Keeping a very large website up-to-date: Some feasibility results
As websites grow large and become more sophisticated, organizations use structured database systems as a source of base data for information on the website. Thus, it has become critical to keep a very large website up-to-date in response to the frequent changes in base data. This gives rise to an im...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Institutional Knowledge at Singapore Management University
2000
|
| Subjects: | |
| Online Access: | https://ink.library.smu.edu.sg/sis_research/978 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Singapore Management University |
| Language: | English |
| Summary: | As websites grow large and become more sophisticated, organizations use structured database systems as a source of base data for information on the website. Thus, it has become critical to keep a very large website up-to-date in response to the frequent changes in base data. This gives rise to an important issue: Can a website be timely refreshed by executing a set of queries against the base data? In this paper, we investigate the feasibility of scheduling a set of queries to refresh a very large website. Based on two types (tight and loose) of feasibility requirements, we present feasibility results when the base data change with uniform, regular and random periods. We found that tight feasibility depends on the interval length between two consecutively raised cell refresh requests while it is NP-Hard to determine loose feasibility when the base data have regular or random update periods. For the case when the base data have the uniform update periods, loose feasibility of a set of refresh queries depends on the sum of execution times of the refresh queries. |
|---|