Modul 6 - Filsystem

Question 1

Benefits of interrupts over polling? (I/0)

Answer 1

• Interrupts allow for overlap of computation and I/O, which improves utilization.
• Instead of polling the device repeatedly the os can issue a request, put the process to sleep and do a context switch.
• When the device is finished with the I/O a interrupt is raised by the hardware and handled by the interrupt handler. Then wakes the process that issued the I/O.

Question 2

When can interrupts be bad?

Answer 2

• If the I/O is performed quickly, interrupts can slow the process because of context switch and interrupt handling.
• In networks, when a huge stream of incoming packets each generate an interrupt, it is possible for the OS to livelock, that is, find itself only processing interrupts and never allowing a user-level process to run and actually service the requests.
• Solution: Wait a little bit before delivering an interrupt to the CPU.

Question 3

How to interact with a device?

Answer 3

• A register to read the status of the device.
• A register to instruct the device to read or
  write.
• A register that holds the data.
• I/O-bus could be separate from memory bus (or the same).
• The driver will use either special I/O instructions or regular load/store instructions.

Question 4

Access time for a HDD? (Seek, rotation, read)

Answer 4

• seek time: time to move arm to the right cylinder
• rotation time: time to rotate the disk
• read time: read one or more sectors

Question 5

What is a file system?

Answer 5

A file system is the user space implementation of persistent storage.

• A file is persistent i.e. it survives the termination of a process
• A file can be access by several processes i.e. a shared resource
• A file can be located given a path name

Question 6

What is a file?

Answer 6

• A sequence of bytes

Attributes, associated meta-data:

• Size and type
• Owner and permissions
• Author
• Created, last written
• Icons, fonts, presentation….

Question 7

What role does the directory have in a file system?

Answer 7

• It maps from name to identifier.
• Contains a list of (user-readable name, low-level name) pairs.
• Each entry in a directory is either a file or other directories.
• Directories also has an inode number (low-level name)

Question 8

What role does the file module have in a file system?

Answer 8

Locates file

Question 9

What role does access control have in a file system?

Answer 9

Interacts with authentication system

Question 10

What role does file operations have in a file system?

Answer 10

Read and write operations

Question 11

What role does block operations have in a file system?

Answer 11

Which blocks on which devices

Question 12

What role does device operations have in a file system?

Answer 12

Operations on physical drive

Question 13

What is an inode number?

Answer 13

It’s a low level name for a file, the user is not aware of the name.

Question 14

How is a file created and what happens?

Answer 14

• By passing the O_CREAT flag to the system call open().
• When a file is created an inode structure is created that will track all relevant information about the file (size, blocks on disk etc.)
• Then link a human-readable name to that file and putting that link into a directory.

Ex:
int fd = open(“foo”, O_CREAT|O_WRONLY|O_TRUNC, S_IRUSR|S_IWUSR);

The routine open() takes a number of different flags. In this example, the second parameter creates the file (O CREAT) if it does not exist, ensures that the file can only be written to (O WRONLY), and, if the file already exists, truncates it to a size of zero bytes thus removing any existing content (O TRUNC). The third parameter specifies permissions, in this case making the file readable and writable by the owner.

Question 15

What is a file descriptor?

Answer 15

• A file descriptor is just an integer, private per process, and is used in UNIX systems to access files.
• Once a file is opened, the file descriptor is used to read or write the file.
• The system call open() returns a file descriptor.

Question 16

What does lseek() system call do?

Answer 16

• Used to read or write to a specific offset within a file. (specific location)

off_t lseek(int fildes, off_t offset, int whence);
First argument is the file descriptor, second is the offset and the third determines how the seek is performed.

Question 17

What is a open file table?

Answer 17

• Open file table contains entries of the file descriptors that each process has (processes maintains these in an array).

. Table is global, one entry per open operation

• Also contains the current offset and information such as whether the file is readable or writable.

Question 18

When does processes share file tables entries?

Answer 18

• When a parent process creates a child process with fork().

Question 19

What happens if different processes reads the same file at the same time?

Answer 19

If the open file table isn’t shared, then each logical reading or writing of a file is independent.

Question 20

What happens when a file table is shared?

Answer 20

• When a file table entry is shared, its reference count is incremented; only when both processes close the file (or exit) will the entry be removed.
• Sharing open file table entries across parent and child is occasionally useful. For example, if you create a number of processes that are cooperatively working on a task, they can write to the same output file without any extra coordination.

Question 21

What does dup() system call do?

Answer 21

• The dup() call allows a process to create a new file descriptor that refers to the same underlying open file as an existing descriptor.
• The dup() call (and, in particular, dup2()) are useful when writing a UNIX shell and performing operations like output redirection;

Question 22

What does fsync() system call do?

