Electrical computers and digital data processing systems: input/ – Input/output data processing – Input/output data buffering
Reexamination Certificate
1998-09-17
2002-04-02
Lee, Thomas (Department: 2182)
Electrical computers and digital data processing systems: input/
Input/output data processing
Input/output data buffering
C710S060000, C706S920000, C716S030000, C711S113000
Reexamination Certificate
active
06366969
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates generally to methods of measuring performance of computer systems, such as personal computers. More particularly, the invention relates to measuring the performance of disk subsystems, such as hard drives.
2. Description of the Related Art
A powerful computer system is often noted for its ability to quickly store, retrieve, communicate, and process a large amount of information, such as conventional data, voice, graphics, and interactive video. With the evolution of intranets and internets, such as the public Internet, the need for more powerful computer systems is generally of paramount consideration to consumers. In assessing the power of a computer system, a consumer often measures the computer system performance using one of several commercially available benchmarks. A benchmark is a computer program which tests and measures performance of a computer system and its subsystems, such as disk, CD-ROM, memory, processor, and graphics/video subsystems.
These benchmark tests are intended to provide information about the overall performance of the computer system. One kind of benchmark testing involves measuring a data transfer rate within the computer system. The data transfer rate is the rate at which the computer (e.g., the main processor) transfers data among its internal components, such as between a hard drive and a random access memory (RAM). The data transfer rate may be of great importance in assessing the power of the computer system. When operating the computer system, certain data is read from and written to the main memory frequently and repeatedly. For instance, recognizing that the disk drive is spending much of its time reading the same information over and over leads to the conclusion that the data transfer rate may be an important criterion in determining the performance of the computer system.
Commercially available benchmarks are often inadequate for measuring computer system performance. Moreover, these benchmarks may provide different and inconsistent test results for the same computer system. For example, a benchmark manufactured by Ziff-Davis may provide test results which are different than those provided by Adaptec Threadmark 32. Additionally, these benchmarks often provide performance results which may not match a computer manufacturer's specifications. Consequently, a consumer may be forced to not rely on performance test results provided by commercially available benchmarks.
Computer manufacturers often use their own benchmarks which test computer subsystem performance, such as data transfer rate. These benchmarks are often designed for a specific product or group of products and, hence, may not be applied to compare computer performance from different manufacturers. Consequently, it is difficult to rely upon these benchmark programs for accurate and comparative results.
There may be other difficulties associated with determining data transfer rates in a computer system. These difficulties are often due to inherent operational characteristics of computer hard drives. More particularly, a hard drive typically transfers data in short read and write bursts, lasting few milliseconds. This is accomplished by reading a relatively small amount of data in a short burst, and immediately thereafter, writing the small amount of data, in a similarly short burst. Hence, a transfer of an entire data file is often broken up into a sequence of very short bursts of read and write operations. This operation renders it difficult to accurately measure the rate at which data is actually transferred between two memory subsystems (and, particularly, across the main/host bus).
Therefore, there is a need for a system and method of analyzing and determining the performance of computer subsystems objectively and, when repeated, with consistent results. The system should be suitable for use with a variety of operating systems having various subsystems and components.
SUMMARY OF THE INVENTION
To overcome the above-mentioned problems, the invention provides a method of assessing performance of a device in a computer. One embodiment of the invention is a method that comprises establishing a virtual drive, and transferring data between the virtual drive and the device. The method further comprises measuring the transfer rate of the device. In another embodiment, the invention provides a method of measuring performance of a hard drive in a computer. The method comprises establishing a RAM drive in the computer, and transferring a file from the RAM drive to the hard drive. The method further comprises measuring the duration of time needed to transfer the file, and determining a data transfer rate of the hard drive. In another embodiment, the method comprises establishing a RAM drive in the computer, and transferring a file from the hard drive to the RAM drive. The method further comprises measuring the duration of time needed to transfer the file, and determining a data transfer rate of the hard drive.
REFERENCES:
patent: 4851998 (1989-07-01), Hospodor
patent: 4937734 (1990-06-01), Bechtolsheim
patent: 5121487 (1992-06-01), Bechtolsheim
patent: 5245638 (1993-09-01), Gustafson
patent: 5265236 (1993-11-01), Mehring et al.
patent: 5291595 (1994-03-01), Martins
patent: 5566206 (1996-10-01), Butler et al.
patent: 5778420 (1998-07-01), Shitara et al.
patent: 5787463 (1998-07-01), Gajjar
patent: 5812564 (1998-09-01), Bonke et al.
patent: 5819066 (1998-10-01), Bromberg et al.
patent: 5857147 (1999-01-01), Gardner et al.
patent: 5944802 (1999-08-01), Bello et al.
patent: 5987400 (1999-11-01), Hirano
patent: 5987479 (1999-11-01), Oliver
patent: 5999995 (1999-12-01), Ueno et al.
patent: 6047115 (2000-04-01), Hohan et al.
patent: 6223267 (2001-04-01), Hodges et al.
patent: 02300932 (1990-12-01), None
patent: 004330665 (1992-11-01), None
patent: 07168673 (1995-07-01), None
Ling, X.-P; WASMII: A Data Driven Computer On A Virtual Hardware; Apr. 5-7, 1993; pp. 33-42.
Elamin Abdelmoniem
Knobbe, Martens, Olson, & Bear,LLP.
Lee Thomas
LandOfFree
Method of determining data transfer rate of a device by... 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 of determining data transfer rate of a device by..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of determining data transfer rate of a device by... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2932769