Wikipedia page for Zilla.app

I was reading the Wikipedia article on Xgrid (https://en.wikipedia.org/wiki/Xgrid) and I noticed that it needs to be updated.  In two different places the article states that the Zilla.app was distributed with OPENSTEP. It was also distributed with NEXTSTEP. I'm not sure which version of NEXTSTEP that Zilla was first released with, but maybe someone here knows. If you have experience with editing Wiki pages please consider updating that article. I would do it myself but I haven't had good luck with the survival rate of my edits. Another thing I noticed is that there doesn't appear to be a Wikipedia page for the Zilla app, one of the programs that started it all.

Hi @Nitro , I did not find the answer to your question, but while digging through my Usenet archives I found some (possibly) interesting slices of information:

QuoteFrom news Sun Jun 14 14:57:43 91
From: ********@acsu.buffalo.edu (********)
Subject: Zilla Unit is a changing....

Prior to the NeXTwarp9 leak a Zilla Unit was defined as 100 '40 NeXTs,
which was equivalent to some Cray model in raw power. How will the 88110
affect this definition?

unfortunately no answers on this Cray comparison topic.

QuoteFrom news Sun Feb 4 02:17:16 91
From: ********@remus.rutgers.edu (********)
Subject: Re: zilla info

In article <130610@midas.encore.com>, ********@encore.com (********) writes:
> I am looking for information on godzilla. Any information or pointers
> to sources would be appreciated.
>
> Thanks,
>       **** ****

Take a look at last week's (Wednesday's?) New York Times.  It had an
article called "Creating Computer Godzillas" with a cute little
cartoon.  It went something like, "Researchers at the computer maker
NeXT Inc. have developed a potent program that harnesses the unused
computing power in a network of computers..."

As for documentation, it's available online in the Zilla application,
located in /NextDeveloper/Demos/Zilla.app on NeXTstep 2.0 Extended
systems.

At NeXT, they've been factoring numbers like F15 - a number with about
10,000 digits.  Nasty stuff.

****** **********
Independent NeXT Developer

So Zilla was at minimum already available in NeXTstep 2.0 and there might be some information in the New York Times around 4th of February 1991.

That's really cool @nuss, thank you for finding that. Maybe we can put together enough information here to build a simple Zilla app page? I found a small amount of info here:

https://markmail.org/message/6mv2m6ufisdfxbyc#query:+page:1+mid:uyflgfg34suntrt3+state:results

Here's the NYTimes article.

https://www.nytimes.com/1991/01/30/business/business-technology-creating-computer-godzillas.html

I also ask ChatGPT about it.

Can you tell me about the Zilla App written for NeXTstep by Joshua Doenias and Richard Crandall?

Quote from: ChatGPTTo the best of my knowledge, there was indeed an application called "Zilla" that was developed by Joshua Doenias and Richard Crandall for the NeXTstep operating system. This application was designed to allow multiple NeXT computers to work together in parallel on tasks, such as performing large-scale mathematical calculations.

The Zilla software used a custom parallel processing architecture that was designed specifically for NeXT computers. It allowed multiple NeXT computers to be connected together over a high-speed network and to work together as a single, powerful computing cluster. The Zilla software was notable for its ability to divide complex mathematical problems into smaller, more manageable tasks that could be distributed among the individual computers in the cluster.

The Zilla application was widely used in scientific and academic research, particularly in fields such as physics, chemistry, and mathematics. It was also used by financial firms for complex financial modeling and analysis.

Overall, the Zilla application was an important development in the field of parallel computing and helped to pave the way for modern high-performance computing clusters.

Here's more Zilla info.

Idle computers hunt for big primes
https://www.newscientist.com/article/mg12917585-600-technology-idle-computers-hunt-for-big-primes/

Quote from:  NeXT Mathematica WhitepaperParallelization

In the summer 1990 issue of NeXT On Campus we described
a supercomputing network comprised of many NeXT com-
puters. There is now a demonstration application called Zilla,
which provides a graphical interface for such supercomput-
ing, with available machines appearing as icons within an
application window. It is now a straightforward matter to par-
allelize Mathematica problems via the Zilla application. You
may, for example, launch Mathematica kernels on many com-
puters with Zilla, each kernel being fed some unique set of
seed parameters, thereby multiplying the power of Mathemat-
ica one-hundred-fold.

If anyone has a copy of the summer 1990 issue of NeXT On Campus (or any other Zilla related information) please post it here. The NeXT On Campus newsletters were originally available at:

ftp://nova.cc.purdue.edu/pub/next/Newsletters (ftp://nova.cc.purdue.edu/pub/next/Newsletters)


Here's a project that's related to Zilla. :)

https://zillion.sourceforge.net/
https://web.archive.org/web/20070525190405/http://wiki.gnustep.org/index.php/Zillion.tool (https://web.archive.org/web/20070525190405/http://wiki.gnustep.org/index.php/Zillion.tool)

Quote from: Nitro on March 13, 2023, 05:37:21 PMIf anyone has a copy of the summer 1990 issue of NeXT On Campus (or any other Zilla related information) please post it here.

