Computer Science 4630

Spring 2000

Quiz 4

 

Circle the letter of the best answer to each question.  You have 15 minutes.

 

  1. An important difference between a hard link and a soft link in a file system is

(a)   Reference counts are kept for hard links but not for soft links.

(b)  Reference counts are kept for soft links but not for hard links.

(c)   Hard links are allowed to create cycles, but soft links are not.

(d)  Soft links require much less disk space than hard links.

 

  1. An advantage of the file allocation table file structure over the linked file structure is

(a)   The file allocation table solution uses much less disk space than the linked structure.

(b)  The linked structure uses much less disk space than the file allocation table structure.

(c)   The file allocation table method allows faster seeks in files.

(d)  The linked allocation method allows faster seeks in files.

 

  1. A Unix inode stores

(a)   Only the name of a file.

(b)  Only file allocation information about a file.

(c)   Only access permission information about a file.

(d)  Access permissions and file allocation information for a file, as well as some other information.

 

  1. Contiguous allocation of files has some advantages and some disadvantages.  Which of the following is NOT a disadvantage of contiguous allocation?

(a)   Fragmentation of the disk free blocks.

(b)  Slow seek time.

(c)   Difficulty of adding new blocks to the end of a file.

(d)  Difficulty of adding new blocks to the beginning of a file.

 

  1. Clustering is

(a)   Creating large logical disk blocks from collections of contiguous physical disk blocks.

(b)  Allocating directories close to one another.

(c)   Allocating files close to one another.

(d)  Moving the blocks of a file closer to one another on the disk.

 

  1. An advantage of clustering for a file allocation table is

(a)   The file allocation table can be made contiguous.

(b)  The file allocation table is smaller.

(c)   The file allocation table can probably fit in a single disk block.

(d)  The file allocation table becomes entirely unnecessary.

 


  1. Indexed allocation of files has the disadvantage that

(a)   Seeks tend to be very slow.

(b)  Large files use much more space than they would in other schemes.

(c)   Even a small file uses at least two disk blocks.

(d)  There are no disadvantages over any other schemes.

 

  1. When striping is used with a RAID disk array, reliability becomes a problem.  How is it compensated for?

(a)   Doing striping at a bit level instead of a byte level.

(b)  Dong striping at the block level instead of at the byte level.

(c)   Storing information redundantly.

(d)  It can only be solved by using highly reliable disk drives.

 

  1. To read or write a block on a disk, a process must

(a)   acquire the disk resource for its private use.

(b)  send a direct i/o request to the disk.

(c)   first check that the disk is powered up and spinning.

(d)  spool the request with the disk manager.

 

  1. A directory is a kind of file.  Which of the following is NOT a reason for not providing a means for processes to directly read and write directories the way ordinary files are read and written?

(a)   Directory organization changes from one operating system to another, so programs that do direct access to directories are non-portable.

(b)  Allowing a process to write a directory will allow the process to replace any permission information that is stored in the directory, and hence will reduce system security.

(c)   Processes that write a directory might destroy the directory structure, causing a large part of the file system to be lost.

(d)  A process is usually not allowed to find out the names of the files that are part of a directory.

 

  1. First-come-first-serve disk scheduling suffers from what problem?

(a)   There tends to be a large amount of disk head motion that can be reduced by using other scheduling algorithms.

(b)  The algorithm is not fair to processes, and can cause some processes to wait for a long time.

(c)   The algorithm causes files to become fragmented.

(d)  The algorithm is difficult to implement.

 

  1. The shortest-seek-time-first disk scheduling algorithm suffers from what problem?

(a)   There tends to be a large amount of disk head motion that can be reduced by using other scheduling algorithms.

(b)  The algorithm is not fair to processes, and can cause some processes to wait for a long time.

(c)   The algorithm causes files to become fragmented.

(d)  The algorithm is difficult to implement.