Answer 22

• When a process calls fsync() for a particular file descriptor, the file system responds by forcing all dirty
  (i. e., not yet written) data to disk, for the file referred to by the specified file descriptor. The fsync() routine returns once all of these writes are complete.

Question 23

What is a hard link?

Answer 23

• A hard link is a way to create a link between a new file name to an old one. Basically you create another way to refer to the same file.

Question 24

What does the system call link() do?

Answer 24

• Creates a hard link between two file names.
• link() takes two arguments, an old pathname and a new one.
• Command-line program ln is used to do this.
• The way link works is that it simply creates another name in the directory you are creating the link to, and refers it to the same inode number (i.e., low-level name) of the original file. The file is not copied in any way

Question 25

What does unlink() system call do?

Answer 25

• Removes a file from the file system.

- By removing the hard links of the file.

Question 26

What are symbolic links (soft links)?

Answer 26

• A way to link to a directory (which hard links can’t).
• Way to link to files in other disk partitions. (Hard links can’t because inode numbers are only unique within a particular files system).
• Symbolic links are actually quite different from hard links. The first difference is that a symbolic link is actually a file itself, of a different type.
• Can leave dangling references if the original file is removed.

Question 27

What does mount do?

Answer 27

• Mount unifies all file systems into one tree.

Question 28

Inode tables?

Answer 28

• Inode table: one table per file system, one entry per file object
• Stores inodes which holds information (owner, access rights, file size etc) of the files in a file system.

Question 29

What does a file table entry hold?

Answer 29

The file table entry holds:
- Reference to inode

• Reference counter, when it reaches 0 the file is closed and the entry is removed
• The current position in the file
• Read and write access rights, determined when opened

Question 30

What is a bitmap?

Answer 30

It’s a simple structure which tracks data blocks are free or allocated. Each bit is used to indicate whether the corresponding object/block is free (0) or in-use (1).

• One for the data region (data bitmap) and one for the inode table (inode bitmap)

Question 31

What is the data region?

Answer 31

• It’s the region of the disk that is used for user data.

- Most of the space in a file system is user data.

Question 32

How is the space in a disk divided for a file system?

Answer 32

It’s divided into blocks, commonly-used size of 4kb.

Question 33

What are the components of a file system?

Answer 33

• Data blocks (D)
• Inodes (I)
• Inode bitmap (i)
• Data bitmap (d)
• Superblock (S)

Question 34

What information does the superblock contain?

Answer 34

• A superblock contains information about the file system, including for example how many inodes and data blocks are in the file system, where the inode table begins etc.

Question 35

What is the inode datastructure?

Answer 35

The inode is a structure that holds the metadata for a given file, such as its length, permissions, and the location of its constituent blocks. It is short for index node, as the inode number is used to index into an array of on-disk inodes in order to find the inode of that number.

Question 36

How are inodes designed?

Answer 36

• To support bigger files, file system designers have had to introduce different structures within inodes.
• One common idea is to have a special pointer known as an indirect pointer. Instead of pointing to a block that contains user data, it points to a block that contains more pointers, each of which point to user data.
• To support even larger files, add another pointer to the inode: the double indirect pointer. This pointer refers to a block that contains pointers to indirect blocks, each of which contain pointers to data blocks.
• You can have triple indirect pointer or more (supports even larger files).
• Overall, this imbalanced tree is referred to as themulti-level index approach to pointing to file blocks.

Question 37

Where are directories stored?

Answer 37

Often, file systems treat directories as a special type of file. Thus, a directory has an inode, somewhere in the inode table (with the type field of the inode marked as “directory” instead of “regular file”). The directory has data blocks pointed to by the inode (and perhaps, indirect blocks); these data blocks live in the data block region of our simple file system.

Question 38

How do we reduce the number of I/O to the (slow) disk when reading, writing or opening files?

Answer 38

• Most systems use system memory (DRAM) to cache important blocks.
• Modern systems, employ a dynamic partitioning approach. Specifically, many modern operating systems integrate virtual memory pages and file system pages into a unified page cache. In this way,memory can be allocated more flexibly across virtual memory and file system, depending on which needs more memory at a given time.
• Write buffering, delaying writes so that the file system can batch some updates into smaller set of I/O’s.
• Most modern file systems bufferwrites in memory
  for anywhere between five and thirty seconds, representing yet another trade-off: if the system crashes before the updates have been propagated
  to disk, the updates are lost; however, by keeping writes in memory longer, performance can be improved by batching, scheduling, and even avoiding writes.