Not that specific issue, but a follow-up article from the spring 1991 issue is available at http://www.kevra.org/TheBestOfNext/page590/files/page590_3.pdf (http://www.kevra.org/TheBestOfNext/page590/files/page590_3.pdf) (p. 20).

That's great, thank you cuby!

Several issues of "NeXT on Campus" (including summer 1990) are available on Peanuts (CD).
If you have no easy access to the files, please ping me.

> ls -1R ./Peanuts/Documents/user-groups/OnCampus
./Peanuts/Documents/user-groups/OnCampus:
NOCFall90
NOCFall91
NOCSpring90
NOCSpring91
NOCSpring92
NOCSum90
NOCWinter91

./Peanuts/Documents/user-groups/OnCampus/NOCFall90:
images.tar.gz
NOCFall90Text.frame.gz
NOCFall90Text.ps.gz

./Peanuts/Documents/user-groups/OnCampus/NOCFall91:
Fall91_Images.tar.gz

./Peanuts/Documents/user-groups/OnCampus/NOCSpring90:
Spring90.frame.gz

./Peanuts/Documents/user-groups/OnCampus/NOCSpring91:
Spring91_Images.tar.gz
Spring91_Text.frame.gz
Spring91_Text.ps.gz

./Peanuts/Documents/user-groups/OnCampus/NOCSpring92:
00_InsideFrtCover.ps.gz
01_Spring_92TOC.ps.gz
02_SciComputing.ps.gz
05_NeXTedge.ps.gz
07_SpringFeature_A.ps.gz
07_SpringFeature_B.ps.gz
39_Rsrc_Infocus.ps.gz
42_Rsrc_Archives.ps.gz
44_Rsrc_3Pproducts.ps.gz
46_BackCover.ps.gz
99_Mailcard.ps.gz
README

./Peanuts/Documents/user-groups/OnCampus/NOCSum90:
OCNoImage.frame.gz
OCNoImage.ps.gz
OnCampus.frame.gz
OnCampus.ps.gz
README.gz

./Peanuts/Documents/user-groups/OnCampus/NOCWinter91:
Winter91_Images.tar.gz
Winter91_Text.frame.gz
Winter91_Text.ps.gz

That's perfect, thank you so much. I have the CD's.  8)

From the Xgrid man page (https://web.archive.org/web/20090131010748/https://developer.apple.com/documentation/Darwin/Reference/ManPages/man1/xgrid.1.html):

Quote from: Xgrid man pageHISTORY
    Xgrid's history can be traced back to Zilla, which was developed by NeXT in the late 80's and was the
    first clustering desktop program to make use of the "noninterventive screen saver" motif, a motif which
    is now commonplace and widely used in projects like Seti@Home.  Zilla won the 1991 national Computer-
    World-Smithsonian Award in the science category for this noninterventive, community-supercomputing par-
    adigm.

    Apple acquired the rights to Zilla in 1997, and later used that as the inspiration for the research
    project which became Xgrid.  Xgrid was publicly launched as a Technology Preview on January 6, 2004 at
    MacWorld San Francisco.

Here's some additional information about Zilla winning the Computerworld Smithsonian Award in 1991.

https://books.google.com/books?id=_djw_FNRO4cC&pg=PA91#v=onepage&q&f=false (https://books.google.com/books?id=_djw_FNRO4cC&pg=PA91#v=onepage&q&f=false)

There's a brief NeXT screenshot at the 13 second mark:

https://www.youtube.com/watch?v=RHfueIlkYpA
The Wikipedia page for the Computerworld Smithsonian Award is missing an entry for 1991.  :(

https://en.wikipedia.org/wiki/Computerworld_Smithsonian_Award (https://en.wikipedia.org/wiki/Computerworld_Smithsonian_Award)

The German Data Becker "Das große NeXT Buch" has a page on Zilla, too:



OCR and machine translated:

Quote6.1.17 Zilla an example of supercomputer capabilities


Supercomputers achieve peak processing speeds primarily through two factors: very high clock rates and the parallel use of many independent processors. While the clock speed of the NeXT cannot be easily increased, parallel processing can be exploited.

In the future, many NeXT computers will work together in smaller networks. This is not only due to the built-in interfaces and the ease of use, but it will be a matter of efficiency to ensure fast data transfer and information exchange in this way. Experience shows that the computers in a local network are rarely all in use at the same time. Often the user is busy with other tasks or not even at his workstation. At the latest in the evening and night hours the network is usually at a standstill.

