Internet-based Distributed Computing Projects
Home   News and Articles   Active Projects   Computing Platforms  
Upcoming Projects   Recently Completed Projects   Past Projects  
Tools   Development Platforms   Conferences   Parody and Fun Sites   Links   Politics
Teams   Books and Journals

Recently Completed Distributed Computing Projects

Project InformationCategoryCompletion DateProject DurationTotal Number of Participants/
Genetic TSP used a Java application that ran through a user's web browser and used genetic algorithms to solve a Traveling Salesman Problem (in a TSP, a salesman must find the shortest route in which he/she can visit each a set of cities once and return to his/her starting city). This project attempted to solve a problem of 15,122 cities of Germany. As of December, 2003, the current record-holders of this problem were Princeton University and Rice University. Puzzles/Games December, 2003 2 years unknown/unknown
GRISK searched for K-optimal lattice rules. It completed a Delta=7 project on November 15, 2000, and a Delta=8 project on December 21, 2001. It began a Delta=11 project on December 21, 2001, and completed 22% of it when the GRISK project ended. Mathematics October, 2003 at least 3 years 604/unknown
emirp The Distributed Emirp Project searched for Emirps, prime numbers whose digits, when reversed, are also a prime number (for example, 13 and 31 are Emirps). The project processed 197 blocks.

Join a discussion forum about the project.

Mathematics August, 2003 2 months 65/unknown
Russia flag MD@home studied the properties of oligopeptides. Note that the site is written in Russian, but babelfish provides a reasonable English translation U. K. flag of it. The project simulated the thermal agitation of molecules within large proteins in order to understand how the design of a protein defines the protein's properties and behavior (and the properties and behaviors of the protein's component parts (oligopeptides). The client software simulated the thermal agitation of a molecule and calculated the special characteristics of that thermal agitation. The project processed 32,000 work units. Life Sciences August, 2003 9 months 1,470/unknown
The NEO Project used the NEO-c (Network Exchange Operation for Charity) platform to participate in various computing challenges and projects, and would have donated any winnings to the charities specified by its users. The project was the first to use Microsoft's .NET architecture. The project ended unexpectedly before any results were found for the challenges below:

The project's first challenge was the RSA 576-bit factoring challenge. The project's first attemp to solve the challenge was through random guesses. For this attemp, volunteers generated 39,033,522 packets (.154885015296E+15 keys checked) for the project. The attempt ended on January 10, 2003. The project's second attempt, Phase 2, would have used a General Number Field Sieve (GNFS) algorithm.

The second challenge was the MD5 project. The MD5 encryption algorithm is widely used in business, secure websites, Unix systems, and the Internet. The challenge would have demonstrated MD5's vulnerability, forcing people who use it to develop a better algorithm. This project began on January 30, 2003, and was stopped on May 20, 2003, due to a lack of interest from the MD5 developers (who were outside of the NEO project). The challenge tested at least 0.53% of the MD5 keyspace.

The third challenge was the Tellurium project, a physics project in which space-time geometry, specifically Isaac Newton's Equivalence Principle (see a simpler explanation), would be tested with the handedness or chirality property of matter. The principle has never been tested this way: if the test had caused it to fail, then Albert Einstein's General Relativity theory would be shown to be subtly incorrect. This challenge began on May 9, 2003, but the alpha client for the challenge was never publicly released.

The fourth challenge was World TSP, a study of the Traveling Salesman Problem. This challenge attempted to find the shortest route which visits all 1,904,711 populated cities and towns on Earth. "The current best lower bound on the length of a tour for the World TSP [was] 7,510,666,782 (Kilometers)." This bound was established on June 18, 2002. The challenge used an evolving artificial intelligence algorithm to attempt to beat that bound. With 97,820 total routes completed, the shortest route discovered was 13,802,932,609 Kilometers.

Cryptography, Science July 31, 2003 10 months ~50,000/~50,000
Operation Project X was a distributed effort to solve the Xbox Linux Project, a challenge to crack the 2048-bit RSA private encryption key Microsoft uses to sign Xbox media. If this key was discovered, Linux could be run on the Xbox without modifying the Xbox hardware. The client used Microsoft's .NET architecture, and was available for many platforms, including Xbox. Over 351.3 trillion keys were tested, but the project ended unexpectedly before the key was found.

Source code for the project is available here and here for anyone who would like to continue the project.

Listen to an April 26, 2003 CBC Radio interview (in RealAudio format) with some of the project coordinators.

