Latency Hiding with Optimistic Computations

Hybinette, Maria; Fujimoto, Richard M.

doi:10.1006/jpdc.2001.1801

Cited by 2 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The VLP is responsible for sending user-defined interactions to the simulation. This method achieves real-time simulation interactions using an optimistic IO approach (Jefferson, 1985;Hybinette and Fujimoto, 2002) where the state of each agent in the SLPs is saved for every time-step that has not yet been received by the VLP. When the VLP broadcasts an event (E t ) to the SLPs, they roll back every agent to the notified time-step (t) and then progress with the simulation.…”

Section: Roll-backs For Real-time Interaction With the D Tumormentioning

confidence: 99%

“…A combination of these two methods has been discussed by Jefferson and Barnes (2017). Hybinette and Fujimoto (2002) have implemented the technique called Optimistic Input-output (OIO) which makes use of Georgia Tech Time Warp (GTW), a Time Warp system designed for shared memory processors as described by Das et al (1994). OIO focuses on latency hiding by allowing the simulation computation to go faster than the visualizer and roll back to the desired point of IO.…”

Section: Introductionmentioning

confidence: 99%

“…The state of the simulation in this example does not change as often as in a densely packed cell physics simulation. Hybinette and Fujimoto (2002) describes the two synchronization protocols as conservative where LPs wait to process events until reception of an out-of-order event is impossible (Bryant, 1977;Chandy and Misra, 1979). The optimistic protocol, in contrast, uses a detect-and-recover scheme (Jefferson, 1985).…”

Section: Introductionmentioning

confidence: 99%

“…We have deployed a dense agent-based simulation inspired by Jefferson's optimistic simulation (Jefferson, 1985;Hybinette and Fujimoto, 2002) and MMOs to support real-time user interactions with a deterministic distributed simulation. We will implement this using the standard Message Passing Interface (MPI) (Clarke et al, 1994) with N ranks, where (N − 1) ranks will be used for the simulation LPs, and 1 rank will be the visualizer LP.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Dense agent-based HPC simulation of cell physics and signaling with real-time user interactions

Merchant

Sampson

Boiko

et al. 2023

Front. Comput. Sci.

View full text Add to dashboard Cite

IntroductionDistributed simulations of complex systems to date have focused on scalability and correctness rather than interactive visualization. Interactive visual simulations have particular advantages for exploring emergent behaviors of complex systems. Interpretation of simulations of complex systems such as cancer cell tumors is a challenge and can be greatly assisted by using “built-in” real-time user interaction and subsequent visualization.MethodsWe explore this approach using a multi-scale model which couples a cell physics model with a cell signaling model. This paper presents a novel communication protocol for real-time user interaction and visualization with a large-scale distributed simulation with minimal impact on performance. Specifically, we explore how optimistic synchronization can be used to enable real-time user interaction and visualization in a densely packed parallel agent-based simulation, whilst maintaining scalability and determinism. We also describe the software framework created and the distribution strategy for the models utilized. The key features of the High-Performance Computing (HPC) simulation that were evaluated are scalability, deterministic verification, speed of real-time user interactions, and deadlock avoidance.ResultsWe use two commodity HPC systems, ARCHER (118,080 CPU cores) and ARCHER2 (750,080 CPU cores), where we simulate up to 256 million agents (one million cells) using up to 21,953 computational cores and record a response time overhead of ≃350 ms from the issued user events.DiscussionThe approach is viable and can be used to underpin transformative technologies offering immersive simulations such as Digital Twins. The framework explained in this paper is not limited to the models used and can be adapted to systems biology models that use similar standards (physics models using agent-based interactions, and signaling pathways using SBML) and other interactive distributed simulations.

show abstract

Section: Roll-backs For Real-time Interaction With the D Tumormentioning

confidence: 99%