Скачать или смотреть GRE Computer Science Question 66

66. Query optimizers typically use summaries of data distributions to estimate the sizes of the intermediate tables
generated during query processing. One popular such summarization scheme is a histogram, whereby the input
range is partitioned into buckets and a cumulative count is maintained of the number of tuples falling in each
bucket. The distribution within a bucket is assumed to be uniform for the purposes of estimation.
The following shows one such histogram for a relation R on a discrete attribute a with domain [1..10].
• Bucket 1: Range = [1..2], Cumulative tuple count = 6
• Bucket 2: Range = [3..8], Cumulative tuple count = 30
• Bucket 3: Range = [9..10], Cumulative tuple count = 10
What is the estimated size of the self-join operation R R ?
(A) 46 (B) 218 (C) 248 (D) 1,036 (E) 5,672