Laserfiche WebLink
It is assumed that the word ?write? implies the appropriate portions of analysis, construction, and <br />testing/debugging for each of the SQL tasks listed above. Each SQL script should be written in such <br />a way as to output the exceptions to the rule/constraint that is being enforced. <br />All of the scripts should be cataloged, organized, and documented in an easy-to-follow, cookbook- <br />style presentation. For any table needing a refresh, the data administration staff would then have a <br />step-by-step guide to follow in verifying and applying the ?refresh? data. <br />Executing the Database Refresh Process <br />The staff involved in carrying out the refresh process should work from a prepared cookbook listing <br />the steps of the refresh to perform, what to look for after each step, and what the consequences of <br />each step will be. While the process cannot be reduced to an automated task, the goal is to reduce <br />the chances of errors being introduced during this process to a minuscule amount. Outlining all of <br />the possible outcomes of each step goes a long way to achieving that goal. <br />If questionable data items are detected, then the overall time needed to complete the process could be <br />variable since the data will need to be reviewed by the State?s Management team. <br />The estimated average execution time of the scripts is 4 to 6 hours per table being refreshed. The <br />time will vary based on the number of records involved in the refresh, the number of problems <br />encountered, and the available hardware resources during the refresh process. Executing the process <br />to refresh the derived tables would average 3 to 6 hours, depending upon the data type and the <br />number of records involved. <br />It is expected that as part of the normal software life cycle, the CRDSS environment will change and <br />evolve over time. The database refresh process must be part of any database maintenance work that <br />is performed in order to assure the continued accuracy and integrity of the data environment. <br />3.0 CONCLUSIONS AND RECOMMENDATIONS <br />The recommendation at this point in the project is to stress the ongoing integrity of the production <br />CRDSS database when processing data from any source outside the database. This focal point is <br />brought about by careful, detailed evaluation of the incoming data and thorough backup procedures <br />carried out against the production data. <br />11 <br />a320/taskmems/ 2-13-01.doc 01/03/97 <br />