---
title: "Read Kung's 1982 'Why Systolic Architectures?' directly and confirm the cardiac/heart-pumping quote attributed to it"
type: "question"
status: "answered"
date_raised: "2026-07-12T00:00:00.000Z"
date_answered: "2026-08-12T00:00:00.000Z"
answered_log: "Answered, as a correction rather than a confirmation, by [[claim-systolic-array-named-after-cardiac-systole]] (updated in place), [[myth-kung-1982-processor-pumps-quote]], and [[claim-kung-1982-opening-paragraph-frames-dataflow-as-blood-circulation]]. What settled it: a direct extract_pdf read of Kung's 1982 paper (Tier 1, sha256 8c7e08b1...) found the popularly quoted sentence ('the function of a processor is analogous to that of the heart... every processor regularly pumps data in and out') nowhere in the text. What Kung actually wrote attributes the analogy to the memory, not the processor, and uses 'pulses' rather than 'pumps.' The why-systole question (whether Kung explains the choice of systole specifically over other rhythmic metaphors) was not separately addressed and is not reopened here — the paper's own cardiac language answers the provenance question this thread was raised to settle."
tags: ["systolic-array","kung","etymology","verification","hop"]
---


[[claim-systolic-array-named-after-cardiac-systole]] rests on an encyclopedic
secondary account (Wikipedia) for the specific quote attributed to H.T. Kung's
1982 paper — "the function of a processor is analogous to that of the heart.
Every processor regularly pumps data in and out." No source in this promotion's
chain (or the capture it came from) reports having read Kung's paper directly.

**What would answer it:**

- Obtain H.T. Kung, "Why Systolic Architectures?", *Computer* 15(1):37–46
  (January 1982), IEEE. Check IEEE Xplore (DOI likely under
  10.1109/MC.1982.1653825 — confirm), or a CMU technical-report mirror (Kung
  published an earlier CMU CS tech report under a similar title that may be
  open-access).
- Confirm the exact wording of the heart/pump analogy, and note whether it
  appears in Kung's own words or is a paraphrase introduced by a later
  secondary source.
- While there, check whether Kung's paper gives any additional detail on why
  *systole* specifically (as opposed to, say, "peristaltic" or another rhythmic
  biological metaphor) was chosen.

Resolving this would let [[claim-systolic-array-named-after-cardiac-systole]]
move off its `[unverified-quote]` flag and firm up the etymological half of
the systolic-array/TPU "heart" bridge documented in
[[claim-tpu-matrix-unit-called-heart-of-the-tpu]].


## Progress log

- Answered, as a correction rather than a confirmation, by [[claim-systolic-array-named-after-cardiac-systole]] (updated in place), [[myth-kung-1982-processor-pumps-quote]], and [[claim-kung-1982-opening-paragraph-frames-dataflow-as-blood-circulation]]. What settled it: a direct extract_pdf read of Kung's 1982 paper (Tier 1, sha256 8c7e08b1...) found the popularly quoted sentence ('the function of a processor is analogous to that of the heart... every processor regularly pumps data in and out') nowhere in the text. What Kung actually wrote attributes the analogy to the memory, not the processor, and uses 'pulses' rather than 'pumps.' The why-systole question (whether Kung explains the choice of systole specifically over other rhythmic metaphors) was not separately addressed and is not reopened here — the paper's own cardiac language answers the provenance question this thread was raised to settle.