“From security to performance to GPU programming..."
summary
Shows that secure allocator wipes pages and locks them to keep secrets out of swap, demonstrates pool allocators (boost PMR and custom pools) that stop fragmentation — one demo goes from gigabytes of heap growth to single-megabyte footprints, and TBB’s size-segregated per-thread allocator avoids contention under concurrency. Explains using boost::interprocess plus offset/fancy pointers for shared memory and scoped_allocator_adaptor to propagate allocator state into nested containers, and notes GPU/pinned-memory allocators require device-aware accessors.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 0:31What do alligators do
- 1:05Two ways to use alligators
- 2:13Overview
- 2:36Secure Alligator
- 3:35Secure Alligator Demo
- 4:34Secure Stream
- 6:39Memory fragmentation
- 7:16Pool alligators
- 8:50Memory alligators
- 9:47Aggregate alligators
- 11:35Shared Memory
- 13:01Vector
- 14:58Nested Containers
- 18:55Pin memory
- 20:37Alligators
- 21:26Questions
- 25:46Linux
related talks
01h 25mTaming Dynamic Memory - An Introduction to Custom Allocators
01h 03mMost Malleable Memory Management Method
58mLocal (Arena) Memory Allocators Part 1
49mMost Malleable Memory Management Method in C++
01h 13mValue Proposition: Allocator-Aware (AA) Software
01h 01mGetting Allocators out of Our Way
01h 08mLocal (Arena) Allocators Part II
01h 01mRecipes for Reducing Allocations by C++ Containers