Queue is a load-balancing system for UNIX that allows intuitive control of remote executing jobs through job control by the local shell. You can suspend, resume, send signals, and learn of termination events at the remote end by sending these to the local 'stub' daemon, which acts as a proxy for the job executing on the remote host. Queue also features a more traditional batch processing mechanism, where users are informed of the status of remote jobs via email. Remote jobs may be scheduled based on a number of factors, including modified load average at the remote end, time of day scheduling, number of slots available for process type, etc., etc. Queue also functions as a handy replacement for rsh when executing jobs within the local cluster. This is the second beta version release of the code. In this release, we have added GNU autoconf scripts for additional portability; consequently, we now support additional platforms, including GNU/Linux, HP-UX, SunOS 5.6, Solaris, and IRIX (alpha support). Addtn'l changes may be found in the changelog, http://bioinfo.mbb.yale.edu/~wkrebs/ChangeLog.txt . Download instructions for Queue may obtained off the home page, http://bioinfo.mbb.yale.edu/~wkrebs/queue.html . The current version may be download from http://bioinfo.mbb.yale.edu/~wkrebs/queue-0.23b-beta.tar.gz . We have formed a mailing list for developers and testers of the Queue system. Currently, our development list is working on several fun projects, including MPI & PVM support, secure socket connections, AFS & Kerberos support. We're also porting and improving a nifty utility that lets you monitor and control the execution of Queue jobs throughout your cluster. The list is a great way to tap into the group's expertise and keep up with the latest developments. So, come join the fun and keep up with the latest developments by sending subject line 'subscribe' to '[EMAIL PROTECTED]' or by clicking here: mailto:[EMAIL PROTECTED]?subject=subscribe .
