[Info-vax] BOINC for VMS

[VAXman- at SendSpamHere.ORG]

In article <3086321.866.1331211956518.JavaMail.geo-discussion-forums at vbas10>, Neil Rieck <n.rieck at sympatico.ca> writes:
>Back in the day, there were techies at Compaq who made sure there were SETI=
>-at-home clients available for Alpha and you could see the reason why: Alph=
>a platforms beat every other platform out there. Around the time that Compa=
>q was acquired by HP, SETI-at-home support was ended. Also at that time, ot=
>her companies like SUN and IBM stepped up with software, hardware donations=
>, and money just so they could get their names associated with distributed =
>computing.
>
>( since then, IBM has developed their own on-ramp to BOINC called "world co=
>mmunity grid" http://en.wikipedia.org/wiki/World_community_grid )
>
>I, too, supported SETI-at-home but must admit that large conservative corpo=
>rations would rather be involved with more practical projects involving thi=
>ngs like: protein analysis, searching for new designer drugs to combat huma=
>n disease, improving climate models, etc. (I can hear it now: IBM will one =
>day announce "we cured cancer" while the others were just being greedy)
>
>There is nothing wrong with thinking about putting some old Alphas to work =
>on a BOINC project but I fear you will just be wasting electrical energy. T=
>he best results come from systems with DDR2 and DDR3 memory (most Alpha sys=
>tems have neither but I don't know about the EV7z generation), fast I/O bus=
>es, large CPU and memory caches (L1/L2/L3), and large amounts of memory. So=
>me science projects now require x86 Streaming Extensions (MMX/SSE2, etc)
>http://en.wikipedia.org/wiki/Streaming_SIMD_Extensions
>"None x86 systems" get around this requirement by employing many thousands =
>of streaming processors found in today's video cards. "x86 system" go this =
>route too. (sure you could use a CPU to generate sound but a sound card can=
> do it many orders of magnitude faster; it is the same thing with doing sci=
>ence on a graphics card)
>
>Now Folding-at-home does support a cool SMP client which would be perfect t=
>o try out on an Itanium box. I wonder if hobbyists have tried getting a BOI=
>NC client to run on one of these.
>
>(p.s. I am a huge fan of distributed processing and, like many others, have=
> set up a small folding-farm to do my bit. In the winter time the additiona=
>l heat from these systems causes me to burn a lesser amount of fossil fuels=
>.. In the summer time they drive out the dampness of my basement while makin=
>g the air-conditioner run a little more ofter. Oh well, you win some, you l=
>ose some)
>
>Neil Rieck
>Waterloo, Ontario, Canada.
>http://www3.sympatico.ca/n.rieck/docs/folding_at_home.html
>=20

So, what you are saying is that Watterson was a visionary:

http://upload.wikimedia.org/wikipedia/en/e/e0/Scientific_Progress_Goes_Boink_%28Calvin_and_Hobbes%29.jpg

-- 
VAXman- A Bored Certified VMS Kernel Mode Hacker    VAXman(at)TMESIS(dot)ORG

Well I speak to machines with the voice of humanity.