1/5
Looks like no tags are added yet.
Latency
Time required for one token to pass through system from tart to end. Represented by Tc
Throughput
The number of tokens that can be produced per unit of time
Latency1=Tc1
Spatial parallelism
Multiple copies of the hardware are provided so that multiple tasks can be done at the same time
Temporal parallelism OR pipelining
A task is broken into stages, like an assembly line, and hence multiple tasks can be spread across the stages
Throughput in non-ideal
1/L
Throughput in an ideal spatially parallel system with n hardware copies
N/L