Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
21 Cards in this Set
- Front
- Back
|
What is the "logical file system", the "basic file system" and what is the difference between them? |
The logical file system deals with the metadata associated with the file but not the data itself, and manages the directory structures and mapping file names to File Control Blocks. Basic File System works directly with device drivers to retrieve and store raw blocs of data on device. The Logical File System sits a couple of layers above the Basic File System. |
|
|
What does the File Organization Module do? |
Translates logical block number to physical block numbers and manages disk allocation. |
|
|
What is ZFS? |
An advanced File System developed for Solaris by Oracle. No need to know more about it though. |
|
|
What is the Boot Control Block? |
It's typically the first sector in a disk partition and contains instructions on how to boot the system. (If no boot data exists for this partition, the block will typically be left blank and not used for anything else) |
|
|
In the layered file system, which layer would receive a command like "retrieve block 123"? |
The Basic File System |
|
|
What information does the Volume Control Block hold? |
Total # of blocks, total free blocks, block size, in Unix - superblock, in Windows - master file table. |
|
|
Diagram of in-memory file systems |
|
|
|
What is a Virtual File System? |
Unix feature, provides an object-oriented way to implement a file system. Allows same API to be used for different types of file systems. (Similar in concept to Java Virtual Machine) |
|
|
What are two ways to implement the directory structure in memory? |
1) Linear List 2) Hash Table |
|
|
What's the problem with using a Hash Table to implement directories in a file system? |
Collisions for two file names that has to the same spot. Hash Table is only a good option if file names are fixed size, or if Hash Table uses chained-overflow |
|
|
What are three different file allocation methods? |
1) Contiguous 2) Linked 3) Indexed |
|
|
What are Extent-Based Systems, with regard to block allocation for files? |
It's a modified version of contiguous allocation scheme. Files are put in extents, which are contiguous blocks on the disk. |
|
|
What is Linked Allocation? |
A file is a linked list of blocks where the file lives. |
|
|
What are the advantages/disadvantages of Linked Allocation? |
Advantages: No external fragmentation, improved efficiency Disadvantages: Longer I/O time, increase internal fragmentation |
|
|
What is Indexed Allocation? |
Unlike Linked Allocation, where you have to start from the first Address to get to any others, Indexed Allocation maintains an index of all the pages in a file in an index table - giving direct access to random blocks |
|
|
How big should the Index block be when using Indexed Allocation? |
A multi-level index could be used to point a main index to smaller more specific indexes which in turn point to the file blocks |
|
|
How does UFS implement allocation? |
|
|
|
How do most file systems know which blocks/clusters are available? |
By maintaining a free-space (bit vector/bit map) list which is stored on disk (not in memory) since it must survive reboots |
|
|
What is consistency checking? |
Comparing data in directory structure with data blocks on the disk to find inconsistencies, that may have arisen for example due to a sudden reboot or failed I/O operation |
|
|
What are Log Structured File Systems? |
All file transactions are written to a log before they are actually performed. If a crash happens, the log can be read and the transaction could be performed. Once the transaction is performed, the log entry is removed. |
|
|
What is "copy-on-write"? |
Writing updates to new blocks and then repointing the inodes to these new blocks and eventually deleting the old one. |
