Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
2002-09-17
2004-06-08
Homere, Jean R. (Department: 2177)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000, C707S793000, C707S793000, C707S793000, C707S793000, C709S201000, C709S219000, C709S229000, C717S108000, C717S116000, C717S165000
Reexamination Certificate
active
06748373
ABSTRACT:
COPYRIGHT NOTICE
A portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document of the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.
CROSS-REFERENCED CASES
The following applications are cross-referenced and incorporated herein by reference:
U.S. Provisional Application No. 60/349,432 entitled “System and Method for Dynamic Querying,” by Matt Shinn et al., filed Jan. 18, 2002.
FIELD OF THE INVENTION
The present invention relates to the querying of data, such as against a database.
BACKGROUND
Data in an Enterprise application is typically stored in a Database Management System (DBMS). Using such a DBMS system to manipulate data is awkward because the native language used by a DBMS is often proprietary and primitive. Enterprise JavaBeans (EJBs), or entity beans, provide a mechanism by which data stored in a DMBS can be manipulated in Java, which is a modern, object-oriented language.
A typical Java 2 Enterprise Edition (J2EE) platform includes one or more containers. A container is a runtime used to manage application components and provide access to J2EE application programming interfaces (APIs). An EJB container typically hosts Enterprise JavaBean components, such as entity beans or session beans. Apart from access to these infrastructure-level APIs, each container also implements a container-specific API, such as an EJB API. EJBs, or other application components, can be developed and hosted in the container.
EJB clients are typically applications that access EJB components in the containers. EJB clients can be application clients and components in the web container, such as Java servlets and JSP pages. A Java Database Connectivity (JDBC) API can provide a developer with the ability to connect to a relational database system. The API can allow transactional querying, retrieval, and manipulation from a JDBC-compliant database.
The EJB 2.0 specification, set forth by Sun Microsystems, Inc., of Santa Clara, Calif., defines a mechanism known as container managed persistence (CMP) by which the EJB implementation maps data in an DBMS to entity beans. This mapping includes both mapping to values in a DBMS and mapping of relations between entities in the DBMS.
Multiple CMP entity beans can have container-managed relationships among themselves using a local model with local interfaces. Instance variables for a bean can be designated as container-managed persistence fields or container-managed relationship fields. These fields can be defined in the appropriate deployment descriptor, for example. The values of these fields can be retrieved and set using public accessor methods, or getter and setter methods, which can be defined in an entity bean. Getter and setter methods can be declared public and abstract in a bean implementation class. The bean itself does not provide an implementation for these methods, as the methods are provided by the EJB container.
To implement a local model, an entity bean such as a customer entity bean can use its home interface to extend an EJB local home interface. The home interface includes finder methods for this bean. The home interface can be used to declare the finder methods, while the deployment descriptor contains the implementation of these methods. The customer bean can also define a local interface through which clients on the container can access methods for that bean. By default, a CMP implementation can load all mapped CMP values in a bean when the bean is first accessed. First access does not also cause related values to be loaded. The related values are instead faulted in when they are first used. Even if a user can control which mapped data should be loaded and which data should be faulted, a problem exists in how to predict or monitor how data will be used so as to improve system performance.
BRIEF SUMMARY
Systems and methods in accordance with the present invention provide for increased performance in data transfer and data storage systems. An entity bean instance having a number of fields, each representing a data item in a database, can be loaded into a container. A container-managed persistence mechanism can be used to monitor the accessing of each of these fields, such as by an EJB application. Using information gathered in the monitoring, a field group for the entity bean can be optimized. An optimized field group can allow the fields and associated entity beans that are most often accessed for a given entity bean can be automatically loaded when an entity bean is loaded into a container in an attempt to reduce the number of hits to the database. Access to a bean can be tracked by the entity and/or transaction causing the entity bean to be loaded. The fields that are least frequently accessed can be excluded from those that are automatically loaded, in order to reduce the amount of data transferred.
For instance, a counter can be updated for any field in the entity bean when that field is accessed, such as through the use of abstract accessor methods surrounding that field. The way in which the entity bean is accessed can be tracked, such as by using finder methods of a container interface used to access the entity bean. The way in which a transaction was started that accessed the entity bean can also be tracked, such as by annotating transaction context to include the way in which the transaction was started.
The optimization of a field group using this information can occur at regular time intervals, after a number of updates, periodically, or each time a bean is accessed. The optimization can include a balancing of the cost of loading a field with the frequency of loading a field. Field groups can also be optimized for applications and transactions that access the entity bean.
Any or all of the above optimization methods can be used to further optimize any such the field groups. This can adaptively optimize queries by pre-loading data that a query is most likely to access, while not loading data that a query is not likely to access.
REFERENCES:
patent: 6298478 (2001-10-01), Nally et al.
patent: 6411969 (2002-06-01), Tam
patent: 6457065 (2002-09-01), Rich et al.
patent: 2002/0049749 (2002-04-01), Helgeson et al.
patent: 2002/0077958 (2002-06-01), Gardner et al.
patent: 2002/0087563 (2002-07-01), Ghemawat et al.
patent: 2002/0124137 (2002-09-01), Ulrich et al.
patent: 2002/0147763 (2002-10-01), Lee et al.
patent: 2002/0152422 (2002-10-01), Sharma et al.
patent: 2003/0028682 (2003-02-01), Sutherland
patent: 2003/0056022 (2003-03-01), Carlson et al.
patent: 2003/0065827 (2003-04-01), Kufca et al.
Cheng-Wei-Chen et al., Runtime compositions and optimizations of parallel Javabean programs on clustering environments, Abstract only, International Conference on Parallel and Distributed Processing Techniques and Applications, vol. 4, Jun. 26-29, 2000.*
Code Generated for Supporting Associations between EJBs at EJB Creation Time, Abstract Only, IBM Corp. Nov. 9, 200.*
A Technique for virtually expanding queryable EJB CMP Entity Bean Fields, Abstract Only, IBM Corp., Feb. 2002.
Messinger Adam
White Seth
Bea Systems Inc.
Fliesler Dubb Meyer & Lovejoy LLP
Homere Jean R.
LandOfFree
System and method for adaptively optimizing queries does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for adaptively optimizing queries, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for adaptively optimizing queries will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3336268