Final pt 2

Shared Memory Interprocess Communication Pros/Cons

Pros:
-Fast Speed/Communications

Cons:
-Protection / Syncing Processes has issues

Message Passing Interprocess Communication Pros/Cons

Pros:
-In a distributed system it's easier to implement than shared memory interprocess communication, no conflicts

Cons:
-Longer to implement because kernel intervention
- Once implemented, can only transfer small amounts of data

Microkernel Approach: Overview & Pros/Cons

Overview:
OS Structure which minimizes processes in kernel by placing them in user mode.

Pros:
- Minimizing processes in Kernel Mode makes OS more secure
- Less shared global variables that could potentially cause OS complications if used incorrectly.

Cons:
- Lots of context switching and interprocess communication can it make inefficient

How many processes are created by the following code?

int main() {
fork();
fork();
return 0;
}

4.

[Parent]
_______
[Parent] [1] //fork 1, 'parent' creates '1'
_______
[Parent] [1] [2] [3] //fork 2, 'parent' creates '2', '1' creates '3'

​Describe​​ the ​​actions ​​taken ​​by ​​a ​​kernel ​​to ​​context-switch​​between​ ​processes P1->P2

1. P1's PCB is saved
2. P2's PCB is retrieved

Short Term vs Long Term Scheduler

Short Term (AKA CPU Scheduler):
- Decides what process to be executed next
- Allocates CPU

Long-Term (AKA Job Scheduler):
- Selects which processes should be brought into the ready queue

Types of Processes (2)

I/O-Bound Process:
- Predominantly does I/O work, short CPU Bursts

CPU-Bound Process:
- Does more computations, few but long CPU bursts

Important resources used when process is created (5)

-State
-Number
-Counter (location of next processes to execute)
-Stack
-Register

Resources used when a thread is created (2)

- Register
- Stack

Deadlock definiton

Prevent sets of concurrent processes from completing their tasks

Process Utilization Steps (3)

1. Request
2. Use
3. Release

4 conditions that cause deadlock

1. Mutual Exclusion: Only one process can use at a time
2. Hold and Wait: A process holding at least one resource is waiting to acquire additional resources held by another
3. No Preemption: A resource can be released only voluntarily by the process holding it, after that process has completed its task
4. Circular Wait: Resources waiting on each other in a manner that a circular dependency is created

Firmware

Program loaded at power-up/reboot that initializes all aspects of system

How does device controller inform CPU that it's finished it's operation?

Causing an interrupt - OS's are interrupt-driven

Caching

Copying information info faster storage system

Upon attempting to access this information, cache is checked first to see if information is there

Multiprocessor Advantages (3)

1. Increased Throughput (processes being executed simultaneously)
2. Economy of Scale
3. Increased Reliability

Multiprocessor Types (2)

Asymmetric - Each processor gets a specific task

Symmetric - Each processor performs all tasks

Essential OS Services (8)

- Program Execution
- I/O Operations
- File Systems
- Communication
- Resource Allocation
- Accounting
- Error Detection
- Protection/Security

Purpose of System-Call Interface

Maintains table of numbers associated with each system call, and invokes the desired system call in kernel mode depending on user application

Core Dump

File capturing memory of the process, generated upon application failure

Crash Dumb

File capturing kernel memory, generated when OS crashes

System Boot Overview

Bootstrap loader, stored in ROM or EEPROM locates the kernel, loads it into memory, and starts it.

Process States (5)

- New: being created
- Running
- Waiting: waiting for event to occur
- Ready: waiting to be assigned to a processor
- Terminated: has finished execution.

Program Counter

Processor Register that points to the next instruction to execute.

Procedure for switching between processes P1 and P2

P1's state is saved into PCB1.
P2's state is retrieved from PCB2.

vice versa occurs when switching back.

Scheduling Queues of Processes (3)

- Job Queue - Set of all processes in the system
- Ready Queue - Set of all processes in main memory, ready and waiting to execute
- Device Queues - Set of processes waiting for an I/O device

Producer/Consumer Problem

Producer and Consumer are sharing a buffer. Producer cannot add data to it if the buffer full, and consumer cannot consume data if the buffer is empty.

Is blocking message passing synchronous or asynchronous?

Synchronous:
Sending is blocked until message is received, receiver is blocked until a message is available.

Socket definition and structure

Endpoint for communication
161.25.19.8:1625

Loopback: Special Case of 127.0.0.1 which refers to system on which process is running.

