors of the requester (i.e. it must check for I owner (l; M ) NCW (C) Global Delegation of locks. All locking mechanisms proposed for nested transactions [33, 17, 25, 8], or for more flexible systems such 6 This concerns 28 of the 232 instructions of the extended JAM s instruction set. as ASSET [7] or the Reflective Transaction Framework of [6] implement delegation of locks by changing the ownership of each delegated lock. The delegation of a single lock ....

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, and R.H. Badani. ECA Rule Integration into an OODBMS: Architecture and Implementation. Technical Report UF-CIS-TR-94-023, University of Florida -- Computer and Information Sciences, May 1994.

.... waterlevel of the river is too low and water temperature of the river is too high and heat output of the reactor is too high ACTION reduce planned power of the reactor by 5 Nowadays there are several prototypes of active object oriented database management systems like REACH [BZBW95] and SENTINEL [CKTB95] both based on TI s Open OODB [WBT92] and SAMOS [GGD95] based on ObjectStore [LLOW91] each of them promising to be a good platform for applications in computer integrated manufacturing, network management, multimedia, etc. Great interest exists in comparing the functionality and performance of ....

Chakravarthy, S., Krishnaprasad, V., Tamizuddi, Z., Badani, R. H.; ECA Rule Integration into an OODBMS: Architecture and Implementation. Proc. 11th Int. Conf. on Data Engineering, Taipeh (Taiwan), 1995.

....to persistent objects. Using a wrapper function is also simpler and cleaner. The wrapper function is declared to be virtual if the corresponding member function is virtual. The wrapper function returns the value, if any, returned by the associated member function. The same technique is used in [9, 7, 4]. 2 Passing type CredCard is only an optimization to speed up finding trigger information at run time. 5.4 Run time Trigger Information We now describe the structures used to represent triggers at run time their states, masks, actions, and FSMs. We will also describe how these data ....

.... separated; however, we would then be forced to same restriction that REACH has namely that composite events can not be used in triggers that must fire immediately after the composite event occurs (i.e. they must be fired in deferred or detached mode) The overall architecture of Ode and Sentinel [7] is quite similar. Both use a pre processor to modify the user s code to post events, and both support a similar set of events. Both allow rule activation and deactivation at run time. Both support composite events, although the languages are somewhat different. However, there are also significant ....

[Article contains additional citation context not shown here]

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, and R. H. Badani. ECA rule integration into an OODBMS: Architecture and implementation. In Proc. Data Engineering, March 1995.

.... user defined actions in the form of Event Condition Action (ECA) rules [DBB 88] ADBMSs have received great attention lately, and several prototypes of object oriented ADBMSs are already available (e.g. ACOOD [Ber91] NAOS [CCS94] Ode [AG89] REACH [BZBW95] SAMOS [GGD94] SENTINEL [CVKB94] We are currently in a position to evaluate the performance of ADBMSs by concentrating on ffl the performance requirements of different architectural approaches; i.e. integrated versus layered, ffl different techniques used for standard tasks of an ADBMS; i.e. rule maintenance, event ....

S. Chakravarthy, Z. Tamizuddin V. Krishnaprasad, and R. H. Badani. ECA rule integration into an OODBMS: Architecture and implementation. Technical Report 94-023, CIS Department, University of Florida, 1994.

....defined in Section 4. When a passive OODB model is extended to provide active rules, inheritance and overriding models must be adapted to also work with such active OODBs. While many active OODB systems support some form of rule inheritance (e.g. Ode [Lieuwen 96] SAMOS [Geppert 95] Sentinel [Chakravarthy 95] and rule overriding is provided by (at least) one system (namely TriGS [Kappel 94] no general model of rule inheritance in active OODB systems has been established as of now. Such treatment is conspicuously missing for rules that are defined over multiple classes, even though most active ....

....rule should be a sub event type of the event of the original rule, though the meaning of subtyping for events is not presented. Similarly, there is no discussion of whether and how class spanning rules may be overridden in NAOS. Other active OODB systems such as SAMOS [Geppert 95] and Sentinel [Chakravarthy 95] support both class specific and class spanning rules, but issues related to rule overriding and the semantics of inheritance for classspanning rules have not been discussed. Similar comments apply to REACH [Zimmerman 96] In Ode, rules are defined within the definition of a class and are ....

