Electrical computers and digital processing systems: multicomput – Computer-to-computer protocol implementing – Computer-to-computer data transfer regulating
Reexamination Certificate
1996-10-01
2003-06-17
Powell, Mark (Department: 2142)
Electrical computers and digital processing systems: multicomput
Computer-to-computer protocol implementing
Computer-to-computer data transfer regulating
C709S233000, C709S213000
Reexamination Certificate
active
06581104
ABSTRACT:
TECHNICAL FIELD
The present invention is directed to managing a large distributed enterprise environment and, more particularly, to regulating how data is distributed among computing resources in the managed environment.
BACKGROUND OF THE INVENTION
The problem of designing the most-efficient message distribution hierarchy in a complex network must take into consideration a number of factors including network topology, network bandwidth and machine resources. A well-designed distribution method would be both efficient and fault tolerant. Because of the varying capabilities of the actual networks in use today, however, it is not possible to select, a priori, a distribution method that will work well for all networks. This problem is exacerbated in a large, geographically-dispersed network environment. One such known environment includes a management framework comprising a server that manages a number of nodes, each of which has a local object database that stores object data specific to the local node. The server is used to perform a variety of system management tasks, including a multiplexed distribution service that performs an efficient one-to-many data distribution. A data distribution is typically initiated at each target node with an object method invocation.
It is also known in such distribution schemes to provide a network load tunable parameter that can be set to limit the amount of data a distribution will write per unit time. Such techniques, however, do not adequately address the serious loading problems that can occur when multiple endpoints seek to obtain distribution service at the same time. The following illustrates this problem in a representative distribution hierarchy. Assume a repeater in the network fans-out to 50 machines, the machines are all on separate T1 links, and the network has a 10 mbit local area network (LAN) that must be crossed to get to a router running the T1 links. The repeater has a sustainable send rate of 750 KB/sec, and the netload tuning parameter is set to 500 KB/sec to keep LAN traffic manageable. In this example, the LAN becomes a “bottleneck” because only 4-5 endpoints (500/1.5) can be kept busy in parallel. If the distribution is made to more than 5 endpoints in parallel, all distributions are slowed down. Sending the distribution to all 50 machines has the effect of making each network only one tenth as busy. The problem is made even worse if the endpoint machines include relatively slow modems (e.g., less than 9600 baud), which is quite common since endpoint machines are typically the last part of the enterprise to be upgraded. On these old networks, even a single 16 K write operation saturates the network for close to 30 seconds, making it impossible to do other work.
Thus, it would be desirable to provide an improved mechanism to distribute data efficiently in a managed network environment.
BRIEF SUMMARY OF THE INVENTION
It is a primary object of the invention to provide an efficient one-to-many data distribution service for a distributed enterprise computing environment.
It is a more specific object of the invention to control distribution of data in a large, geographically-dispersed managed enterprise having a plurality of endpoint machines.
It is another object of the invention to enable an enterprise to place substantially all of its computing resources on a network that is managed in a reliable, cost-effective manner, and to provide an efficient data distribution service within such enterprise.
It is a further object of the invention to provide a multiplexed distribution scheme within a managed environment wherein distribution loads within one or more networks connected to respective endpoint machines are balanced.
Still another object of the invention is to provide load control for each network that connects an endpoint machine to a management node in the enterprise environment.
Another object of the invention is to facilitate parallel distribution to endpoint machines in a large, distributed enterprise environment.
It is another object of the invention to use actual network load information during a data distribution to ensure that each network does not exceed a desired load value.
It is still another object of the invention to meet the needs of customers with very large and geographically-dispersed networks and, more particularly, to significantly expand the data distribution capabilities of the network administrators.
These and other objects are achieved in a large distributed enterprise that includes computing resources organized into one or more managed regions, each region being managed by a management server servicing one or more gateway machines, with each gateway machine servicing a plurality of endpoint machines. A system management framework is preferably “distributed” on the gateway machines and the one or more endpoint machines to carry out system management tasks.
To facilitate balanced distribution of data, a network administrator first sets a load parameter identifying an amount of network bandwidth that may be consumed by a particular data distribution over each subnet of each network path between the gateway machine and the endpoints serviced by that gateway. Prior to initiating a data distribution, a calculation is made of the effective load for each subnet that will be affected by (i.e. that will handle) the data distribution. A determination is then made regarding whether the effective load in any subnet that is affected by the data distribution exceeds the load parameter for that subnet. If so, the data distribution is altered by inserting one or more delays in the rate at which data is transmitted over the affected network path from the gateway. The data distribution, as altered, is then initiated. These steps are repeated for a new data distribution, or at predetermined intervals.
The foregoing has outlined some of the more pertinent objects of the present invention. These objects should be construed to be merely illustrative of some of the more prominent features and applications of the invention. Many other beneficial results can be attained by applying the disclosed invention in a different manner or modifying the invention as will be described. Accordingly, other objects and a fuller understanding of the invention may be had by referring to the following Detailed Description of the preferred embodiment.
REFERENCES:
patent: 4551833 (1985-11-01), Turner
patent: 5142531 (1992-08-01), Kirby
patent: 5204949 (1993-04-01), Yasue et al.
patent: 5283897 (1994-02-01), Georgiadis et al.
patent: 5287508 (1994-02-01), Hejna, Jr. et al.
patent: 5295139 (1994-03-01), Palmer
patent: 5377327 (1994-12-01), Jain et al.
patent: 5406559 (1995-04-01), Edem et al.
patent: 5422879 (1995-06-01), Parsons et al.
patent: 5425023 (1995-06-01), Haraguchi et al.
patent: 5434848 (1995-07-01), Chimento, Jr. et al.
patent: 5440719 (1995-08-01), Hanes et al.
patent: 5452350 (1995-09-01), Reynolds et al.
patent: 5491801 (1996-02-01), Jain et al.
patent: 5506834 (1996-04-01), Sekihata et al.
patent: 5515371 (1996-05-01), Venters
patent: 5517643 (1996-05-01), Davy
patent: 5581610 (1996-12-01), Hooshiari
patent: 5598566 (1997-01-01), Pascucci et al.
patent: 5600637 (1997-02-01), Kikuta
patent: 5652751 (1997-07-01), Sharony
patent: 5712981 (1998-01-01), McKee et al.
patent: 5717745 (1998-02-01), Vijay et al.
patent: 5734652 (1998-03-01), Kwok
patent: 5742587 (1998-04-01), Zornig et al.
patent: 5774668 (1998-06-01), Choquier et al.
patent: 5793976 (1998-08-01), Chen et al.
patent: 5799002 (1998-08-01), Krishnan
patent: 5805072 (1998-09-01), Kakemizu
patent: 5889761 (1999-03-01), Yamato
patent: 5923849 (1999-07-01), Venkatraman
patent: 5925100 (1999-07-01), Drewry et al.
patent: 5978851 (1999-11-01), Kayama et al.
patent: 0498967 (1991-02-01), None
patent: 9524812 (1995-09-01), None
Iida et al, Autonomous routing scheme for large scale network based on neural processing IEEE, 1989.*
Douglas C.Schmidt, Tim Harrison and Ehab Al-Shaer. [www.acl.lanl/CORBA/#DOCS], Jun. 1995.*
IBM Technical Disclosure Bulletin “Connection E
Burwell Joseph R.
International Business Machines - Corporation
Judson David H.
LaBaw Jeffrey S.
Powell Mark
LandOfFree
Load balancing in a distributed computer enterprise environment does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Load balancing in a distributed computer enterprise environment, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Load balancing in a distributed computer enterprise environment will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3114701