tàu bản sao của tất cả các mối quan hệ với trang web SI. Sử dụng các kỹ thuật của Chương 13, hãy chọn một chiến lược để xử lý toàn bộ các truy vấn cục bộ tại trang web SI. • tàu một bản sao của mối quan hệ tài khoản để trang web S2, và tính toán Temp1 = tài khoản 1 gửi tiền tại S2. Tàu Temp1 từ S2 S3, và tính toán Temp2 = Temp1 1 chi nhánh tại S3. | 14733 Silberschatz-Korth-Sudarshan I VI. Database System Database System Architecture Concepts Fourth Edition 19. Distributed Databases The McGraw-Hill Companies 2001 Distributed Query Processing 737 Ship copies of all three relations to site Si. Using the techniques of Chapter 13 choose a strategy for processing the entire query locally at site Si . Ship a copy of the account relation to site s2 and compute temp1 account M depositor at s2. Ship tempi from s2 to S3 and compute temp2 tempiM branch at s3. Ship the result temp2 to Si . Devise strategies similar to the previous one with the roles of s1 s2 s3 exchanged. No one strategy is always the best one. Among the factors that must be considered are the volume of data being shipped the cost of transmitting a block of data between a pair of sites and the relative speed of processing at each site. Consider the first two strategies listed. If we ship all three relations to Si and indices exist on these relations we may need to re-create these indices at Sị. This re-creation of indices entails extra processing overhead and extra disk accesses. However the second strategy has the disadvantage that a potentially large relation customer M account must be shipped from s2 to s3. This relation repeats the address data for a customer once for each account that the customer has. Thus the second strategy may result in extra network transmission compared to the first strategy. Semijoin Strategy Suppose that we wish to evaluate the expression nN r2 where ri and r2 are stored at sites S1 and s2 respectively. Let the schemas of ri and r2 be R1 and R2. Suppose that we wish to obtain the result at si. If there are many tuples of r2 that do not join with any tuple of ri then shipping r2 to S1 entails shipping tuples that fail to contribute to the result. We want to remove such tuples before shipping data to si particularly if network costs are high. A possible strategy to accomplish all this is 1. Compute temp1 nR1 n R2 r
