The comparison across Registers, Main Memory, and Disk Storage is best quantified by the punishing reality of latency, the time delay between the CPU requesting data and its arrival. This difference is not linear; it is exponential, defining a strict performance hierarchy where the cost of a data retrieval error (a cache miss) increases astronomically with each lower tier. Registers, being located directly on the CPU die, possess access latency measured in fractions of a nanosecond, typically a single clock cycle. This near-zero latency allows for uninterrupted instruction execution.

Moving to Main Memory (DRAM), the latency cost jumps hundreds of times higher, typically requiring fifty to over one hundred nanoseconds. In terms of processor cycles, this latency translates to the CPU being stalled for hundreds of cycles, creating the severe performance limitation known as the Memory Wall. The processor is forced to rely heavily on its cache hierarchy and out-of-order execution to mask this significant access time.

The cost escalates catastrophically when data must be fetched from Secondary Storage (Disk). For an SSD, access time is measured in tens of thousands of nanoseconds (microseconds), while for an older mechanical Hard Disk Drive (HDD) that requires physical seek time, latency can exceed ten million nanoseconds (tens of milliseconds). This magnitude of latency means that a single disk access stalls the CPU for millions of clock cycles, effectively halting the computational work. The massive difference in access time underscores why operating systems invest immense effort in virtual memory and paging strategies, minimizing disk access and ensuring that data needed for immediate execution resides as high up the memory hierarchy as possible.