Suppose 19-bit instructions are copied to a CPU's instruction register, and suppose that each instruction dedicates the first 9 bits for the "opcode", with the remaining bits for specifying a memory location. What is the total number of possible memory locations that could be referenced by these instructions?

Using the information in the table below, what is the parent process ID for the process that printed this table to the Linux shell?

Suppose that a single-CPU system uses round robin scheduling with a time slice of 1 millisecond. The ready queue for this system always contains n processes that make no I/O requests, and the time taken to swap processes assigned to the CPU is negligible. A process (with a burst time of t milliseconds) arrives in the ready queue immediately after a process is assigned to the CPU. This newly arrived process eventually successfully terminates before any of the other processes. How long (in milliseconds) did the process take to execute in this system? 

 Suppose 2 programs -- A and B -- with service times 156 time units and 168 time units respectively, are executing on a system with a single CPU. The programs make no I/O requests.  Using a Round Robin policy with a time slice of 12 time units, the OS schedules the programs to run on the CPU -- program A runs first, then program B, then back to A, and so on. The OS uses timer interrupts to reassign the CPU from one program to another. 

Assume 1) no other program/OS activity consumes CPU cycles, 2) both programs will consume their service times and successfully terminate, and 3) no interrupts occur before the CPU executes the first instruction from program A. 

Based only on the information given, how many mode switches due to interrupts have occurred by the time program A has spent 155 time units running on the CPU?

Under the 5-state process model, processes with IDs 3, 8, 13, 16 and 25 are executing on a system. They arrived in the ready queue in the order they are listed, starting with process 3. The processes do not fail and do not make I/O requests. The short-term scheduler uses the following preemptive priority scheduling algorithm (processes with larger priority numbers have higher priority):

- the process with the highest priority runs on the CPU (FCFS is used, in case of a tie);

- a process has priority 0 when it first enters the ready queue;

- the priorities of all processes in the ready queue continuously decrease at the same rate;

- the priority of the process running on the CPU also continuously decreases, but more slowly than the priorities for processes in the ready queue.

Based only on this information, what is the order in which the processes finish executing?

An OS is using round-robin CPU scheduling (time slice=3ms) on a system with 10 CPUs. A process has just arrived on the system and immediately creates 11 threads. These threads have the same service time of 21ms each, they make no I/O requests, and they will all successfully terminate. We assume: 1) the 5-state process model, 2) no other user programs are executing, 3)  thread management and OS activities consume negligible time, and 4) the process immediately terminates when all of its threads have terminated.

 What is the turnaround time (in ms) for this process if these are:

1) user-level threads (ULTs)? 

2) kernel-level threads (KLTs) that can be assigned to different CPUs in parallel?

An OS is using round-robin CPU scheduling (time slice = 3ms) on a system with 9 CPUs. A process has just arrived on the system and immediately creates 10 threads. These threads have the same service time of 24ms each, they make no I/O requests, and they will all successfully terminate. We assume: 1) the 5-state process model, 2) no other user programs are executing, 3)  thread management and OS activities consume negligible time, and 4) the process immediately terminates when all of its threads have terminated.

 What is the turnaround time (in ms) for this process if these are:

1) user-level threads (ULTs)? 

2) kernel-level threads (KLTs) that can be assigned to different CPUs in parallel?

An OS is using round-robin CPU scheduling (time slice=3ms) on a system with 9 CPUs. A process has just arrived on the system and immediately creates 10 threads. These threads have the same service time of 21ms each, they make no I/O requests, and they will all successfully terminate. We assume: 1) the 5-state process model, 2) no other user programs are executing, 3)  thread management and OS activities consume negligible time, and 4) the process immediately terminates when all of its threads have terminated.

 What is the turnaround time (in ms) for this process if these are:

1) user-level threads (ULTs)? 

2) kernel-level threads (KLTs) that can be assigned to different CPUs in parallel?

If the answer involves "I/O tables", what is the question?

If the answer involves "the medium-term scheduler", what is the question?