Chris Higgins
Richard Feynman and the Connection Machine
by Chris Higgins - June 4, 2008 - 5:32 PM

Connection Machine CM-5 FrostburgDanny Hillis is best known for his Connection Machine, a massively parallel supercomputer that led to breakthroughs in computational science and parallel computing. (It also was the subject of Hillis’s book The Connection Machine, which was a highly technical but fascinating version of his thesis paper on parallel computing. If you want the only halfway mind-blowing version, check out The Pattern on the Stone.) Hillis later led R&D at Walt Disney Imagineering, but in my book the Connection Machine was his awesomest project.

The Connection Machine was actually a series of supercomputers, labeled CM-1, CM-2, and so on. A CM-5 (codenamed FROSTBURG) is pictured above left — originally installed at the National Security Agency, it was used to break codes and is now on display at the National Cryptologic Museum. Looking surprisingly like a classic “movie computer” (one did appear in Jurassic Park), the CM-5 was covered in blinky lights that communicated the status of various processing nodes, and could be used in diagnostics. (So they’re useful for something after all….)

Anyway, the point of this blog is that Hillis wrote an essay for Physics Today about physicist Richard Feynman’s involvement in the Connection Machine’s development — and now the article is available online via The Long Now Foundation. Hillis’s article reveals how Feynman was instrumental in designing the Connection Machine’s router, which was key in distributing communications within the massive machine. From the article:

Richard’s interest in computing went back to his days at Los Alamos, where he supervised the “computers,” that is, the people who operated the mechanical calculators. There he was instrumental in setting up some of the first plug-programmable tabulating machines for physical simulation….

The router of the Connection Machine was the part of the hardware that allowed the processors to communicate. It was a complicated device; by comparison, the processors themselves were simple. Connecting a separate communication wire between each pair of processors was impractical since a million processors would require $10^{12]$ wires. Instead, we planned to connect the processors in a 20-dimensional hypercube so that each processor would only need to talk to 20 others directly. Because many processors had to communicate simultaneously, many messages would contend for the same wires. The router’s job was to find a free path through this 20-dimensional traffic jam or, if it couldn’t, to hold onto the message in a buffer until a path became free. Our question to Richard Feynman was whether we had allowed enough buffers for the router to operate efficiently.

Read the rest if you’re interested in Feynman, math, blinky lights, or just crazy engineering projects. See also: more on the Connection Machine, more on Danny Hillis.

Click here to get a Risk-Free issue of mental_floss magazine
Comments (4)
  1. I’ve read a few of Feynman’s books / books about Feynman. The math and science are usually way over my head, except when they give examples of his ability to break things down into wonderfully concise, yet not condescending, explanations. I think he must have been a real pain in the a$$ at times, and I think I would have absolutely loved the guy.

  2. The article mentions Feynman’s setting up some of the first “plug-programmable tabulating machines”. His work was likely influence by Wallace Eckert at Columbia. Wallace was an astronomer responsible for creating tables of the movement of heavenly bodies. He initially used young women with adding machines. He kept pushing the envelope, using punch cards and beyond. You can click on the link from my name to read an article on Eckert and computing.

  3. They had a Connection Machine at FSU when I was there during high school. I asked the grad student showing us around if the blinking lights indicated anything about processor activity. He said you could program them to, but mostly they were there “just to impress the generals.”

  4. Einstein’s Nemesis: DI Her Eclipsing Binary Stars Solution
    The problem that the 100,000 PHD Physicists could not solve

    This is the solution to the “Quarter of a century” Smithsonian-NASA Posted motion puzzle that Einstein and the 100,000 space-time physicists including 109 years of Nobel prize winner physics and physicists and 400 years of astronomy and Astrophysicists could not solve and solved here and dedicated to Drs Edward Guinan and Frank Maloney
    Of Villanova University Pennsylvania who posted this motion puzzle and started the search collections of stars with motion that can not be explained by any published physics
    For 350 years Physicists Astrophysicists and Mathematicians and all others including Newton and Kepler themselves missed the time-dependent Newton’s equation and time dependent Kepler’s equation that accounts for Quantum – relativistic effects and it explains these effects as visual effects. Here it is

    Universal- Mechanics

    All there is in the Universe is objects of mass m moving in space (x, y, z) at a location
    r = r (x, y, z). The state of any object in the Universe can be expressed as the product

