Главная
Study mode:
on
1
Intro
2
Ground Rules
3
Information - Integrity Matters
4
Information - Ordering Matters
5
Obligatory Thread Ordering Example
6
Thread Communication
7
First-On First-Out (FIFO) Queue
8
Chunk of Memory Types
9
Commonalities
10
Quiz Time
11
Coherence Network + Coherence
12
Experiment
13
DDR to Core
14
Prefetch Maybe?
15
Blue Pill of Virtual Memory
16
Move Pages (remember NUMA)
17
Darth Mutex
18
The Dark Side of Mutex
19
Alignment
20
Two socket system, Producer - Consumer Pair
Description:
Explore the intricacies of First-In-First-Out (FIFO) queues and their impact on thread communication in parallel programming during this C++Now 2017 conference talk. Delve into the micro-architectural implications of FIFO implementations, examining various open-source examples and their behavior on different hardware. Follow along as the speaker dissects code and assembly, revealing the inner workings of processor cores. Gain valuable insights into how FIFOs affect performance, covering topics such as thread ordering, coherence networks, prefetching, virtual memory, NUMA considerations, and the challenges of mutex usage. Learn to optimize thread communication for improved parallelism and scalability in your C++ programs.

Good FIFOs Make Good Thread Neighbors

CppNow
Add to list