Scribd
Upload
9 views2 pages

Fixed Parallel Computing Notes

The document provides comprehensive notes on parallel computing, explaining the division of workload among multiple processors to enhance problem-solving speed. It distinguishes between concurrency and parallelism, discusses the limitations of single-core CPUs, and highlights the advantages of multi-core processors. Additionally, it addresses the cache coherence problem and outlines cache coherence protocols like MESI and MOESI.

Uploaded by

FACTS And KNOWLEDGE
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views2 pages

Fixed Parallel Computing Notes

The document provides comprehensive notes on parallel computing, explaining the division of workload among multiple processors to enhance problem-solving speed. It distinguishes between concurrency and parallelism, discusses the limitations of single-core CPUs, and highlights the advantages of multi-core processors. Additionally, it addresses the cache coherence problem and outlines cache coherence protocols like MESI and MOESI.

Uploaded by

FACTS And KNOWLEDGE
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Comprehensive Notes: Parallel Computing (CS 526)

Based on Lectures from Muhammad Nadeem Nadir

Department of Computer Science, The University of Lahore

Slide 1: Introduction to Parallel Computing


Parallel computing means using multiple processors or computers to solve a problem faster. Instead
of one processor doing all the work, the workload is divided.

Example: Instead of one person making an entire pizza, one person prepares the dough, another
adds toppings, and a third bakes it.

Slide 2: Concurrency vs. Parallelism


Concurrency: Multiple tasks execute in overlapping time periods but not necessarily at the same
time.
Example: A person eating and talking-they pause between bites to talk.

Parallelism: Multiple tasks execute at the same time.


Example: A chef cooking while talking on the phone, both actions happen simultaneously.

Slide 5: Growth in CPU Transistor Count


Moore's Law states that the number of transistors doubles every 18 months, improving performance.
However, increasing clock speed causes excessive heat and power consumption.

Example: A car engine cannot run indefinitely faster because of overheating and fuel limitations.

Slide 6: Single-core CPU Limitations


Single-core CPUs have three major problems:
- Power Consumption: Higher clock speeds require more power.
- Heat Dissipation: More power generates more heat.
- Limited Memory: A single-core CPU has limited memory access.

Example: A single cashier handling customers vs. multiple cashiers for efficiency.

Slide 8: Multi-core vs. Single-core Performance


Single-core processors rely on higher clock speeds, which increase heat and power consumption.
Multi-core processors distribute the workload.

Example: A group of people solving a puzzle together is faster than one person solving it alone.

Slide 10: The Cache Coherence Problem


Cache coherence ensures all processor cores see the same data.

Example: If a teacher updates a class schedule but only some students get the new version, others
still follow the old one.

Slide 11: Cache Coherence Protocols (MESI & MOESI)


MESI Protocol states that cache blocks can be in one of four states:
- M (Modified): Updated locally but not in main memory.
- E (Exclusive): Present only in one cache, matches main memory.
- S (Shared): Available in multiple caches.
- I (Invalid): Not valid.

MOESI adds an 'Owned' state to allow direct sharing between caches.


Example: A library with multiple copies of a book-some are borrowed (Modified), some are reserved
(Exclusive), and some are on shelves (Shared).

You might also like