From 9c400b403ac6b3189adba52e4784ebfde9486c52 Mon Sep 17 00:00:00 2001 From: Nicola Castelletto Date: Thu, 16 May 2024 09:50:57 -0700 Subject: [PATCH] Fixing reference --- src/docs/JOSS/paper.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/docs/JOSS/paper.md b/src/docs/JOSS/paper.md index b127bbe4ad7..808d2c7aa98 100644 --- a/src/docs/JOSS/paper.md +++ b/src/docs/JOSS/paper.md @@ -144,7 +144,7 @@ Simulations such as these play a critical role in predicting the performance of As an example of the weak scalability of GEOS on exascale systems, we present two weak scaling studies on a simple wellbore geometry using the exascale Frontier supercomputer located at Oak Ridge National Laboratory (ORNL). -The results from the weak scaling study (Figure \ref{fig:Frontier_Mechanics}a) shows flat scaling of the GEOS processes (assembly/field synchronization) up to 16,384 MPI ranks and 81.3e9 degrees-of-freedom (1/4 of Frontier). +The results from the weak scaling study (Figure \ref{fig:Frontier_scaling}a) shows flat scaling of the GEOS processes (assembly/field synchronization) up to 16,384 MPI ranks and 81.3e9 degrees-of-freedom (1/4 of Frontier). There is a moderate decrease in efficiency with the application of the hypre preconditioner setup and solve, but given the complexity of those algorithms this level of scaling efficiency is expected. The compositional flow study presented in Figure \ref{fig:Frontier_scaling}b shows similarly good weak scaling.