Since indexes speed up searches why wouldnt the dbms


1. Give brief and clear answers to the following:

(a) Explain how non-volatile RAM can help speed up disk writes

(b) Explain the main advantage(s) and disadvantage(s) of keeping the data file sorted.

(c) 17.28 parts (a) and (b)

(d) 18.3

(e) 18.7

(f) 18.8

(g) Since indexes speed up searches, why wouldn't the DBMS automatically create an index for every column of a table?

(h) Explain the main disadvantage(s) of static hashing.

(i) Briefly describe one situation in which ordered files may be better than static hashing.

(j) Explain the main advantage B-trees have over a multilevel index of the type shown in Figure 18.6.

2. In this problem, you have to calculate the number of disk accesses required in the worst case for different index organizations. The data base keeps track of different pieces of information for a student: name (25 bytes), idnumber (15 byes), gpa (10 bytes) and address (90 bytes). There are five hundred thousand students (it is a large university!). Each block can store 2000 bytes of information. A block pointer requires 10 bytes. The data file is sorted by the primary key, the student name. The idnumber is a secondary key. How many disk accesses are required if:

(a) access is through student name and there is a primary index on student name (similar to the index in Figure 18.1). You can assume that the index is small enough to be kept in RAM, and is already in RAM.

(b) access is through student name and there is a primary index on student name (similar to the index in Figure 18.1) which also has to be kept on disk.

(c) access is through student name and there is no index

(d) access is through student id and there is a secondary index on the student id (similar to the index in Figure 18.4) which also has to be kept on disk.

(e) access is through student name and there is a 2-level primary index on student name (similar to the index in Figure 18.6) in which the second level of the index is in memory, while the first level of the index is on disk. This is somewhat different from the way the Figure 18.6 index is organized in the sense that we have only 2 levels; the second level already in RAM, and the first level on the disk.

Solution Preview :

Prepared by a verified Expert
Basic Computer Science: Since indexes speed up searches why wouldnt the dbms
Reference No:- TGS01363469

Now Priced at $40 (50% Discount)

Recommended (97%)

Rated (4.9/5)