S. Chakravarthy, et. al., "ECA Rule Integration into an OODBMS: Architecture and Implementation, " Proc. of 11th ICDE, 1995, 341-348.

....constructs for the modeling of reactive behavior, the development of concepts for rule execution, or the investigation of architectural issues for the construction of an active DBMS. A number of prototypes have been built, including Starburst [19] POSTGRES [16] ACCOD [5] REACH [6] Sentinel [7], NAOS [8] Ode [11] Ariel [13] TriGS [14] Chimera [19] and SAMOS [10] Although the advantages of active DBMS are nowadays well known, they are still not widely used in practice. This is even true for simple active mechanisms like triggers in relational systems which are commercially ....

....[1, 2, 3, 18] The existing approaches deal, however, only with simple rule languages, that means languages supporting only a limited set of constructs for specifying active behavior. Expressive rule languages like those as supported in, for example, SAMOS [10] ACCOD [5] REACH [6] Sentinel [7], NAOS [8] Ode [11] TriGS [14] and Chimera [19] offer more elaborate features (e.g. composite events, time events, complex actions, coupling modes) They generally allow the specification of more application semantics in the form of rules. However, the price to be paid is that rule analysis is ....

[Article contains additional citation context not shown here]

Chakravarthy S., Krishnaprasad V., Tamizuddin Z., Badani R.H.; ECA Rule Integration into an OODBMS: Architecture and Implementation; Proc. of the 11th Intl. Conf. on Data Engineering, Taipei, Taiwan, March 95.

....passive OODB model is extended to support active rules, inheritance and overriding models must be adapted to also work with such active OODBs. While many active OODB systems support some form of rule inheritance (e.g. Ode (Lieuwen, Gehani Arlien, 1996) SAMOS (Geppert, et al. 1995) Sentinel (Chakravarthy, et al. 1994)) and rule overriding is provided by (at least) one system (TriGS (Kappel, et al. 1994) no general model of rule inheritance in active OODB systems has been established to date. Such treatment is conspicuously missing for rules that are defined over multiple classes, even though most active ....

....rules (class external rules in SAMOS terminology) are identified with event 14 definitions rather than class definitions. Rules are inherited by the subclasses. The semantics of inheritance for class spanning rules and rule overriding though have not been discussed. The rule model for Sentinel (Chakravarthy, et al. 1994) is fairly similar to the SAMOS rule model. In Chimera (Ceri, et al. 1996, Meo, Psaila Ceri, 1996) again a distinction is made between targeted rules (with events belonging to a single class, i.e. class specific rules in our terminology) and untargeted rules (class spanning rules in our ....

Chakravarthy, S. et. al. (1994). ECA Rule Integration into an OODBMS: Architecture and Implementation (Report No. UF-CIS-TR-94-023). Dept. of Computer & Information Science, University of Florida, Gainesville, Florida.

....transaction models have been proposed to define the semantics of rule evaluation [2, 3] In these schemes, the action part of the rule is always fired as part of the event generating transaction i.e. the events are always intra transactional. In another Active Object DBMS (AODBMS) named Sentinel [4], an effort has been made to take care of inter transactional events i.e. where transaction generated events cross transaction and application boundaries. Thus an application transaction can take action using the events generated by other concurrently running transactions by defining detached ....

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, and R. H. Badani, "ECA Rule Integration into an OODBMS: Architecture and Implementation, " in Tech. Report UF-CIS-TR-94-023, Dept. of C.I.S., Univ. of Florida, Guinesville, Canada, 1994.

.... and performs corresponding user defined actions typically in the form of Event Condition Action (ECA) rules [16] ADBMSs have received great attention lately, and several prototypes of objectoriented ADBMSs are already available (e.g. ACOOD [4] NAOS [14] Ode [1] REACH [7] SAMOS [23] SENTINEL [13]) We are currently in a position to evaluate the performance of ADBMSs by concentrating on ffl the performance requirements of different architectural approaches; i.e. integrated versus layered, ffl different techniques used for standard tasks of an ADBMS; i.e. rule maintenance, event ....

