1、附件： Spatial data management has been an activearea of research in the database field for two decades， with much of the research being focused on develop-ing data structures for storing and indexing spatialdata. However, no commercial database system pro-vides facilities for directly defining and sto
2、ring spa-tial data, and forniulating queries based on researchconditions on spatial data. We believe the followingare the relevant issues on which near-term researchshould be focused (in the order of decreasing impor-tance and urgency) First, relational query optimization techniquesneed to be extend
3、ed to deal with spatial queries， thatis， queries that contain search conditions on spatialpredicates to be developed. Second, more work needs to be done on experi-mental validation of the relative performance of someof the more promising data structures and indexingstructures proposed thus far, with
4、 consideration of amuch broader set of operations than just a few opera-tions that have typically been used in the limited per-formance studies conducted thus far. Third, it is difficult to build into a single data-base system multiple data structures for spatial index-ing, and all spatial operators
5、 that are useful for awide variety of spatial applications， as such, it isdesirable to build a database system so that it will beas easy as possible to extend the system with addi-tional data structures and spatial operators If the DBMS provides a way to interactively and update the database, as wel
6、l as interrogate it capability allows for managing personal data-Aces however, it does not automatically leave an audit trail of actions and does not provide the kinds of controla necessary in a multiuser organization. These-controls are only available when a set of application programs are customiz
7、ed for each data entry and updating function. Software for personal computers which perform me of the DBMS functions have been very popular arsenal computers were intended for use by individuals for personal information storage and process- These machines have also been used extensively small enterp
8、rises, professionals like doctors, acrylics, engineers, lawyers and so on By the nature of. intended usage, database systems on these machines except from several of the requirements of full doge database systems Since data sharing is not： Tended, concurrent operations even less so, the) fewer can b
9、e less complex. Security and integrity7aintenance are de-emphasized or absent. As data-) limes will be small, performance efficiency Is also important. In fact, the only aspect of a database system that is Important is data Independence. Data-.dependence, as stated earlier. Means that applicant prog
10、rams and user queries need not recognizantphysical organization of data on secondary storage. The importance of this aspect, particularly for the personal computer user, is that this greatly simplifies database usage. The user can store, access and manipulate data at a high level (close to the appli
11、cation) and be totally shielded from the low level (close to the machine) details of data organization. We will not discuss details of specific PC DBMS software packages here Let us summarize in the following the strengths and weaknesses of personal computer data-base software systems： The most obvi
12、ous positive factor is the user friendliness of the software. A user with no prior computer background would be able to use the system to store personal and professional data, retrieve and perform relayed processing. The user should, of course, satiety himself about the quality of software and the f
13、reedom from errors (bugs) so that invest-merits in data arc protected. For the programmer implementing applications with them, the advantage lies in the support for applications development in terms of input screen generations, output report generation etc offered by theses stems The main negative p
14、oint concerns absence of data protection features Unless encrypted, data cane accessed by whoever has access to the machine Data can be destroyed through mistakes or maliciousintent. The second weakness of manv of the PC-basedsystems is that of performance. If data volumes growup to a few thousands
15、of records， performance couldbe a bottleneck For organization where growth in data volumesis expected, availability of, the same or compatiblesoftware on large machines should be considered. This is one of the most common misconceptionsabout database management systems that are used inpersonal compu
16、ters Thoroughly comprehensive andsophisticated business systems can be developed indBASh, Paradox and other DBMSs However, theyare created by experienced programmers using theDBMSs own programming language. That is not thesame as users who create and manage personal filesthat are not part of the mai
17、nstream company system Data security prevents unauthorized users from viewing or updating the database Using passwords, the database, called subschema(pronounced sub-scheme)， For example, an employee database can contain all the data about an individual employee, but one group of users may be author
18、ized to view only payroll data, while others are allowed access to only work history and medical data. The DBMS can maintain the integrity of the database by not allowing more than one user to up-date the same record at the same time, The DBMS can keep duplicate records out of the database, for exam
19、ple, no two customers with the same customer numbers (Key fields) can be entered into the Data-Base. When a DBMS is used, the detailed knowledge of the physical organization of the data does not have to be built into every application program. The application program asks the DBMS for data by field
20、pine, for example, I coded representation of give customer name and balance due would be sent to he DBMS Without a DBMS, the programmer must secrecy space for the full structure of the in the program, Any change in data structure requires hangs in all the applications programs The multiple-database
21、model is represented by proposals for shared and private database architect rues, checkout and checking of data to and from shared and private databases Each user may populates/her private database with data checked out of the shared database, perform updates against the data. and check them back in
22、to the shared Database. The multiple-database model can be used to work around the conflict situations inherent in long lunation database sessions Since each user from the shared database and work against. is/her private database, disconnected from the、 hared database (at least on the surface)， the
23、usersan avoid the conflict situations In particular, multile users may be simultaneously updating the same thought having to wait for other users to comate late their updates However, when updated is to benecked into the shared database, it may have to be checked in as a new version, necessitating v
24、ersion management. Further, when data in a private data-base references data in the shared data， or vice versa, a private database is not really disconnected from the shared database. For example, the evaluation of query in general will require the database system to access both a private database a
25、nd the shared data-base, even if the query may have been formulated against a private database The multiple-database model is more appropriate than the single-database model in an environment where it is easy to determine in advance logical partitions of the database that correspond to work to be pe
26、rformed. The single-database model has received considerably more attention than the multiple- database model The focus of research into single-database model has been on addressing the twin problem of long waits and loss of work that the long duration of transactions brings about. The single-databa
27、se model requires the intro-diction of a notion of database consistency, and protocols for concurrency control and recovery, that are different from those supported in traditional database systems If a reasonable notion of database consistency is to be supported (I c. the database system is to enfor
28、ce it automatically), there are bound to be” conflict situations where one transaction comes into an access conflict against some other transaction. If await is to be avoided， some means of a negotiated settlement of the conflict must be provided, thereby dragging the users into the details of concu
29、rrency control The single-database model is more appropriate-ate than the multiple-database model in an environment where it is difficult to determine in advance logical partitions of the database that correspond to work to be performed, and the users closely cooperate. The objective of long-duration transactions is to model long-duration, interactive Database access sessions in application environments The fundamental assumption about short-duration of transactions that underlies the traditional model of transactions is inappropriate for long-duration
