The system is modeled as a network of functions that exchange data through message queues. For example, a function represents users that send/get requests or post requests to access web applications. The number of users is a parameter whose value can be tuned to control the amount of requests that are sent. The percentage of post and get requests is also controlled. Requests are based on payload information and do not contain real data. These requests are routed either to the cloud front or a warehouse that contains web servers, application servers, and databases. The number of cloud fronts, web servers, application servers, and databases can be scaled based on Intel® CoFluent™ model parameters. All elements in the Intel® CoFluent™ model are annotated with processing speed, memory size, power consumption, and cost information.
See how Intel® CoFluent™ studio reduces costs and development time modeling capacity, memory, and more.
Models and simulations measure the impact of architecture decisions before hardware is available.