Operating Systems

Question 1

Operating System Structures

Answer 1

1. Monolithic
2. Layered
3. Microkernel
4. Modular

Question 2

Monolithic Structure

Answer 2

All devices are implemented
by a large kernel. Any new feature is added to the kernel. “Everything is connected to everything”.

Pros and Cons of the Monolithic Structure
Communications within the kernel is fast. However, the OS may be difficult to understand, difficult to modify and less secure.

Question 3

Layered Structure

Answer 3

An Operating System with a layered structure organizes its services into distinct layers. Each layer can only communicate with adjacent layers.

Pros and Cons of the Layered Structure
The OS is easy to debug. Development can easily be organized into separate units. However, performance may be poor since access to a specific service often requires traversal through multiple layers.

Question 4

Microkernel Structure

Answer 4

An OS with a microkernel structure has services implemented by servers, and a small kernel delivering messages between them.

Pros and Cons of the Microkernel Structure
The OS is secure and reliable. Performance may be poor due to increased communication overhead.

Question 5

Modular Structure

Answer 5

The operating system begins with a small kernel and adds or removes additional modules as needed

Pros and Cons of the Modular Structure
The OS is fast since unnecessary services do not need to be loaded and since it allows for direct communication between modules. However, as the number of loaded module increases, the OS begins to resemble a monolithic structure.

Question 6

Process Control Blocks

Answer 6

Process Number
Process State
Process address space
Process I/O

Each process is represented by a process control block (PCB). The process control block is stored in memory (or on disk).

Question 7

How does Context Switch work (A,B)

Answer 7

Stop running process A and start process B:
1. change the process schedule state of process A
2. save the context of process A
3. load the context of process B
change the process scheduling state of process B

Question 8

Context Switches (ctd)

Answer 8

• A context switch is pure overhead
  The time it takes depends on several factors, including the number of registers that must be copied.

Question 9

Sections of Process Address Space

Answer 9

A process addresses space has four sections:
1. A stack for temporary data
2. A heap for dynamically-allocated data
3. A data section for static data
a. size is constant, so data can be changes, but not in the size bigger than initial
a text section for program code

Question 10

The Heap

Answer 10

In the heap, blocks of memory are allocated and removed in an arbitrary order. The heap is used for variables that need to persist beyond the scope of a single function or need a flexible amount of memory.

Question 11

Process Spawning

Answer 11

A process is created at the request of a different process. In Linux there are four system calls for process spawning.

fork, exec, wait, exit

Question 12

Interprocess Communication can be implemented using? (0/2]

Answer 12

Interprocess communication may be implemented using either:
1. shared memory
message passing

Question 13

Shared Memory

Answer 13

Interprocess communication by shared memory involves processes writing data to/reading data from an agreed area of memory.

Question 14

Message Passing

Answer 14

Interprocess communication by message passing involves processes sending messages to/receiving messages from an agreed mailbox.

Question 15

Shared Memory vs Message Passing

Answer 15

Message passing requires regular system calls, while shared memory only requires system calls to set up the shared memory space. As a result, shared memory is generally more efficient for large data transfers, whereas message passing is better suited for smaller data exchanges.

Question 16

Controller Registers

Answer 16

The CPU communicates with the controller by reading and writing to the controller’s registers. (data in, data out, status, control)

Question 24

Port-based I/O vs Memory-mapped I/O

Answer 24

The CPU can access the controller registers using special instructions (port-based I/O) or standard memory instructions (memory-mapped I/O). Memory-mapped I/O is today much more common than port-based I/O.

Question 25

Polling

Answer 25

The CPU repeatedly checks the controller's status register to see whether the controller is busy. When the controller is ready and the CPU wants to give new instructions, the CPU writes to the data-out register and signals that is has done so through the control register.

Question 26

Interrupts

Answer 26

The CPU regularly senses an interrupt-request line. When the CPU gets an interrupt signal through the interrupt-request line, it stops the current process and initiates a response.

Question 27

Polling vs Interrupt

Answer 27

With interrupts, the CPU can use an interrupt controller (IC) to monitor the status of several devices at the same time, and serve them based on priority. With polling, the CPU needs to check each device individually in a round-robin fashion.

Question 28

When pollling may be preferable?

Answer 28

May if preferable if: * the controller and the device are fast * the I/O rate is high * some I/O data can be ignored the CPU has nothing better to do

Question 29

Direct Memory Access (DMA)

Answer 29

A CPU may offload large data transfers to a direct memory access (DMA) controller. The CPU writes a command block into memory, specifying the source and destination of the transfer. The DMA controller can then perform multiple transfers via a single command. When the transfer is complete, the CPU receives an interrupt from the DMA controller.

Question 30

Character I/O (System Calls)

Answer 30

Example devices: keyboards, computer mice, microphones, speakers. Characters must be processed in order that they arrive in the stream. The interface includes read and write operations. Character devices are low volume devices.

Question 31

Block I/O (System Calls)

Answer 31

Typically non-volatile mass storage devices. Block devices are used for to transfer blocks of data. The interface includes read and write operations. Block devices are high volume devices.

Question 32

Block I/O: Memory Mapped Files

Answer 32

A memory-mapped interface provides access to disk storage via a location in main memory: there is a system call that maps a file on the device into memory. The OS deals with transferring data between memory and the device and can perform transfers when needed, thereby reducing the number of mode switches.

Question 33

Network I/O

Answer 33

The interface includes operations for creating and connecting sockets, sending and receiving packets, checking whether a transfer was successful, etc. The main difference between network I/O devices and other I/O devices is that with network I/O devices things routinely go wrong (missing packets, etc.).

Question 34

Define wear-levelling and its purpose

Answer 34

Spreads writes across flash blocks to prevent early cell failure and extend SSD lifespan.

Question 35

Waht is dynamic wear leveling

Answer 35

Remaps only rewritten (hot) blocks to fresh blocks; static data stays put.

Question 36

What is static wear-levelling?

Answer 36

Moves cold data periodically so all blocks age evenly.

Question 37

Compare dynamic vs static.

Answer 37

Dynamic = simpler, fewer moves, uneven wear. Static = more moves, controller overhead, uniform wear → longer life.