Threads

A thread is a unit of execution that lives inside a process. While a process has its own isolated address space, multiple threads within the same process share certain resources — making threads lighter and faster to create and switch between than full processes. [OS Course Notes — Weeks 3 & 5.md]

[OS Course Notes — Weeks 3 & 5.md]

Each thread runs independently but operates inside the same process memory. [OS Course Notes — Weeks 3 & 5.md]

• Responsiveness — one thread can keep the UI running while another does heavy work.
• Resource sharing — threads share memory cheaply without needing IPC mechanisms.
• Economy — creating and context-switching a thread is cheaper than doing so for a whole process.
• Multicore utilization — threads can run truly in parallel on multiple CPU cores.

[OS Course Notes — Weeks 3 & 5.md]

Because threads share the heap and data segment, concurrent access to shared variables can produce race conditions — bugs where the outcome depends on the exact interleaving of thread execution. Proper synchronization (locks, semaphores, monitors) is required to prevent them. [OS Course Notes — Weeks 3 & 5.md]

⚠️ Race conditions are a key risk of multithreading. See the Key Concepts Glossary for a one-line definition.

• Threads live inside a process, which is tracked by a PCB (Process Control Block).
• The CPU Scheduling algorithms (FCFS, SJF, Round Robin, Priority) generally operate at the thread/process level — the scheduler decides which thread gets the CPU next.
• Context switching between threads of the same process is cheaper than switching between different processes because the memory space does not change. [OS Course Notes — Weeks 3 & 5.md]

Threads are covered in the Week 3 material and are in scope for the midterm (Wed 22 April 2026). Review how threads differ from processes, what they share, and why synchronization is needed. See Exam Prep — Midterm for a full checklist. [OS Course Notes — Weeks 3 & 5.md]