RM: threads loops whenever not needed for solving the problem. Making…

… the instantiation slightly longer but reducing RAM usage significantly.
lanl-ansi · Aug 15, 2024 · 52f7322 · 52f7322
1 parent 872516d
commit 52f7322
Showing 1 changed file with 4 additions and 2 deletions.
diff --git a/src/core/base.jl b/src/core/base.jl
@@ -359,7 +359,8 @@ function instantiate_model_decomposition(
     master_boundary_vars_vector = Vector{Vector{Vector{JuMP.VariableRef}}}(undef, number_of_subproblems)
 
     # Threaded loop for instantiating subproblems
-    Threads.@threads for i in 1:1:number_of_subproblems
+    # Threads.@threads for i in 1:1:number_of_subproblems # faster but causes +RAM usage
+    for i in 1:1:number_of_subproblems
         ckt_name = ckts_names_vector[i]
         ckt_data = ckts_data_vector[i]
 
@@ -491,7 +492,8 @@ function instantiate_model_decomposition(
     boundary_vars_vector = Vector{Vector{Vector{JuMP.VariableRef}}}(undef, number_of_subproblems)
 
     # Threaded loop for instantiating subproblems
-    Threads.@threads for i in 1:number_of_subproblems
+    # Threads.@threads for i in 1:1:number_of_subproblems # faster but causes +RAM usage
+    for i in 1:number_of_subproblems
         ckt_name = ckts_names_vector[i]
         ckt_data = ckts_data_vector[i]