Systolic arrays (Kung & Leiserson, 1978) take their name from cardiac systole — Kung's 1982 paper likens each processor's rhythmic data-pumping to a heartbeat

H.T. Kung and Charles Leiserson introduced the systolic array — a grid of simple processing elements that rhythmically compute and pass data to their neighbors — in 1978. The name comes from cardiac physiology: systole is the heart's rhythmic contraction, and Kung's own 1982 paper, "Why Systolic Architectures?" (Computer 15(1):37–46), is reported to explain the choice as an analogy — "the function of a processor is analogous to that of the heart. Every processor regularly pumps data in and out."

The design idea itself is straightforward and well-attested: each processing element performs a small computation and passes its result to a neighbor on a fixed rhythm, so data flows through the array the way blood flows through a circulatory system, without needing to return to a central memory between steps. This structural description is uncontested in the secondary literature and clears the sourcing floor for a definitional/etymological claim.

The specific wording attributed to Kung — the exact phrase comparing processors to a pumping heart — is a different matter. It was recorded in this promotion from an encyclopedic secondary account (Wikipedia), not from a direct read of Kung's 1982 Computer article, so it is held here as an attributed quote rather than a verified primary one. See claim-tpu-matrix-unit-called-heart-of-the-tpu for the closing half of the metaphor: three decades later, Google's engineers independently called the same structure "the heart of the TPU."