Data processing: measuring – calibrating – or testing – Measurement system – Performance or efficiency evaluation
Reexamination Certificate
1998-08-13
2001-07-17
Shah, Kamini (Department: 2857)
Data processing: measuring, calibrating, or testing
Measurement system
Performance or efficiency evaluation
C711S167000
Reexamination Certificate
active
06263298
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a method for analyzing the performance of a computerized system, in general and to a method for analyzing a computer system, both software and hardware, in particular.
BACKGROUND OF THE INVENTION
Database related software environments as well as hardware environments are known in the art. A large scale software environment often includes a plurality of software applications, for managing a plurality of storage devices, such as disk drives and memory devices, which include storage units, such as files.
One popular database related software environment is the Oracle software package, manufactured and sold by the Oracle Corporation. The Oracle software is generally directed to managing resources, both in software sessions and in the processed data. This is a large scale software product which is capable of processing and managing large amounts of data, for a plurality of users.
Conventionally, database related software environments provide a high level interface language, for use by an end-user or a developer, in the form of statements which are associated with the systems database. These statements control input and output flow, as well as execution of database objects, which are associated with the software environment.
It will be appreciated by those skilled in the art, that the tuning of such software environment is required to optimize their performance, especially, for large scale systems.
Methods for tuning such software applications are known in the art. One such tuning method is implemented in the Oracle Trace software product, manufactured and sold by the Oracle Corporation. According to this method, the length of a queue of statements which access a selected database object, is measured. This length is directly proportional to the load which is currently applied on the selected objet.
Storage device managing systems, are known in the art, for managing storage hardware environment, which includes a plurality of storage devices. Such systems attempt to optimize the performance of a storage hardware environment by detecting the amount of I/O activity through each storage device, which is directly proportional to the load applied thereon.
Reference is now made to
FIG. 1
, which is a schematic illustration of a method for tuning a software application, known in the art.
In step
102
, the processing activities within the system are sampled. The sampled processing activities include software procedures as well as operating system sessions. The results of this step are then provided to a user in a visual format (step
106
).
In step
104
, the access activities within the system are sampled. The sampled access activities include data object access requests. The results of this step are then provided to a user in a visual format (step
108
).
In step
110
, the user which operates according to the method of the prior art, decides upon the action which is to be taken. Such an action can include a changing or updating the configuration of the system (step
112
), duplicating an item which is heavily accessed (step
114
), moving an item from a loaded sub-environment (step
116
), such as a processor or a storage device and adding a storage device (step
118
).
Reference is now made to
FIG. 2
, which is a schematic illustration of a method for tuning a storage hardware system, known in the art.
In step
202
, the hardware storage activities are sampled. These hardware storage activities include storage hardware device access requests, as well as storage data distribution over and between the plurality of storage devices.
In step
204
, the performance characteristics of each of the storage devices are detected. Conventionally, this step is performed by processing the information gathered in the previous step.
In step
206
, the detected performance characteristics are displayed to a user.
In step
208
, the data distribution is rearranged according to the decision of the user.
None of the above methods provide a complete and optimized solution to the problem presented by such complex systems.
SUMMARY OF THE PRESENT INVENTION
It is an object of the present invention to provide a novel method for monitoring a computer system, both software and hardware, and for analyzing, detecting and retrieving information relating thereto which overcomes the disadvantages of the prior art.
In accordance with the present invention, there is thus provided a method for monitoring a computer system having a plurality of storage devices and processing means connected thereto. The storage devices containing a plurality of storage units. The processing means processing a plurality of access requests to the storage units.
The method in accordance with the present invention comprise the steps of:
analyzing each said access requests, thereby determining its respective waiting time period, detecting from said waiting time periods selected waiting time periods which exceed a predetermined threshold value, retrieving information relating to waiting access requests which are associated with said selected waiting time periods, retrieving information relating to at least one of said storage units, which is associated with said waiting the access requests, and retrieving information relating to at least one storage device, containing said at least one storage units.
In accordance with another aspect of the present invention, the method further includes the preliminary steps of detecting the distribution of said storage units among said storage devices, and storing information indicative of said detected distribution.
In accordance with another aspect of the present invention, the method further includes the step of providing said information to a user.
In accordance with another aspect of the present invention, the method further includes the step of adjusting resource distribution within said computer system.
In accordance with another aspect of the present invention, the method wherein said step of adjusting resource distribution further includes a sub-step of distributing access requests of the same nature of said detected waiting access requests.
In accordance with another aspect of the present invention, the method wherein said step of adjusting resource distribution further includes a sub-step of relocating said at least one storage unit to others of said storage devices.
In accordance with another aspect of the present invention, the method wherein said step of adjusting resource distribution further includes a sub-step of duplicating said at least one storage unit to others of said storage devices.
In accordance with another aspect of the present invention, the method further includes the step of detecting when the number of said waiting access requests exceeds a predetermined value.
In accordance with another aspect of the present invention, the method further includes the step of detecting when the number of said access requests which are associated with a selected one of said storage units, exceeds a predetermined value.
In accordance with another aspect of the present invention, the method further includes the step of detecting when the number of said access requests which are associated with a selected one of said storage devices, exceeds a predetermined value.
REFERENCES:
patent: 5062055 (1991-10-01), Chinnaswamy et al.
patent: 5495607 (1996-02-01), Pisello et al.
patent: 5557548 (1996-09-01), Gover et al.
patent: 5768500 (1998-06-01), Agrawal et al.
patent: 5936794 (1999-08-01), Mizukami et al.
Kerman Shiri
Kopans Haim
Shain Tomer
Barbee Manuel L.
Davidson Davidson & Kappel LLC
Precise Software Solutions Ltd.
Shah Kamini
LandOfFree
Method for analyzing system performance does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for analyzing system performance, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for analyzing system performance will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2557482