Cryptography July 31, 2003 4 months ~4,000/unknown
The search for Wieferich prime numbers looked for numbers of the form ap-1 = 1 (mod p2) for a = 2 or 3. The only two known Wieferich primes are 1,093 and 3,511 and there are no other Wieferich primes less than 2 * 1014. The project extended the search limit to 1.25 * 1015, but did not find any new Wieferich primes. 131,429 total ranges (37,424,648,092,395 primes) were checked at an average speed of 621,457 primes per second. 131 near-misses were found. Mathematics June 19, 2003 14 months 304/unknown
dchess logo The DIstributed Chess Project tried to create better chess-playing artificial neural networks. The project software implemented a genetic algorithm "to train multi-layer-perceptron neural networks on sets of chess positions with known best continuations (e.g. endgame studies, mate in n moves, white to move and win, ...)." The software was available as a screensaver or command-line client. It allowed the user to "view the status of the evolution any time in terms of computation time, current generation, current best fitness and population diversity" and to "modify the key parameters of the algorithm before a new task starts (e.g. number of generations, population size, number of hidden layers, number of nodes per hidden layer, ...)" if the user wanted to take an active role in shaping its chess-playing neural networks. Volunteers contributed 16 years 57 days of computing time to the project. Puzzles/Games April, 2003 10 months 658/unknown
Cell Computing Japanese flag Cell Computing was a non-profit project sponsored by NTT Data Corporation, with two sub-projects: finding disease-causing genes, and finding good materials for creating optical microprocessors. Note that this site is written completely in Japanese, but the text translates to English reasonably well in the babelfish translation. Tetsuya Matsushita wrote an excellent English translation U.K. flag of the major information about the project and provided screenshots with translations of important information and buttons on each screen. The project was developed on the United Devices distributed computing platform. By the end of the project, 12,000 PCs contributed results to one or both of the sub-projects. A paper detailing the results of the sub-projects should be available soon.

The sub-projects:

  • BOLERO (Bio Odyssey of Lateen Explorer for Repeated Objects), which searched "for the huge repeat of a human genome," and analyzed "a relation with a cause-of-a-disease gene." (see babelfish translation)
  • OPAL (Optical Property AnaLyzer (of photonic crystals)) which looked "for the material which can manufacture an optical microprocessor." (see babelfish translation)
Life Sciences April 30, 2003 4 months unknown/12,000
The Triangles project found difference triangles with the smallest (optimal) span for a given sequence. The project didn't have a website. The project evolved from a programming contest, sponsored by Al Zimmerman, which ran from July, 2002 to October 15, 2002. See the final results of the contest. The project used a modified version of Jean-Charles Meyrignac's client for the Minimal Equal Sums of Like Powers project and a modified version of Stephen Montgomery-Smith's Dispense Package distributed computing platform.

567,847 entries were submitted for the project. See the smallest known spans discovered by the project.

Join a discussion group about the project.

Mathematics March 7, 2003 4 months 44/unknown
Project Dolphin tracked the total number of keystrokes users made during the use of their computers. It was just for fun. It tracked a total of 34,935,065,880 from over 36,000 users (1,000 times more users than the project coordinator originally planned for). Human January 17, 2003 1 year 36,000/not applicable
DClient was a distributed, brute-force attempt to find the secret "backdoor" password for Tivo's version 3.2 software. This password would allow a Tivo device owner to enable hidden features in the software. The project ended before a key was found. You can read the project post-mortem, download the server source code or the last version of the client application or see the final stats page. The project generated about 2.7 billion blocks of keys using about 85 CPU years.

Version 1 of this project was known as Tivocrack. 320,679 packets were completed for Tivocrack. Tivocrack began from a discussion forum.

Version 2 became active around November 8, 2002.

Cryptography January 20, 2003 3 months unknown/unknown
the smallest remaining Sierpinski problem candidate k=4847 project searched for prime numbers of the form 4847.2n+1 for n > 1,000,000 (n <= 1,000,000 had already been checked). The project was coordinated by Payam Samidoost, an active researcher of Fermat numbers. It used George Woltman's PRP software. The project was merged with the Seventeen or Bust project in November, 2002. Mathematics November, 2002 4 months 9/unknown
The ECCp-109 Challenge was a distributed effort to solve Certicom's ECCp-109 challenge and set "a new world record in characteristic p elliptic curve discrete log computation." The project won Certicom's $10,000 (US) prize: each of the two people who found the winning curve received $1,000 (US) and the remaining $8,000 (US) was given to the Free Software Foundation.

After 68,228,567 total distinguished points were found, the solution was discovered to be k=281183840311601949668207954530684. See more information about the solution.

Join a discussion group about this project.

Cryptography October 15, 2002 6 months 10,308/unknown
Care2's Race for the Pandas had a free button for people to click to save 0.3 acres (2.2 square meters) of endangered panda habitat per click. The project generated 4,185,000 donations, which saved about 1.2 million acres (about 486,000 hectares) of habitat in the Wanglang Nature Reserve in China. Charity July 22, 2002 458 days over 600,000/unknown
Give Water, sponsored by Thames Water, gives safe drinking water to needy people around the world. The project had a free donation button that it asked to be clicked 4 million times. After that goal was reached, the project donated £200,000 to Water Aid, enough money to provide safe drinking water for 13,000 people. Note that the original goal of the project was to donate £50,000 to help 4,600 people. Charity June 26, 2002 6 months unknown/unknown
DECRYPTHON Décrypthon, hosted by the French organization Téléthon 2001, decrypted proteomes to fight against neuromuscular and other diseases. It compared and classified about 550,000 proteins, and made the results available in a free, public database on September 16, 2002. The database website is available in French and English. The project began in March, 2002.