Types of Sockets (3)

UDP - Connectionless (Data is sent, receiving end is not considered)
TCP - Connection-oriented (Send/Receive confirmation is important)
Multicast Socket - Data can be sent to multiple recipients

External Data Representation Types (2)

Big-Endian and Little-Endian

Amdahl's Law

speedup

Multithreading Techniques

- Many-to-one: Multiple user-level threads mapped to single kernel thread.
- One-to-one: Creating a user-level thread creates a kernel thread, however # of threads per process is restricted due to overhead
- Many-to-many: A lot of user-level threads mapped to a lot of kernel threads
- Two-level: Blends many-to-many and one-to=one

Implicit Threading Definition

Compiles and run-time libraries manage threads instead of programmers

Thread Pool Advantages (3)

1. Quicker to access pre-made thread
2. Number of threads can be bound by pool size
3. Task is not restricted by overhead like creating a thread pool, allows room for different strategies

Parallel Regions definiton

Blocks of code that can run in parallel

Thread cancellation options (2)

Async Cancellation: Terminates thread immediately

Deferred Cancellation: Periodically checks to see if thread can be cancelled, which is signified by the cancellation point

Upcall

Kernel notifying thread's library scheduler about events (i.e. when thread is about to block), and thread library decides what to do.

Atomic Instructions

Non-interruptable

Bounded Waiting

There exists a bound, or limit, on the number of times other processes are allowed to enter their critical sections after a process has made request to enter its critical section and before that request is granted.

Layman's terms: Each process gets a chance to enter the critical section at least once, to prevent starvation

Priority Inversion

Scheduling Problem when lower-priority process holds a lock needed by a higher-priority process.

Dining-Philosopher Problem&Solution

Problem:
- 5 Philosophers / 5 Chopsticks,
- Need 2 Chopsticks to eat

Solution:
- Make picking up chopsticks part of the critical section
- Have odd-numbered neighbors pickup left then right chopstick
- Have even-numbered neighbors pickup right then left chopstick

//Right Chopstick First
do {
wait(chopstick[i]);
wait(chopstick[(i+1)%5];

// critical section - eating

signal(chopstick[i]);
signal(chopstick[(i+1)%5];
} while (true)

//Left Chopstick First
do {
wait(chopstick[(i+1)%5];
wait(chopstick[i]);

// critical section - eating

signal(chopstick[(i+1)%5]);
signal(chopstick[i];
} while (true)

Banker's Algorithm

What is an Operating system?

Execute user programs and make solving user problems easier.
Make the computer system convenient to use.
Use the computer hardware in an efficient manner.

It is a resource allocator and a control program.

Kernel

Another name for the OS. The one program running on the computer at all times.

System Programs

Associated the the OS but not necessarily part of the kernel.

Bootstrap Program

Program loaded up at power on/reboot. Typically stored in ROM or EPROM. Generally know as firmware. Initializes all parts of the system. Loads OS and starts it.

Interrupt

When the CPU is interrupted, it stops what it is doing and immediately
transfers execution to a fixed location.

Interrupt Vector

It is a table of pointers to interrupt service routines. Vector of addresses is indexed by a unique device number, given with
the interrupt request, to provide the address of the interrupt service routine for the interrupting device.

Trap/Exception

Software generated interrupt

Polling

Also know as busy waiting. It is in a loop, reading the
status register over and over until the busy bit becomes clear.

System Call

Also called a monitor call. Request to the OS to allow user to wait for I/O completion. Six major categories: process
control, file manipulation, device manipulation, information maintenance,
communications, and protection.

Device-status table

contains entry for each I/O device indicating its type, address, and state

Storage Device Hierarchy (Fastest->Slowest)

Registers>Cache>Main Memory>SSD>HDD>Optical Disk>Magnetic Tapes

Caching

Information in use copied from slower to faster storage temporarily to be used directly from cache.

Direct Memory Access (DMA)

After setting up buffers, pointers, and counters for the I/O device, the device controller transfers an entire block of data directly to or from its own buffer storage to memory, with no intervention by the CPU. Interrupt only generated once per block to let driver know op is completed.

Asymmetric Multiprocessing vs. Symmetric Multiprocessing

Asymmetric: each processor is assigned a specific task.

Symmetric: each processor performs all tasks.

Clustered System

Multiple computers connected together that usually have a shared storage-area network (SAN).

Multiprogramming (Batch System)

One job is selected and run via job scheduling. When it has to wait on I/O, OS switches to another job.

Timesharing (Multitasking)

is logical extension in which CPU switches jobs so frequently that users can interact with each job while it is running, creating interactive computing.

Emulation

can allow an OS to run on non-native hardware.

Semaphore

A semaphore S is an integer variable that, apart from initialization, is
accessed only through two standard atomic operations: wait() and signal().

wait()

wait(semaphore *S) {
S->value--;
if (S->value < 0) {
add this process to S->list;
block();
}
}

signal()

signal(semaphore *S) {
S->value++;
if (S->value

mutex lock

Protect a critical section by first acquire() a lock then release() the lock. Requires busy waiting.

Amdahl's Law

Identifies performance gains from adding additional cores to an application that has both serial and parallel components.

Multithreading Models

Many-to-one, one-to-one, many-to-many.

Thread Local Storage (TLS)

allows each thread to have its own copy of data.