Evaluating hardware models having resource contention

Abstract

A performance prediction simulator gives effect to the resource contention among multiple resources in a simulated system by adjusting event durations appropriately. A resource topology tree defining the resource configuration of the system is input to the simulator. The simulator includes an evaluation engine that determines the amount of resource used during each simulation interval of the simulation and records the resource usage in a resource contention timeline, which can be displayed to a user. The amount of resource used during a simulation is also used to adjust the event duration calculations of the hardware models associated with each event.

Description

RELATED APPLICATIONS [0001] The application is related to U.S. patent application Ser. No. ______, entitled “LATE BINDING OF RESOURCE ALLOCATION IN A PERFORMANCE SIMULATION INFRASTRUCTURE” [Docket No. MS#164198.1 / 40062.162US01], specifically incorporated herein for all that it discloses and teaches.TECHNICAL FIELD [0002] The invention relates generally to computer system performance simulation, and more particularly to a performance simulation infrastructure that considers the impact of resource contention. BACKGROUND OF THE INVENTION [0003] Performance simulation of software systems running on one or more computers is a crucial consideration for developers deploying network-based services, such as those services available over the Internet, an intranet, or an extranet. For example, developers often want to determine how their software design decisions will affect future system performance. Likewise, system users want to determine the optimal mix of hardware to purchase for an expec...

AI Technical Summary

Benefits of technology

[0007] Embodiments of the present invention solve the discussed problems by providing a performance simulation infrastructure that generates resource contention timelines (“RCTs”) representing load levels (i.e., resource usage) on various system resources over the course of a performance simulation. Such systems typically includes both software and hardware elements. A performance prediction simulator gives effect to the resource contention among multiple resources in the simulated system by adjusting event durations appropriately. A resource topology tree defining the resource configuration of the system is input to the simulator. The simulator includes an evaluation engine that determines the amount of resource used during each simulation interval of the simulation and records the resource usage in a resource contention timeline, which can be displayed to a user. The amount of resource used during a simulation is also used to adjust the event duration calculations of the hardware models associated with each event. Each RCT can be presented to a user in a graphical or text format to provide rich feedback about system bottlenecks resulting from resource contention.

Problems solved by technology

One problem with existing approaches for modeling performance of a system is that contention for use of system resources is not accurately modeled at high load levels.

Under such load conditions, a resource may be subject to events generated by multiple other resources within the system.

Such events may exceed the performance capabilities of the resource; therefore, some of the events may be rejected or delayed by the resource until it is capable of handling the events.

Another problem with existing approaches for modeling performance of a system is that feedback about such resource contention is not easily and accurately communicated to the user.

However, neither approach offers any explanation as to why these simulation times resulted.

It would be more helpful to offer feedback indicating that, for example, the system's ability to access a given disk or communication channel was overloaded, thereby resulting in a slower than expected performance.

Images