....Ode [1] is a very good example of an ADBMS that is developed from scratch. ffl integrated architecture An existing passive DBMS is modified and customized to offer active functionality. This approach seems to be a compromise between development cost and functionality. REACH [7] and SENTINEL [13] are two objectoriented ADBMS prototypes that have integrated architectures. Both utilize the extensible DBMS Open OODB [34] as the underlying passive database system. NAOS [14] is another prototype ADBMS which is integrated in the O 2 object oriented database system [3] ffl layered ....

[Article contains additional citation context not shown here]

S. Chakravarthy, Z. Tamizuddin V. Krishnaprasad, and R. H. Badani. ECA rule integration into an OODBMS: Architecture and implementation. Technical Report 94-023, CIS Department, University of Florida, 1994.

....database. If this rule exists in the rule database, it will update the parameters of the existing rules in the rule database. The database uses trigger to monitor the parameters in the rule database. Once the condition is met, the specific trigger will be fired. ECA (Event Condition Action) model [CHA1995] can be used for more complex triggers. Figure 1.2 shows how active data mining works. One example of active data mining is that the supermarket can use association rule algorithm and active data mining to monitor the sale trends. Once the support or confidence parameter for a specific rule ....

Chakravarthy S., Krishnaprasad V., Tamizuddin Z., and Badani R. H., ECA Rule Integration into an OODBMS: Architecture and Implementaion, 11th International Conference on Data Engineering, Taipei, Taiwan, March 1995.

....Section 4 describes the implementation aspects. In section 5 we show several screen dumps produced by the tool at different stages of a sample application execution. Section 6 contains conclusions. 2 Sentinel Architecture In this section, we briefly discuss the overall architecture of Sentinel [CM94, AMC93, Bad93, CKAK94, CKTB95, Tam94] and its functional modules. The Sentinel architecture shown in Figure 1 extends the passive Open OODB system [WBT92, OOD93] The Open OODB toolkit uses Exodus as the storage manager and supports persistence of C objects. Concurrency control and recovery for the top level transactions are ....

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, and R. Badani. ECA Rule Integration into an OODBMS: Architecture and Implementation. In Proceedings, International Conference on Data Engineering, Feb. 1995.

....applications and transaction semantics to be modified independently of each other as well as use of existing applications without major modifications. For concreteness, we show how various transaction models can be translated into a set of ECA rules in the context of Sentinel [AMC93, CKAK94, CKTB95] an active OODBMS developed at UF. However, our framework and approach [CA95] is general and can be applied to any DBMS supporting active capability at the system level (relational or otherwise) The remainder of this paper is structured as follows. In section 2 we present our general approach ....

....of withdrawals appropriately and either continue or abort the transaction. To summarize, not only does the active database paradigm allow for the specification of transaction semantics but arbitrary semantics as well in an extensible manner. 2. 1 Overview of Sentinel Sentinel [CM94, AMC93, CKAK94, CKTB95] is an active object oriented DBMS that seamlessly integrates ECA rules into the object oriented paradigm. The Sentinel architecture is an extension of the passive Open OODB system architecture [OOD93] The behavior of objects has been extended to facilitate support of both system level and ....

[Article contains additional citation context not shown here]

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, and R. Badani. ECA Rule Integration into an OODBMS: Architecture and Implementation. In Proceedings, International Conference on Data Engineering, Feb. 1995.

No context found.

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, and R. H. Badani. ECA Rule Integration into an OODBMS: Architecture and Implementation. In Philip S. Yu and Arbee L. P. Chen, editors, Proc. of ICDE, pages 341--348, Taipei, Taiwan, March 1995. IEEE Computer Society.

No context found.

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, and R. H. Badani. ECA Rule Integration into an OODBMS: Architecture and Implementation. In Philip S. Yu and Arbee L. P. Chen, editors, Proc. of ICDE, pages 341--348, Taipei, Taiwan, March 1995. IEEE Computer Society.

No context found.

S. Chakravarthy, V. Krishnaprasad, Z. Tamizuddin, and R. Badani. ECA Rule Integration into an OODBMS: Architecture and Implementation. In Proceedings of the Eleventh International Conference on Data Engineering, 1995.

No context found.

Chakravarthy, S., Krishnaprasad, V., Tamizuddin, Z., and Badana, R.H., "ECA Rule Integration into an OODBMS: Architecture and Implementation", IEEE International Conference on Data Engineering, February 1995.

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