Browse Prior Art Database

Efficient Process Scheduling in the Workstations in the Cooperative Client Server Processing

IP.com Disclosure Number: IPCOM000107638D
Original Publication Date: 1992-Mar-01
Included in the Prior Art Database: 2005-Mar-22
Document File: 2 page(s) / 89K

Publishing Venue

IBM

Related People

Parikh, SN: AUTHOR

Abstract

In the cooperative processing and client server architecture applications, a work station typically requests the completion of a transaction (or a request) from the host ("server").

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Efficient Process Scheduling in the Workstations in the Cooperative Client Server Processing

       In the cooperative processing and client server
architecture applications, a work station typically requests the
completion of a transaction (or a request) from the host ("server").

      Often, multiple processes execute in the workstation, each of
them simultaneously requesting service from the host. These processes
typically wait for the responses from the host. This delay can be on
the order of several hundred milliseconds or more.

      The operating system in the requesting workstation is not
cognizant of this wait time-related delay.

      There could be multiple processes in the workstation making
these requests (initiating transactions) to the host, each of which
will ultimately end up waiting for a certain amount of time for the
service from the host.

      The operating systems operating in these computing environments
do not have the capability of knowing in advance that the process on
the requester will be waiting for a certain amount of computing/
execution time, etc.

      As a result, assuming all processes have equal priority, all
processes will be allocated an equal number of clock cycles while
executing instructions on the workstation. With availability of this
information to the operating system scheduler before the execution of
these processes begins, a better, more efficient utilization of the
processing resources will be possible.

      Consider the following example:

      Assume two processes A and B are running in the workstation.
Each has the following 'execution characteristic': Process A: Execute
for 1000 ms.  Wait for 1000 ms for response from server. Execute for
1 ms.  Process B: Execute for 200 ms.  Wait for 100 ms for response
from server.  Execute for 1 ms.  Assume both are at same priority,
i.e., when both are active, each will get half of the CPU cycles.

      With this scenario, process A will complete in 2202
milliseconds...