Livermore Computing Production Control System, 3.0: Product description. Revision 1

The purpose of this document is to introduce and describe the Livermore Computing Production Control System, commonly called the PCS. Resource management is very difficult in typical UNIX and related systems. The weakness involves the allocation, control and delivery of the usage of computer resources. Accounting also is weak, being generally limited to reporting that a user used an amount of time on a process. A process accounting record is made only after the completion of a process and then only if the process does not terminate as a result of a system panic. Resources can only be allocated to users. No cognizance can be taken of organizational structure or projects. Allocations cannot be managed at these levels. Denying service to a user who has access to a machine is crude. Large computing centers frequently have thousands of users working on hundreds of projects. These users and the projects are funded by several organizations with varying ability or willingness to pay for the computer services provided. With only typical UNIX tools, the appropriate delivery of resources to the correct organizations, projects, tasks and users requires continual human intervention. UNIX and related systems have become increasingly more reliable over the past few years. A consequence of this is that resource managers and accountants have been presented with an attendant problem of accurate accounting for resources used. Processes can now run for days or months without terminating. A run-away process or a malicious or uninformed user can use an organization`s budgeted resource allocation many times over before an accounting record to that effect is written. If a process`s accounting record is not written because of a panic, the computer center is faced with possibly significant financial loss.