FIT2100 Operating Systems
The Complete Study & Assessment Guide · Semester 2 2026FIT2100 Overview
- Monash University
- Semester 2 2026
- 12 chapters
- Computer Science
Operating Systems explains the software layer that turns processor, memory, storage and devices into safe services for applications. The unit begins with instruction execution, cache, buses and interrupts, then introduces operating system services, privilege modes and system calls.
- Assessed by 1. Demonstration 30% · 2. Exercise 20% · 3. Quiz / Test 50%
- Key terms Interrupt vector, System call, Direct memory access, Process control block
- Assessment move Build one physical revision sheet around mechanisms rather than lecture titles.
How FIT2100 is assessed
| Component | Weight | Format |
|---|---|---|
| 1. Demonstration | 30% | Laboratory submissions in even teaching weeks. Work is submitted and checked during the allocated session, with explanation of functionality and understanding. |
| 2. Exercise | 20% | Applied-class submissions in odd teaching weeks. Written work and in-class discussion or explanation contribute to the evidence. |
| 3. Quiz / Test | 50% | Problem Solving Assessments with calculation, single-answer multiple choice and multiple-answer multiple choice questions based on workshop content. |
What FIT2100 covers
Follow the system from processor and memory through devices, files, processes, scheduling, concurrency, virtual memory, communication and security.
Computer Architecture and Interrupts
CPU registers, instruction cycles, cache, main memory, buses and the interrupt path that makes slow devices manageable.This chapter connects the processor, registers, cache, main memory, buses and input or output controllers to the operating system that manages them. The essential story is control transfer. Normal work follows the fetch, decode and execute cycle.
- CPU registers and the instruction cycle
- Program counter and instruction register
- Main memory and volatility
- Cache and locality
- System buses and controllers
- Interrupt sources and vectors
Operating Systems and System Calls
Operating system services, privilege modes, abstractions and the controlled boundary between applications and kernel work.An operating system offers abstractions while controlling shared resources. Applications see processes, files, address spaces and communication endpoints instead of raw controller registers.
- Operating system objectives
- Resource management and abstraction
- User mode and kernel mode
- Protected instructions and memory
- System-call entry and return
- Process and file services
I/O Devices and DMA
Block and character devices, controller registers, polling, interrupts, direct memory access and disk scheduling.Input and output hardware varies by transfer unit, addressing and timing. Block devices move addressable blocks, while character devices expose streams. Controllers translate between those device behaviours and the system bus.
- Block and character devices
- Controller registers
- Memory-mapped input and output
- Polling
- Interrupt-driven transfers
- Direct memory access
Files, Directories and Storage
Persistent files, directory resolution, metadata, allocation methods, block size and the cost of storage access.A file system provides persistent, named information that can outlive a process and be shared under controlled rules. Directories map path components toward file objects. Metadata records type, ownership, permissions, size, times and block locations.
- Persistent storage requirements
- File operations
- Directory trees
- Path resolution
- File metadata
- Contiguous allocation
Processes and State Transitions
Execution context, process control blocks, creation, termination, ready and blocked states, and context switching.A process combines an executing program with its address space, processor context, identity, resource references and scheduling state. Two executions of the same program are distinct because their registers, stacks, memory and open objects differ.
- Program versus process
- Address space and context
- Process control blocks
- Process creation
- Parent and child relationships
- Running, ready and blocked
CPU Scheduling and Performance
Eligibility, preemption, FCFS, shortest job, round robin, turnaround, response and policy trade-offs.CPU scheduling chooses among ready processes. The policy must first respect eligibility: a process cannot be selected before arrival or while blocked.
- Scheduling decision points
- Preemptive and non-preemptive policies
- FCFS
- Shortest job policies
- Round robin and quantum
- Priority and fairness
Threads, Races and Mutual Exclusion
Shared process resources, private thread state, failing interleavings, critical sections and precise lock scope.Threads separate execution from process resource ownership. Threads in one process share code, data, heap and open resources, while each retains a private program counter, registers and stack.
- Thread model
- Shared process resources
- Private thread state
- Thread libraries
- Race conditions
- Critical sections
Semaphores and Deadlock
Atomic semaphore operations, deadlock conditions, prevention, avoidance, detection and safe sequence reasoning.Semaphores coordinate concurrent work through atomic wait and signal operations. A binary semaphore can enforce one-at-a-time entry, while a counting semaphore represents several available resource instances.
- Semaphore operations
- Binary and counting semaphores
- Mutexes
- Deadlock definition
- Four necessary conditions
- Prevention and avoidance
Memory Allocation and Paging
Logical and physical address spaces, page tables, address translation, fragmentation and replacement policies.Memory management gives each process a protected logical address space and maps active content into physical memory. Contiguous allocation is easy to translate but suffers from external fragmentation and difficult growth.