The website is written in French, but an English translation of the main page is available, and babelfish provides a reasonable translation of the text. More information in English may be found at AFM, Association Francaise contre les Myopathies.

Life Sciences May 2, 2002 2 months 75,000/unknown
qoopy logo qoopy uses a single infrastructure to support many kinds of client applications (similar to the Parabon Computation project. The site is hosted by the University of Dortmund in Germany and is written in German, but an English version is also available.

qoopy's first project, EvoChess, evolved chess-playing programs. Each user's client generated some programs. The more successful programs survives and combines with other users' chess programs to speed up the evolutionary process. Users could play against the evolved chess programs and see information about the best evolved programs in the stats pages. The last version of the project client only allowed programs which looked ahead 5-10 moves to survive. In the end, "the first evolution converged quite fast. This was due to the fact that the individuals faced an immovable enemy (the minimax algorithm)."

A diploma thesis "Verteilte Evolution von Schachprogrammen" (Distributed Evolution of Chess Programs) based on this project and written by Roderich Groß and Keno Albrecht, is available online (it is available only in German).

Puzzles/Games February 23, 2002 6 months 1,700 users/unknown (2,228,343 games played)
Muon logo Stephen Brooks' Distributed Particle Accelerator Design project designed "a channel of magnets to produce a particle beam of muons as efficiently as possible from a source of pions spreading in almost all directions. The pions decay into muons and change direction as they move through the apparatus, making this a particularly challenging problem." Before the project, the highest efficiency was 2.9%: the project achieved an efficiency of almost 7.1%. Version 3 of the client application used a genetic algorithm to improve its design and to learn from previous records in the results file, so it found more efficient designs even faster. Science March 1, 2002 7 months 148/unknown (7.066% best muon transfer rate for 6,373,866,047 total particles simulated) runs health, science, and Internet-related research projects.'s first completed project was a bioinformatics research project for the Washington University in St. Louis, Missouri. The project, called HMMER, used the Hidden Markov Modeling technique to compare known DNA sequences (amino acids) against the data from the Human Genome Project to find similar sequences.'s second completed project searched for potential drugs to fight the toxic properties of anthrax so that the disease can be treated in humans in its advanced stages. Any likely drug candidates from the project will be given to the U.S. government and other U.S.-friendly governments for further development into actual drugs. This project began on January 22, 2002 and the screening phase concluded successfully on February 14, 2002. From a pool of 3.57 billion molecules it found over 300,000 drug candidates.

Life Sciences February 14, 2002 4 weeks unknown/unknown (2,867,618 results)
The SaferMarkets project, which ran on the entropia platform, studied the causes of stock market volatility. The project began on April 2, 2001 and ended on January 18, 2002. According to a Business Week article, the goal of the project was to find a formula that can "predict the likelihood, degree, and duration of volatility in the Nasdaq and S&P indexes and in five currency exchanges where the U.S. dollar is half the equation," first using Bayesian statistics regarding human behavior to create a random fictional history of volatility, then fine tuning the formula against real, historical data. Eventually the project would use the formula to predict the volatility of individual equities. The project coordinators will analyze all of the data generated by the project and will publish its final results in economical journals and make the results available to the public for free "to help people improve their finances through better planning tools."

The Safer Markets URL is redirected to entropia's home page, and the SaferMarkets was taken offline immediately when the project concluded.

Financial January, 2002 9 months unknown/9,335 (970,885 tasks completed)
DataSynapse, Inc. DataSynapse built a better P2P web searcher by joining with the National Center for Supercomputing Applications (NCSA) and Lehigh University to develop an approach called Hierarchical Distributed Dynamic Indexing (HDDI TM). Participants were entered into prize sweepstakes drawings. The project was designed only for users with broadband Internet connections and was only available for the Windows platform. Internet December 17, 2001 less than 1 year over 10,000/unknown
Folderol Folderol was a volunteer project that used a screensaver, command-line client or system client application to simulate protein folding of the data from the Human Genome Project. The project completed 36 years of simulation before the project coordinators placed it on hold indefinitely. They may restart it if and when they have time to support it again. Life Sciences September 27, 2001 less than 1 year unknown/unknown
Popular Power Popular Power searched for a more effective influenza vaccine. The company went out of business on March 17, 2001, but the founders continued the influenza vaccine project until September, 2001. The client used Java for task implementation to provide a secure "sandbox" area within which its customers could run their own code without being able to acces the rest of your system (the way a browser provides a secure area for a Java applet).

Background of the influenza vaccine modeling project.

The last unofficial Stats Page created by Mike Rosack.

Life Sciences September, 2001 14 months ?/?