Electrical computers and digital processing systems: multicomput – Computer-to-computer data routing – Least weight routing
Reexamination Certificate
1997-06-19
2001-07-31
Banankhah, Majid A. (Department: 2151)
Electrical computers and digital processing systems: multicomput
Computer-to-computer data routing
Least weight routing
C709S241000
Reexamination Certificate
active
06269391
ABSTRACT:
BACKGROUND
1. The Field of the Invention
This invention relates to operating systems for computers of a multi-processor type and, more particularly, to novel systems and methods for providing schedulers within such operating systems.
2. The Background Art
An operating system is software to be executed by a processor of a computer. A multi-processing operating system operates to facilitate operation of multiple processors, usually in a single computer. A multi-processor operating system may be relied upon by executables, such as applications, library routines, and the like, in order to determine an order of execution for each executable.
An operating system provides an interface between the hardware or processor and its associated components, and a higher level executable such as an application. Thus, an operating system provides services to an application by communicating to the hardware on behalf of the higher level application.
Within an operating system, a kernel provides a scheduling function. In general, a kernel is sometimes defined as including not only scheduling software but also memory-management functions, interrupt services, and the like. It is also proper to speak of the scheduling kernel as the kernel or the scheduler, as will be done herein.
Currently known scheduling kernels are not architected to manage effectively, efficiently, nor completely satisfactorily. Many problems arise in scheduling multiple processes, programs, or applications, each having multiple threads.
For example, a thread may run on one processor, while at another time the same thread runs on a second processor. This may result in certain supporting data being moved between caches associated with the two different processors. Coordinating, updating, synchronizing, controlling, passing and validating instances of data structures (e.g. objects) may result in cache thrashing.
Cache thrashing may be thought of as wasting inordinate amounts of valuable processing time administering the coordination of necessary data structures. Scheduling procedures often result in cache thrashing in multi-processor schedulers.
BRIEF SUMMARY AND OBJECTS OF THE INVENTION
In view of the foregoing, it is a primary object of the present invention to provide a multi-processor scheduling kernel for a multi-processing operating system.
Consistent with the foregoing objects, and in accordance with the invention as embodied and broadly described herein, a multi-processor scheduling kernel is disclosed in one embodiment of the present invention as including a multi-processor scheduling module and a virtual machine scheduling module.
The multi-processor scheduling module may select, from a plurality of virtual machines, a virtual machine to be executed by a processor included in the multi-processor. The virtual machine scheduling module may select an order in which to execute threads of a virtual machine. The scheduling kernel may also have an execution exclusion set module for selectively limiting the execution of threads in an execution exclusion set to only one thread at a time.
REFERENCES:
patent: 5717926 (1998-02-01), Browning et al.
patent: 5721931 (1998-02-01), Gephardt et al.
patent: 5774668 (1998-06-01), Choquier et al.
Customizable Policy Management in the Sting Operating System (James Philbini, Oct. 1992).
Banankhah Majid A.
Haynes and Boone LLP
Novell Inc.
LandOfFree
Multi-processor scheduling kernel does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Multi-processor scheduling kernel, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multi-processor scheduling kernel will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2563653