- Logical and physical memory
- Contiguous placement
- External fragmentation
- Paging and frames
- Page tables
- Address translation
Virtual Memory and IPC
Demand paging, fault cost, thrashing, signals, queues, shared memory, pipes and communication choices.Virtual memory lets a process use a logical address space larger than its current physical allocation. Demand paging loads a page when a valid reference finds it absent.
- Virtual address spaces
- Demand paging
- Page-fault handling
- Effective access time
- Thrashing
- Signals and queues
Pipes, Shared Memory and Sockets
Descriptor lifetime, shared regions, blocking streams, client and server calls, framing and partial input or output.IPC channels differ in storage, naming and guarantees. Pipes are kernel-managed byte streams with read and write ends. Unnamed pipes naturally connect related processes that inherit descriptors, while named pipes provide a persistent namespace entry.
- Unnamed pipes
- Named pipes
- Descriptor inheritance
- Blocking and end-of-file
- Shared-memory mappings
- Socket domains and types
Operating System Security
Authentication, authorisation, isolation, access control, malicious software, maintenance and complexity risk.Operating-system security layers authentication, authorisation, isolation, auditing and maintenance around valuable resources. Authentication establishes an identity. Authorisation determines what that identity may do.
- Security goals
- Authentication
- Authorisation
- Access control
- Memory protection
- Malicious software
It continues through device controllers, direct memory access, file systems, process state, CPU scheduling, threads and synchronisation, memory management, inter-process communication and security.
The assessed skills are both conceptual and quantitative.
Students trace control transfers, compare policies, calculate scheduling metrics and disk movement, translate logical addresses, evaluate page-fault cost and determine whether a resource state is safe. The strongest responses show the method and then explain the trade-off.
For example, a short scheduling average is incomplete without an eligibility timeline, and a deadlock verdict is incomplete without Need, Available and the sequence used to test progress.
The published assessment is entirely in semester. Demonstration contributes 30%, Exercise contributes 20%, and Quiz / Test contributes 50%. Laboratory and applied work require submission plus an explanation during the allocated class.
The captured Problem Solving Assessments use numerical calculation, single-answer multiple choice and multiple-answer multiple choice. Their instructions permit physical notes and a handheld scientific calculator while prohibiting online notes, software calculators, secondary devices and generative AI.
This guide follows the published teaching route and uses original practice values.
Each chapter pairs a system map with a worked mechanism, two checks and a demonstration move. Use it to rebuild reasoning in your own words, then confirm current task settings in Moodle.
Worked example · free
Original scheduling practice: build the timeline before the average
- +2Draw the FCFS timeline. A is the only ready process at time 0, so it runs from 0 to 5. B then runs from 5 to 7, followed by C from 7 to 10.
- +2Read completion times from the right edge of each final segment: A completes at 5, B at 7 and C at 10.
- +2Subtract arrival from completion: A turnaround is 5, B is 6 and C is 6. Keep arrival and service columns separate.
- +2Compute the mean: (5 + 6 + 6) divided by 3 equals 5.67 time units after final rounding. Check that no turnaround is below its service time.
Key terms
- Interrupt vector
- A table or mapping used to locate the service routine for an interrupt source. It supports controlled transfer to trusted handler code.
- System call
- A controlled request by a user process for a kernel service such as file access, process creation or communication.
- Direct memory access
- A controller mechanism that transfers a programmed block between a device and memory without processor instructions for every byte.
- Process control block
- The kernel record containing saved execution context and management information needed to suspend and resume a process.
- Turnaround time
- The elapsed time from process arrival or submission to completion. It equals completion time minus arrival time.
- Race condition
- A defect in which the result depends on an uncontrolled interleaving of operations on shared state.
- Critical section
- The complete region that accesses a shared invariant and must be protected from unsafe concurrent execution.
- Safe sequence
- An order in which every process can obtain its remaining need, finish and release resources without deadlock.
- Page fault
- An exception raised when a valid referenced page is not resident and needs operating-system service before the instruction can continue.
- Shared memory
- An IPC method that maps the same memory region into multiple processes and therefore requires explicit synchronisation for conflicting access.
FIT2100 FAQ
How is assessment completed in this unit?
The published unit introduction states that assessment is conducted in semester and explicitly says there is no final exam. The official assessment structure instead allocates the grade across Demonstration, Exercise and Quiz / Test components. Confirm the live Moodle task list for any operational updates.
What makes the unit quantitatively difficult?
The calculations span several models rather than one formula set. Scheduling requires eligibility and timelines, disk questions require absolute head movements, paging requires quotient and remainder, and deadlock safety requires matrices and a repeatable sequence. Unit conversions and hidden state assumptions create more errors than arithmetic complexity.
What do the problem-solving assessments look like?
The captured instructions describe numerical calculation, single-answer multiple choice and multiple-answer multiple choice. Questions are based on workshop content. Multiple-answer items require particular care because an incorrect extra selection can remove credit for the whole item.
What materials are permitted in a problem-solving assessment?
The captured assessment instructions permit physical notes and a handheld scientific calculator. They prohibit online notes, software or online calculators, secondary devices and generative AI. Always check the current task page because the live instructions control the sitting.
