Introduction to RAM

• Load → provide the address and get the corresponding data
• Write → Address + Data → update the state of the memory

Modern Technology Options and the Chart

• SRAM: super fast, can be on chip, next to the cpu pipeline (volatile - data is lost when the device is turned off)
• DRAM: denser than sram, offchip (volatile)
• Flash: super dense and cell wears out with each write
• Disk: tiny magnetic substrate with the highest density

<aside> 💡 On chip caches are cache for the memory and are managed by the hardware.

Registers are a cache for the on chip cache and is managed by the compiler.

Main memory is a cache for the disk and is managed by the operating system.

</aside>

Conducting a Memory Read Without Caches

1. CPU places address A (rax) on the memory bus
2. Main memory reads A from the memory bus, retrieves word val, and places it on the bus.
3. Main memory reads data word val from the bus and stores it at address A.

Locality

• Principle of Locality: Programs tend to use data and instructions with addresses near or equal to those they have used recently
• Temporal Locality: Recently referenced items are likely to be referenced again in the near future
  • Knowing that a loop will have a repeated action (reference sum or end condition)
  • only has to do with branching and loops
• Spatial Locality: Items with nearby addresses tend to be referenced close together in time
  • reference array elements in succession when looping through an array