    S = m r; State = mass x location

    P = d S/d t = m (d r/dt) + (dm/dt) r = Total moment

    = change of location + change of mass

    = m v + m’ r; v = velocity = d r/d t; m’ = mass change rate

    F = d P/d t = d²S/dt² = Force = m (d²r/dt²) +2(dm/d t) (d r/d t) + (d²m/dt²) r

    = m γ + 2m’v +m”r; γ = acceleration; m” = mass acceleration rate

    In polar coordinates system

    r = r r(1) ;v = r’ r(1) + r θ’ θ(1) ; γ = (r” – rθ’²)r(1) + (2r’θ’ + rθ”)θ(1)

    F = m[(r"-rθ'²)r(1) + (2r'θ' + rθ")θ(1)] + 2m’[r'r(1) + rθ'θ(1)] + (m”r) r(1)

    F = [d²(m r)/dt² - (m r)θ'²]r(1) + (1/mr)[d(m²r²θ')/d t]θ(1) = [-GmM/r²]r(1)

    d² (m r)/dt² – (m r) θ’² = -GmM/r²; d (m²r²θ’)/d t = 0

    Let m =constant: M=constant

    d²r/dt² – r θ’²=-GM/r² —— I

    d(r²θ’)/d t = 0 —————–II
    r²θ’=h = constant ————– II
    r = 1/u; r’ = -u’/u² = – r²u’ = – r²θ’(d u/d θ) = -h (d u/d θ)
    d (r²θ’)/d t = 2rr’θ’ + r²θ” = 0 r” = – h d/d t (du/d θ) = – h θ’(d²u/d θ²) = – (h²/r²)(d²u/dθ²)
    [- (h²/r²) (d²u/dθ²)] – r [(h/r²)²] = -GM/r²
    2(r’/r) = – (θ”/θ’) = 2[λ + ỉ ω (t)] – h²u² (d²u/dθ²) – h²u³ = -GMu²
    d²u/dθ² + u = GM/h²
    r(θ, t) = r (θ, 0) Exp [λ + ỉ ω (t)] u(θ,0) = GM/h² + Acosθ; r (θ, 0) = 1/(GM/h² + Acosθ)
    r ( θ, 0) = h²/GM/[1 + (Ah²/Gm)cosθ]
    r(θ,0) = a(1-ε²)/(1+εcosθ) ; h²/GM = a(1-ε²); ε = Ah²/GM

    r(0,t)= Exp[λ(r) + ỉ ω (r)]t; Exp = Exponential

    r = r(θ , t)=r(θ,0)r(0,t)=[a(1-ε²)/(1+εcosθ)]{Exp[λ(r) + ì ω(r)]t} Nahhas’ Solution

    If λ(r) ≈ 0; then:

    r (θ, t) = [(1-ε²)/(1+εcosθ)]{Exp[ỉ ω(r)t]

    θ’(r, t) = θ’[r(θ,0), 0] Exp{-2ỉ[ω(r)t]}

    h = 2π a b/T; b=a√ (1-ε²); a = mean distance value; ε = eccentricity
    h = 2πa²√ (1-ε²); r (0, 0) = a (1-ε)

    θ’ (0,0) = h/r²(0,0) = 2Ï€[√(1-ε²)]/T(1-ε)²
    θ’ (0,t) = θ’(0,0)Exp(-2ỉwt)={2Ï€[√(1-ε²)]/T(1-ε)²} Exp (-2iwt)

    θ’(0,t) = θ’(0,0) [cosine 2(wt) - ỉ sine 2(wt)] = θ’(0,0) [1- 2sine² (wt) - ỉ sin 2(wt)]
    θ’(0,t) = θ’(0,t)(x) + θ’(0,t)(y); θ’(0,t)(x) = θ’(0,0)[ 1- 2sine² (wt)]
    θ’(0,t)(x) – θ’(0,0) = – 2θ’(0,0)sine²(wt) = – 2θ’(0,0)(v/c)² v/c=sine wt; c=light speed

    Δ θ’ = [θ'(0, t) - θ'(0, 0)] = -4Ï€ {[√ (1-ε) ²]/T (1-ε) ²} (v/c) ²} radians/second
    {(180/Ï€=degrees) x (36526=century)

    Δ θ’ = [-720x36526/ T (days)] {[√ (1-ε) ²]/ (1-ε) ²}(v/c) = 1.04°/century

    This is the T-Rex equation that is going to demolished Einstein’s space-jail of time

    The circumference of an ellipse: 2Ï€a (1 – ε²/4 + 3/16(ε²)²—) ≈ 2Ï€a (1-ε²/4); R =a (1-ε²/4)
    v (m) = √ [GM²/ (m + M) a (1-ε²/4)] ≈ √ [GM/a (1-ε²/4)]; m

Comment

commenting policy