Why is an in-class demonstration important?
The laboratory instructions and rubric make explanation part of the assessed evidence. A complete submission may receive reduced or zero marks if the student does not demonstrate it or cannot explain how it works. Prepare a normal run, an edge case and a concise mechanism explanation.
How should scheduling calculations be checked?
Draw the service timeline first, rebuild the eligible set at every decision and read completion times from the final segment. Then subtract arrival for turnaround and divide by service only when normalised turnaround is requested. A process cannot complete before arrival plus its own service.
How are unsafe and deadlocked states different?
A deadlocked set cannot progress because each process waits for an event another member must cause. An unsafe state means the system cannot guarantee a completion sequence under the declared maximum demands, even though processes may still finish if their future requests are smaller.
Questions students ask in each FIT2100 chapter
- Why are interrupts better than waiting in a loop?
- A waiting loop repeatedly consumes processor cycles while checking whether a slow device has finished. Interrupt-driven completion allows useful work to run until the controller signals the event. There is still entry and handler overhead, but the CPU is not occupied by continuous status checks. Chapter 01 · Computer Architecture and Interrupts
- Why can an application not access every device directly?
- Direct access would let untrusted code bypass protection, interfere with other processes and depend on device-specific register layouts. Kernel services validate requests, coordinate sharing, report failures and present a stable abstraction above different hardware. Chapter 02 · Operating Systems and System Calls
- Does direct memory access eliminate interrupts?
- No. The controller can move a bulk block without processor instructions for every byte, but completion or failure is commonly reported through an interrupt. The CPU still configures the transfer and handles the resulting event. Chapter 03 · I/O Devices and DMA
- Why is a file name separate from the file data?
- A directory manages the namespace, while metadata and block pointers identify the stored object. Separating these roles supports links, renaming and cached open state without rewriting every byte whenever a name changes. Chapter 04 · Files, Directories and Storage
- How is a process different from a program?
- A program is executable code and related static data, while a process is one active execution with changing registers, stack, address space, open resources and state. Several processes can run the same program without sharing the same execution context. Chapter 05 · Processes and State Transitions
- Which scheduling algorithm is best?
- There is no universal winner. Batch systems value throughput and turnaround, interactive systems value response, and real-time systems care about deadlines. A recommendation must name the workload objective and discuss fairness or overhead. Chapter 06 · CPU Scheduling and Performance
- Why are threads cheaper to communicate between?
- Threads already share their process address space, so they can exchange data through common memory without a separate inter-process transfer. That convenience also means concurrent access needs an explicit rule whenever at least one thread writes. Chapter 07 · Threads, Races and Mutual Exclusion
- How does a counting semaphore differ from a mutex?
- A counting semaphore represents a quantity of available capacity and need not encode ownership. A mutex expresses exclusive ownership more directly and is released by the owner under the intended discipline. The primitive should match the constraint. Chapter 08 · Semaphores and Deadlock
- Why does paging avoid external fragmentation?
- Every physical frame has the same size, so any free frame can hold any logical page. A process does not need one adjacent physical region. Partly used final pages can still waste space internally. Chapter 09 · Memory Allocation and Paging
- Is every page fault an invalid access?
- No. A valid page may simply be absent under demand paging, in which case the operating system loads it and restarts the instruction. An invalid address or denied access is a protection failure and follows a different outcome. Chapter 10 · Virtual Memory and IPC
- When does an empty pipe return end-of-file?
- The buffer must be empty and every write descriptor referring to that pipe must be closed. One inherited write end held by the reader is enough to prevent the kernel from concluding that no more bytes can arrive. Chapter 11 · Pipes, Shared Memory and Sockets
- How are authentication and authorisation different?
- Authentication supplies evidence about who is acting. Authorisation checks whether that identity has the requested right over a particular object. A valid login does not imply unrestricted file or kernel access. Chapter 12 · Operating System Security
How to prepare for the assessments
Build one physical revision sheet around mechanisms rather than lecture titles. Put process states beside scheduling metrics, race conditions beside semaphore rules, and paging beside effective access time. For every formula, retain one original worked example with units. For every policy, write the workload objective and one trade-off. Rehearse multiple-answer questions by judging each statement independently.
Before applied and laboratory sessions, run your work from a clean start, test one failure path and practise a concise explanation of the control flow. The rubrics reward clarity, correctness, initiative and demonstrated understanding. Do not rely on an output screenshot or a memorised script.
For Problem Solving Assessments, use the permitted physical notes and handheld scientific calculator only, and confirm current instructions in Moodle. The final revision loop is simple: map the resource, trace the state change, compute the result, name the trade-off and check the boundary condition.
Your AI Computer Science tutor for FIT2100
Stuck on a hard FIT2100 question? Sia is AskSia’s AI Computer Science tutor — ask any FIT2100 Operating Systems question and get a clear, step-by-step explanation grounded in how the course is actually taught and assessed. Read this whole study guide free, then take your hardest questions to Sia.
