Average Memory Access Time
Effective access time from hit time, miss rate, and penalty.
Formula
Live Calculator
MembersVariables
Example
Cache performance metric
How it works
Average Memory Access Time (AMAT) gives the expected time for a memory reference when some accesses hit quickly and others miss and require extra work. It comes from weighting the fast path and the slow path: every access pays the hit time H, and only the fraction M of accesses that miss pay an additional penalty P, so the average becomes H + M × P. This works because the miss penalty is an extra cost added only on misses, not on every access.
When to use it
Use this formula when evaluating cache, TLB, or memory-system performance in computer architecture. It is especially useful for comparing design choices such as changing cache size, associativity, block size, or replacement policy.
Worked example, step by step
Suppose a processor has a cache hit time H = 2 ns, a miss rate M = 4% = 0.04, and a miss penalty P = 60 ns. AMAT = H + M × P AMAT = 2 + 0.04 × 60 AMAT = 2 + 2.4 AMAT = 4.4 ns So the average memory access time is 4.4 ns per access.
Common mistakes to avoid
- Using the miss rate as a percentage number 4 instead of converting it to 0.04
- Adding the miss penalty to every access instead of only to misses
- Confusing hit time with miss penalty, which are different parts of the memory access cost
Did You Know?
A high cache-hit rate is what makes computers feel fast, since main memory is agonisingly slow by comparison.
Share this formula
More in Computer Science
View allShannon Entropy
AdvancedAverage information content of a random source (bits).
Master Theorem (form)
AdvancedAnalyzes running time of divide-and-conquer algorithms.
Hamming Distance
AdvancedNumber of positions at which two strings differ.
Bits to Represent n
BasicMinimum number of bits to represent n distinct values.
Binary to Decimal
BasicConverts a binary number to decimal.
Nodes in a Full Binary Tree
IntermediateTotal nodes in a perfect binary tree of height h.