ChessBrain: a Linux-Based Distributed Computing Experiment

If one computer already beats you at chess, wait until 646 of them gang up on you.
Bots, Presence and Autonomous Play

The SuperNode server has a Bot called Shannon (implemented as a thread) that connects to on-line game servers and maintains a presence. Members of the game server type commands to challenge ChessBrain or to watch the current game being played. It has been fun programming Shannon, which now understands a variety of commands. There is a great deal of potential in on-line bots that can be instructed to perform actions on behalf of their hosts.

During the development of ChessBrain, I downloaded the source code to a Free Internet Chess Server (FICS) and compiled it on an old Pentium 200 MMX Toshiba laptop running Linux. FICS is written in C, and it compiled using GCC without incident. The game server allows users to telnet to port 5000 and sign in with a user name and password. After a few months the traffic increased, and we moved the FICS server to another ChessBrain machine at our secondary domain at distributedchess.net. Users now have several options for watching ChessBrain play on-line. They can telnet directly to the game server where ChessBrain is playing or use one of our viewer programs.

After ChessBrain could play on an on-line game server, I wrote a Java game viewer to allow people to watch live games. As an alternative to the Java viewer, I also wrote viewers based in PHP and Macromedia Flash (www.chessbrain.net/viewers.html). ChessBrain contributor Anthony Bravo wrote a Java-based network viewer to show the active PeerNodes throughout the world. Users can click on nodes to see how many machines are active in a given country. All of the viewers on the ChessBrain site use SOAP to communicate with the SuperNode.

As a security precaution, browser plugins such as Java and Macromedia's Flash ActionScript don't allow the program to connect to a server other than the one from which the applet was downloaded. To work around this issue, I wrote a simple XML proxy script that accepts an HTTP GET request on one server and connects to the SuperNode server on behalf of the client. For example, if you wanted to query the SuperNode server for the current game position, you could enter the following URL into your browser: http://www.chessbrain.net/xmlproxy.php?command=CBSGetPos. The server would respond with a SOAP package like the one in Figure 4. On Mozilla you can view the page source to see the actual SOAP document.

Monitoring the SuperNode

Monitoring a server's health is an important part of system administration. Fortunately for developers, Linux offers many ways to tackle server monitoring. The Linux /proc virtual filesystem contains a goldmine of valuable system data, offering developers an easy way to profile and monitor system behavior. /proc/net/dev offers device data such as the number of bytes and packets sent and received on a network interface, and /proc/meminfo offers loads of memory statistics. If data mining the /proc isn't your thing, sysinfo() offers a quick and easy way to fill a structure with system statistics, such as system load, freeram and the total number of processes.

The SuperNode server offers a SOAP request that returns system information similar to what is shown in Listing 2. ChessBrain member Greg Davis wrote the first SuperNode monitor in Perl, which issues the SOAP request and displays a screen similar to the top command.

______________________

Comments

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Four years

Blog Client's picture

Reading that article today is like watching some old star trek series :D

Gameserver

Gameserver's picture

Nice Script :P
I hope it will bring the Thing further.

Gameserver

Webcast
How to Build an Optimal Hadoop Cluster to Store and Maintain Unlimited Amounts of Data Using Microservers

Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.

Learn More

Sponsored by AMD

White Paper
Red Hat White Paper: Using an Open Source Framework to Catch the Bad Guy

Built-in forensics, incident response, and security with Red Hat Enterprise Linux 6

Every security policy provides guidance and requirements for ensuring adequate protection of information and data, as well as high-level technical and administrative security requirements for a system in a given environment. Traditionally, providing security for a system focuses on the confidentiality of the information on it. However, protecting the data integrity and system and data availability is just as important. For example, when processing United States intelligence information, there are three attributes that require protection: confidentiality, integrity, and availability.

Learn more about catching the bad guy in this free white paper.

Learn More

Sponsored by DLT Solutions