Computer Hardware Flashcards
CPUs and device _______ are connected through a common bus to a _______ memory
controllers, shared
What is a uniprocessor?
one processor with a single core
High _____ is important for a uniprocessor
frequency
Pre early 2000’s, ______ doubled in processors every 18 months
frequency
Post 2005, ______ ____ doubled in processors every 18 months
core count
What is a multicore processor?
one processor with multiple cores
A multicore processor contains both shared and core-private ______
caches
________ allows a core to execute more than one thread
hyper-threading
Multiple threads on one core still share caches, ______, etc
registers
Extra threads from hyper threading show up to the OS as additional _____ or ___
cores, CPUs
Mainstream servers using a multisocket motherboard usually have - sockets
2, 4
What does NUMA stand for?
Non-Uniform Memory Access
What is a NUMA node?
A processor and its local memory
Accessing _____ memory is faster than _____ memory in other NUMA nodes. This is known as the NUMA effect.
local, remote
An _________ is a communication medium that connect processors to processors or processors to memory in NUMA nodes
Interconnect
What does numactl do?
Control NUMA policy for processes or shared memory
What does htop do?
Interactive process viewer
What is Moore’s law?
the principle that the speed and capability of computers can be expected to double every two years, as a result of increases in the number of transistors a microchip can contain
Parallelism is offered by ______ ______ and ________
Hyper threading, Multi-core and multi-socket
What are the seven layers in the memory/storage hierarchy. (smallest to biggest)
Registers, L1, L2, L3, Main memory, Persistent memory, Flash/disk
The lower levels of the storage hierarchy contains the ______ of higher levels
contents
____, __, __, __ are examples of volatile SRAM
Registers, L1, L2, L3
Volatile DRAM is ______ memory
main
The _______ ______ points to the next instruction to execute
program counter
Instructions and data moved between cache and memory in ______ ____ units
cache line
Describe the fetch, decode, execute cycle in a CPU
- Fetch next instruction: Instruction contains op-code and possibly data
- Decode op-code
- Execute op-code (using data if necessary)