Method to Identify Transactions and Manage the Capacity to Support the Transaction

• US 20090037580A1
• Filed: 10/17/2008
• Published: 02/05/2009
• Est. Priority Date: 04/08/2004
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kumhyr, David Bruce, Childress, Rhonda L., Pennell, Neil Raymond
• Granted Patent: US 7,836,187 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: G06F 11/3419 by assessing time
• CPC Class Codes: G06F 11/3433 for load management allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/875 Monitoring of systems including the internet
• CPC Class Codes: G06F 2201/88 Monitoring involving counting
• CPC Class Codes: G06F 2201/885 Monitoring specific for caches
• CPC Class Codes: G06F 9/5033 considering data affinity
• CPC Class Codes: G06F 9/505 considering the load

Abstract

A monitoring program contains a method for detecting a load imbalance in a group of servers and providing notification so that corrections can be made. An embodiment of the invention reads from the address resolution protocol (ARP) cache to determine which server addresses are present, then clears the ARP cache. The address resolution protocol will repopulate the cache from traffic it snoops and from new requests. By performing this query and clear operation periodically, the tracking program can form a picture of the activity levels of the servers. When a significant imbalance exists, the program notifies a control console for human intervention to solve the problem. The tracking program will periodically write its results to a log file, allowing reports for various time periods to be produced as needed. By this mechanism, not only can an imbalance be quickly detected, but underutilization can be picked up by comparing known capability of a resource to its actual usage.

Claims

1-4. -4. (canceled)

5. A network, comprising:

a first server connected to the Internet and having an address resolution protocol cache;

a plurality of second servers connected to form a network with said first server, wherein said first server forwards a plurality of requests, received from the Internet, to respective ones of said plurality of second servers;

a program, running on said first server, said program performing the steps of;

preparing a first table that contains, for each of said plurality of second servers, an address and a respective counter;

dumping said address resolution process cache to a cache dump;

for each address in said cache dump, if said address in said cache dump matches an address in said first table, incrementing by one a respective counter;

comparing said plurality of counters with each other to determine if a load imbalance between said second servers is greater than a threshold value;

if said load imbalance is greater than said threshold value, sending a message to an operator console regarding said load imbalance.

6. The network of claim 5, wherein said program performs said steps repetitively and said address resolution protocol cache is cleared after each dumping step.

7. The network of claim 5, wherein said preparing step is performed initially and then only when a server is added to or removed from said plurality of second servers.

8. A computer program product comprising:

a computer recordable-type medium storing computer usable program code for monitoring a comparative load for servers within a group of servers in a network in which a first server forwards requests to a plurality of second servers, the computer program product comprising;

first instructions for preparing a first table that contains, for each of said plurality of second servers, an address and a respective counter;

second instructions for dumping a copy of an address resolution protocol cache to a storage area to form a cache dump;

third instructions for determining, from said cache dump, which of said plurality of second servers are present in said cache dump, and for each of said second servers present, incrementing a respective counter;

fourth instruction for, after said determining and incrementing are completed, comparing said plurality of respective counters with each other to determine if a load imbalance between said second servers is greater than a threshold value; and

fifth instructions for, if said load imbalance is greater than said threshold value, sending a message to an operator console regarding said load imbalance.

9. The computer program product of claim 8, wherein said program performs said instructions repetitively and said address resolution protocol cache is cleared after each dumping step.

10. The computer program product of claim 8, wherein said first instructions are performed initially and then only when a server is added to or removed from said plurality of second servers.

11. The computer program product of claim 8, and further comprising:

sixth instructions for providing a report of a relative amount of time each server of said plurality of second servers was active over a given time period.