Cloning

Cloned simulations of epidemiological outbreaks like COVID-19 can now rapidly evaluate millions of what-if scenarios at national and world scales, and efficiently exploit 1000s of GPUs.

Cloned eXecution of simulations is a powerful decision-making system to incrementally evaluate millions of what-if scenarios and scale to 1000s of GPUs.

A mesoscopic modeling approach is described that strikes a middle ground between macroscopic models based on coupled differential equations and microscopic models built on fine-grained behaviors at the individual entity level. Execution of our implementation scaled to 8192 GPUs of supercomputing platforms demonstrates the ability to rapidly evaluate what–if scenarios several orders of magnitude faster than the conventional methods.

This seminar conveys some directions and technical ideas being pursued by the Discrete Computing Systems Group of the Computer Science and Mathematics Division, in collaboration with others at the lab. Some of the scalable computational tools ready for applying to challenging computational problems are presented. Actual working codes ready for customization to COVID-related efforts are described, which are built for scaling to supercomputing platforms such as Summit.

HPC facilities used for scientific computing draw enormous energy, consuming many megawatt-hours. Using simulation cloning, we reduce energy consumption and track the power drawn by thousands of GPUs, achieving significant aggregate energy savings from cloned simulations.

We propose a new redundancy reduction technique for large-scale discrete event simulations, called exact-differential simulation, which simulates only the altered portions of scenarios and their influences in repeated executions while still achieving the same results as the re-execution of entire simulations. We evaluate our approach by using two case studies, PHOLD benchmark and a traffic simulation of Tokyo.

Cloning is a technique to efficiently simulate a tree of multiple what-if scenarios that are unraveled during the course of a base simulation. We present the conceptual simulation framework, algorithmic foundations, and runtime interface of CloneX, a new system we designed for scalable simulation cloning.