Public service for emulation of application load based on synthetic data generation derived from preexisting models

摘要

An improved technique involves providing application workload simulation capability via a centrally located synthetic data generation service to an external party. Along these lines, a client inputs a description of an application scenario into a web form which sends a request to a central server to generate a synthetic data simulation program. The central server extracts parameters values from the request corresponding to the application scenario and finds a model in a model database based on the parameter values. From the model, the central server derives a Java program that generates synthetic data for simulating the operation load of the application scenario described by the client, and sends the Java program to the client. The client is then able to generate synthetic data using the Java program on various configurations to determine which will successfully handle the synthetic data load continuously over a given period of time.

主权项

1. A method performed by a computer of providing an external client capability of simulating application workloads on storage system devices, the method comprising:
receiving, from an external client at a remote location, a request to simulate an application workload on a storage system device resulting from an execution of an application, the request including values of a set of application workload parameters; obtaining, with reference to a Model database, an application model based on the values of the set of application workload parameters, the application model being configured to reproduce application workload conditions on the storage system device; compiling an application simulation program based on the application model; and after compiling the application simulation program, transmitting, to the external client, the application simulation program which, when executed, causes a client device to generate synthetic data that is indicative of how the storage system device performs under the application workload conditions; wherein obtaining the application model includes:
locating a first model stored in the Model database having first values of the set of application workload parameters that match the values of the set of application workload parameters of the request to within the predefined tolerance;locating a second model stored in the Model database having second values of the set of application workload parameters match the values of the set of application workload parameters of the request to within the predefined tolerance; andperforming a best fit operation on the first values and the second values to produce the application model.