Zilla is an experimental approach to concentrate the computing power of all available NeXTs within a network on one task. Of course, only problems that can be broken down into many parallel processes, each of which is then processed by a complete computer, are suitable for this. In the field of numerical analysis, the Scientific Computation Group of NeXT achieved processing speeds in the range of 100 to 1000 megaflops with Zilla. By way of comparison, a single NeXT can achieve 2.5 megaflops. So Zilla achieves speeds equivalent to a high-performance computer.

Zilla is only an experiment, but an executable one. The direction in which this project is heading is extremely interesting and promises to soon bear fruit on a professional level. There are enough areas of application. For example, the complex ray tracing for the creation of photorealistic computer graphics could be carried out at night on a network. Thanks to the high processing speed, the entire day's production would be available the next morning.

Zilla has a detailed help function that specifically describes the possibilities of the program. Among them are also two examples that can be tried out immediately in a network.

Hello NeXT Community:  https://www.lakeshorecsd.org/Page/5425     

Richard Crandall Using Mathematica probably invented the following on a NeXT and an educated guess is :
"discovered ways to double the speed of what are called convolutions -- essentially big multiplication operations" = Zilla?
Amazing as the footprint is still used today :)
#####
    There is a unique history to the arithmetic algorithms underlying the GIMPS project. The programs that found the recent big Mersenne finds are based on a special algorithm. In the early 1990's, the late Richard Crandall, Apple Distinguished Scientist, discovered ways to double the speed of what are called convolutions -- essentially big multiplication operations. The method is applicable not only to prime searching but other aspects of computation. During that work he also patented the Fast Elliptic Encryption system, now owned by Apple Computer, which uses Mersenne primes to quickly encrypt and decrypt messages. George Woltman implemented Crandall's algorithm in assembly language, thereby producing a prime-search program of unprecedented efficiency, and that work led to the successful GIMPS project.

 School teachers from elementary through high-school grades have used GIMPS to get their students excited about mathematics. Students who run the free software are contributing to mathematical research. Historically, searching for Mersenne primes has been used as a test for computer hardware. The free GIMPS program has identified hidden hardware problems in many PCs.


######

Oh and this gem :
https://www.wolfram.com/customer-stories/mathematica-helps-discover-the-largest-known-prime-number.html

For such arithmetic, the underlying algorithm—an "irrational base discrete weighted transform" (IBDWT)—was developed in the early 1990s by R. Crandall at Reed College and NeXT, Inc., using Mathematica as a prototyping environment. It was this algorithm that Woltman cast in the mid-1990s as an assembler variant, thus completing the prototyping/development cycle for which Mathematica is ideal.

The idea of the IBDWT is to expand giant numbers in an irrational base. One then applies a discrete weighted transform—a variant of the classical FFT—to multiply or square numbers in this base with unprecedented speed. As Crandall puts it: "Mathematica is an ideal environment for the requisite mixing of symbolics and numerics and the kind of interactive exploration that leads to such a new, nonstandard algorithm."

$$$ This prime is 2,098,960 decimal digits long and therefore qualifies for a $50,000 prize from the Electronic Frontier Foundation. One interesting sidelight is that in the early 1990s Mathematica played an important part in making the discovery possible.

   

You Got the Power | WIRED

https://www.wired.com/2000/08/comcomp/

Doing this sort of distributed computation on a global scale wasn't possible in the early 1980s. The Net's population was measured in the thousands, and the bandwidth was puny. But by 1989, Richard Crandall, now Distinguished Scientist at Apple (and once my roommate at Reed College), started networking NeXT computers to find, factor, and test gargantuan prime numbers.

"Community supercomputing occurred to me one day at NeXT engineering headquarters," Crandall recalls. "I thought we ought to make these machines do what they were designed to do, which is to work when we humans are not working. Machines have no business sleeping."

Crandall installed software that allowed idle NeXT machines to perform computations, combining their efforts across the network. He called this software Godzilla, but after a legal inquiry from the company that owned the rights to the movie character, he renamed it Zilla. Crandall put Zilla to work on huge prime numbers, which are crucial in cryptography. It was then used to test a new encryption scheme at NeXT - a scheme now employed at Apple, which acquired NeXT. In 1991, Zilla won the Computerworld Smithsonian award for science.

Later, Crandall and several colleagues used distributed processing to complete the deepest computation ever performed, asking the question: Is the 24th Fermat number (which has more than 5 million digits) prime? "It took 10**17 machine operations - 100 quadrillion," Crandall says proudly. "With that level of computational effort, you can create a full-length movie. In fact, that's about the same number of operations Pixar required to render A Bug's Life."

Just a quick note to confirm that the first version of Zilla was indeed in NS release 2.0 (along with many other demos that would later be relegated to the Education Sampler CD-ROM) and is not included with 1.0a. The copyright date further supports this, as 1.0a was released in 1989.