Method to identify performance and capacity bottlenecks of complex systems

A method for determining capacity in complex systems by setting capacity values for service transaction types that assist in identifying performance bottlenecks comprising:

• listing the service transaction types;
• identifying Workflow Manager (WFM) data where invocation data records for each of the service transaction types are located;
• retrieving the WFM data;
• parsing the WFM data to acquire the invocation data records for each of the service transaction types;
• for a given transaction type;
• determining average response time RTh aggregation levels and volume counts counth, wherein the volume counts counth is the traffic load and is a count of successful transactions within a predetermined measurement interval h and the average response time RTh is calculated for successful transactions within the predetermined measurement interval h;
• performing a linear regression of data pairs (counth, RTh);
• determining a Response Time Model RTM (x) to be
• z
• 0
• +
• z
• 1
• 1
• -
• z
• 2
• ⁢
• x
• ,
• wherein the RTM (x) models the average response time RTh of a specific function for a given load (x) as the sum of the sojourn time in each service node required to process the function, wherein z0 is the expected amount of time to traverse non-bottlenecked nodes, z1 is the expected amount of time to traverse a bottleneck node under no transaction load and z2 is the average amount of time of a bottleneck node per transaction type;
• estimating the response time model z parameters z0, z1 and z2 from the data pairs (counth, RTh)
• calculating a maximum throughput
• 1
• z
• 2
• that the complex system can handle;
• calculating a capacity warning value (linear knee) L1;
• calculating a capacity limit value (quadratic knee) L2;
• and
• analyzing a time trend of the volume counts counth to determine when the capacity warning value L1 and the capacity limit value L2 will be reached if the current volume counts counth trends continue.