Automated capacity provisioning method using historical performance data

A system for automatically allocating resources in a computer system, the system comprising:

• at least one processor;
• a non-transitory computer-readable medium storing instructions, when executed by the at least one processor, are configured to implement,
• a data collection tool configured to collect performance data relating to processing nodes of a computer system which provide services via one or more applications;
• a recommendation tool configured to analyze the performance data to generate an operational profile characterizing resource usage of the processing nodes over at least one time period,
• the recommendation tool configured to receive a set of attributes characterizing expected performance goals in which the services are expected to be provided between a provider and a consumer, the set of attributes including an attribute defining a response time for transactions to be below a time value for one or more of the processing nodes,
• the recommendation tool configured to generate at least one provisioning policy based on an analysis of the operational profile in conjunction with the set of attributes, the at least one provisioning policy specifying a condition for re-allocating resources associated with at least one processing node in a manner that satisfies the performance goals of the set of attributes, the at least one provisioning policy indicating to allocate at least one additional server when a response time of the one or more processing nodes is greater than the response time defined by the attribute;
• and
• a provisional tool configured to re-allocate, during runtime of the computer system, the resources associated with the at least one processing node when the condition of the at least one provisioning policy is determined as satisfied including allocating the at least one additional server such that the response time of the one or more processing nodes is less than or equal to the response time defined by the attribute,
• wherein if the expected performance goals cannot be achieved, the resources are re-allocated to distribute the transactions based on calculated recommended arrival rates of the processing nodes.

A non-transitory computer-readable medium storing instructions, when executed by at least one processor, are configured to perform the following operations:

• collect performance data relating to processing nodes of the computer system which provide services via one or more applications;
• analyze the performance data to generate an operational profile characterizing resource usage of the processing nodes over at least one time period;
• receive a set of attributes characterizing expected performance goals in which the services are expected to be provided between a provider and a consumer, the set of attributes including an attribute defining a response time for transactions to be below a time value for one or more of the processing nodes;
• generate at least one provisioning policy based on an analysis of the operational profile in conjunction with the set of attributes, the at least one provisioning policy specifying a condition for re-allocating resources associated with at least one processing node in a manner that satisfies the performance goals of the set of attributes, the at least one provisioning policy indicating to allocate at least one additional server when a response time of the one or more processing nodes is greater than the response time defined by the attribute;
• and
• re-allocate, during runtime of the computer system, the resources associated with the at least one processing node when the condition of the at least one provisioning policy is determined as satisfied including allocating the at least one additional server such that the response time of the one or more processing nodes is less than or equal to the response time defined by the attribute,
• wherein if the expected performance goals cannot be achieved, the resources are re-allocated to distribute the transactions based on calculated recommended arrival rates of the processing nodes.