Exercises to be handed-in marked with a star. Hand-in friday morning.

1. * Consistency: which cases of inconsistent data does PIER tolerate (does PIER not check for), and what could the possible causes of these inconsistencies?
2. What are the drawbacks of traditional central datewarehousing in comparison to an architecture like PIER?
3. What are the main functions of the application layer?
4. What are the main functions of the PIER layer?
5. What are the main functions of the DHT layer?
6. Specify some applications, which will tolerate the degree of query inconsistency provided by PIER (if You think such applications exist. Otherwise argue why not).
7. * You get the assignment to construct a Web site comparing computer hardware prices from different shops. For each type of hardware, You shall present some important product information and price (e.g. capacity in GBytes and prices for harddiscs). The sources are data from the individual shops' product and price information IT-systems. Describe how You can do this using the PIER architecture.
8. Specify some features of PIER architecture, which results in good performance for PIER distributed query, and some features of PIER architecture, which results in bad performance.
9. Specify some areas, where PIER can be improved, together with some ideas of how to make this improvement.
10. * Do You think, that PIER when using another DTH than CAN can have a better or worse performance? You could consider the areas routing, data redundancy, data inconsistency, reliability in storing and retrieving data.