Term
| consider the following cache for a main memory system: cache access time is 25ns, Penalty time is 300ns, and Hit ratio is 75% Please calculate the Average Access Time for this cached mem. system |
|
Definition
|
|
Term
| Define the difference between user-level threads and kernel-level threads. |
|
Definition
| Kernel threads need not be associated with a process whereas every user thread belongs to a process. User-level threads are unknown to the kernel while kernel-level threads are |
|
|
Term
| What benefits do thread pools provide ? |
|
Definition
| eliminates the need for destroy/create routines. threads can be used over and over again. Eliminates a lot of overhead |
|
|
Term
Which of the following components of program state are shard across threads in a multi threaded process ?
A. Register Values
B. Heap Memory
C. Global Variables
D. Stack Memory |
|
Definition
| Heap mem & global variables |
|
|
Term
Process Burst Time
P1 10
P2 1
P3 2
P4 1
P5 5
processes arrive in order: 1,2,3,4,5
Do FCFS and SJF
Do TT |
|
Definition
FCFS:
P1 | P2 | P3 | P4 | P5 |
0 10 11 13 14 19
SJF:
P2 | P4 | P3 | P5 | P1 |
0 1 2 4 9 19
What is the turnaround time of EACH process for each of the
scheduling algorithms in part a?
Turn around time:
FCFS SJF
P1 10 19
P2 11 1
P3 13 4
P4 14 2
P5 19 9 |
|
|
Term
| What are the differences between a trap and a hardware-generated interrupt ? |
|
Definition
| a trap is generated by software (an error or a system call). A hardware generated interrupt is caused by hardware (such as I/O completion) |
|
|
Term
| Can traps be generated intentionally by a user program ? If so for what purpose ? |
|
Definition
| yes. A user program may start a system call to request an operating system service to be performed. (Since a user program can only execute non-privileged instruction, so when it needs to run some privileged instruction, it has to ask OS to do that for it) |
|
|
Term
Why are caches useful ?
If caches are so spiffy, why not make a chache that is the size of M. M. ? |
|
Definition
SPeed ! much faster to have cache in between the CPU & M. M.
Expensive to manufacture & memory is lost one the power is shut off |
|
|
Term
| Describe a mechanism for enforcing memory protection in order to prevent a program from modifying the memory associated with other programs |
|
Definition
| Use 2 registers, base and limit, to specify the beginning address and the size of the memory allocation of a process. Every time when the process makes a visit to the main memory, the MMU checks to make sure the address is in the range: [base, base+limit] |
|
|
Term
| The interface of an RMI server and client must extend ________ interface |
|
Definition
|
|
Term
| What is the purpose of command "rmic" ? |
|
Definition
builds a stub or skeleton of the class file for RMI
"rmic" is no longer needed with JDK 6.0 !! |
|
|
Term
| What is the command that starts the registry of the remote objects ? |
|
Definition
|
|
Term
TRUE OR FALSE:
After a process is loaded into the main memory of a sstem that doent have a memory scheduler, the process changes state among READY, SUSPENDED and BLOCKED, until it terminates |
|
Definition
|
|
Term
TRUE OR FALSE:
ADMISSION scheduler determines the degree of multiprogramming |
|
Definition
|
|
Term
TRUE or FALSE:
Unix uses FORK() system call to create a child process, and uses EXEC() to replace the child process' memory space with a new program |
|
Definition
|
|
Term
TRUE or FALSE
Win32 uses CREATEPROCESS system call to create a new process |
|
Definition
|
|
Term
TRUE or FALSE
MULTIPROGRAMMING - is a processing technique that loads many jobs in the main memory that will run concurrenlty |
|
Definition
|
|
Term
| Compare "multiprogramming" and "time-sharing" systems (list similarities and diff) |
|
Definition
time-sharing is an extension to multiprogramming
time-sharing gives the illusion tot eh user that he has total acces the the CPU when in reality program time is broken up into "time-slices:
Multiprog and time-sharing are similar in that they create a multitasking environment |
|
|
Term
| Draw an expanded state transition diagram |
|
Definition
|
|
Term
| Explain how semaphores can be used by a server to limit the number of concurrent connections |
|
Definition
| Only N sockets are allocated, acquire() and release() semaphores are used to acquire a socket, if all sockets are being used, acquire() must wait until a socket is free |
|
|
Term
| Explain why spin locks are not appropriate for single- processor systems yet are often used in multiprocessor systems |
|
Definition
single-processor systems: spinlocks waist CPU cycles
Multi-processor system use spinlocks because a thread can sit and "spin" on one processor while another thread/proc is executing on another processor |
|
|
Term
| Look at quiz 6... does it prevent a race condition ? |
|
Definition
| no, the "lock" idea doesn't work because two processes could set the lock bit at the same time |
|
|
Term
| What are the 4 necessary conditions for deadlock ? |
|
Definition
mutual exclusion
hold and wait
no preemption
a "circular" wait pattern |
|
|
Term
| explain the differences between internal and external fragmentation |
|
Definition
internal - the memory allocated to a process might be bigger than what the process originally requested
external - many processes cause little chunks of memory to not get used. The more processes, the worse frag gets (50% rule) |
|
|
Term
|
Definition
|
|
Term
| Consider a paging system with the page table stored in memory. If a memory reference takes 200ns, how long does a paged memory reference take? why ? |
|
Definition
| it takes 400ns, 200ns to lookup the page table reference and 200ns to look up the frame contents in memory |
|
|
