Electrical computers and digital processing systems: multicomput – Computer network managing – Computer network monitoring
Reexamination Certificate
2000-08-07
2004-11-23
Cuchlinski, Jr., William A. (Department: 2144)
Electrical computers and digital processing systems: multicomput
Computer network managing
Computer network monitoring
C709S203000, C709S228000
Reexamination Certificate
active
06823380
ABSTRACT:
TECHNICAL FIELD
The present invention relates generally to computer systems, and more particularly to a system and method for providing a consistent server load wherein the load may be adjusted and controlled in order to determine server capacity—thereby facilitating server performance testing and capacity planning.
BACKGROUND OF THE INVENTION
With the rise of Internet and other related technologies, system requirements for servicing ever increasing network traffic have dramatically changed. Ever more powerful server systems, which are often configured as an array of servers, are often provided to service requests originating from external sources such as the World Wide Web, for example. As local Intranet systems have become more sophisticated thereby requiring servicing of larger network loads and related applications, internal system demands have grown accordingly as well. Unfortunately, as demands for more complicated and higher performance systems have increased, conventional tools and/or methods for measuring, predicting, and planning whether a given server is capable of servicing a desired network load have not been adequate.
Presently, many performance measurement systems rely on a connection model in order to determine whether a server may have the requisite capacity to service a desired network load. According to the connection model, a plurality of client systems may be configured to generate a large number of concurrent connections. However, this type of model may not adequately stress a given server and/or application and is thus inadequate for a number of reasons. One such problem is the connection model does not provide a controllable amount of stress for the server. Since each client operates independently, a suitable mechanism, for example, would be needed to coordinate the various load requests. Unfortunately, such mechanisms are presently unavailable. Another problem relates to providing a sustained and/or predictable rate of requests. Generally, since clients are generating load requests open loop (e.g., without monitoring the number of requests), peaks and valleys of requests are likely to occur. Therefore, clients may likely wait while outstanding connections remain open wherein a given server may not be stressed to its fullest capacity. Thus, performance measurements based on an inconsistent load are difficult to determine.
Still another problem associated with conventional systems relates to adjusting desired loads. Currently, in order to increase the number of load requests, an additional client system may have to be installed to provide additional connections. Consequently, costs are increased and resulting granular adjustment of combined server load difficult to attain. Still, conventional systems that provide loosely coupled/controlled and/or unregulated output stream of requests do not enable accurate predictions to be made as to whether an application and/or system may operate at a desired capacity or stress level. In view of the above problems, there is an unsolved need for a system and/or methodology to generate a consistent network load, provide a more realistic load model, and thereby improve server performance testing to achieve improved capacity planning.
SUMMARY OF THE INVENTION
The present invention relates to a system and method to provide a predictable and user definable load/tool for stressing and/or testing a server's performance. In accordance with the present invention, a user may define/specify a control rate (e.g., requests per second) for a continual stream of requests (e.g., HTTP requests) to be sent from a client to a web server, for example. In this manner, a consistent and predictable load may be established for the server in order to determine if a given server may have the capacity to process the desired load. By establishing a predetermined load at a consistent rate for the server, a user may determine: whether particular code and/or applications operate under load, whether any negative results are produced from the load, whether breaking points are produced from excessive load, and/or whether new applications and/or hardware have the capacity to serve the desired load.
More specifically, the present invention provides a queuing architecture to store and retrieve data—in the form of HTTP requests, in a rapid manner. The architecture may provide a queue for processing the requests, a component for sending/receiving the requests, and a feedback loop for controlling the desired rate of the requests, and a scheduler for determining a continual rate of requests for an upcoming period of time, for example. Data items designated for the queue may be sorted according to a predetermined criteria (e.g., time to execute the request) and inserted into the queue such that the queue may remain time sorted. As will be described in more detail below, the scheduler may be provided to enable scheduling of requests and interfacing to the component. The component may then send/receive requests by managing threads via the scheduler (e.g., scheduling API (Application Program Interface) to enable when the requests are actually sent to the server. The feedback loop may reside in a separate thread and may be responsible for calculating requests per second being generated and thus supply the scheduler with information for increasing and/or decreasing the rate of requests.
The following description and the annexed drawings set forth in detail certain illustrative aspects of the invention. These aspects are indicative, however, of but a few of the various ways in which the principles of the invention may be employed and the present invention is intended to include all such aspects and their equivalents. Other advantages and novel features of the invention will become apparent from the following detailed description of the invention when considered in conjunction with the drawings.
REFERENCES:
patent: 5583792 (1996-12-01), Li et al.
patent: 6012090 (2000-01-01), Chung et al.
patent: 6078943 (2000-06-01), Yu
patent: 6078960 (2000-06-01), Ballard
patent: 6173322 (2001-01-01), Hu
patent: 6175869 (2001-01-01), Ahuja et al.
patent: 6314463 (2001-11-01), Abbott et al.
patent: 6330602 (2001-12-01), Law et al.
patent: 6446120 (2002-09-01), Dantressangle
patent: 6526448 (2003-02-01), Blewett
patent: 6571288 (2003-05-01), Sarukkai
G. Banga and P. Druschel, “Measuring the capacity of a web server”, In Proc. of USITS, Dec. 1997.*
P. Goyal, H. Vin, and H. Cheng, “Start-time fair queuing: A scheduling algorithm for integrated services packet switching networks”, In Proc. of ACM SIGCOMM, pp. 157-168, Aug. 1996.*
G. Banga and P. Druschel, “Measuring the capacity of a web server under realistic loads”, World Wide Web, 1999.*
D. Mosberger and T. Jin, “httperf—A Tool for Measuring Web Server Performance”, 1998.*
P.Goyal, H.M. Vin, H.Cheng, “Start-time fair queuing: A scheduling algorithm for integrated services packet switching networks”, IEEE/ACM Transactions on networking, vol. 5, No. 5, pp. 690-704, Oct. 1997.*
Daniel A. Menasce et al, “Capacity Planning for Web Performance: metrics, models, and methods”, Prentice Hall, 1998. chapter 12, p. -264-289.*
Cheng ST (REPRINT), et al.; “Dynamic Quota-Based Admission Control with Sub-Rating in Multimedia Servers”,Multimedia Systems, vol. 8, No. 2, 2000, p. 83-91.
Kuo-Hsiung Wang, et al.; “A Recursive Method to the Optimal Control of an M/G/1 Queueing System with Finite Capacity and Infinite Capacity”,Applied Mathematical Modelling, vol. 24, No. 12, p. 899-914.
Wang KH (REPRINT), et al.; “A Queueing System with Queue-Dependent Servers and Finite Capacity”,Applied Mathematical Modelling, vol. 24, No. 11, 2000, p. 807-814.
Nace Eric W.
Patiejunas Kestutis
Amin & Turocy LLP
Cuchlinski Jr. William A.
Gerezgiher Yemane M.
LandOfFree
System and method providing continual rate requests does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method providing continual rate requests, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method providing continual rate requests will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3344977