Julia module memory and global variable issue

How to add memory feature in this reserve calculation or add a global variable type in a Julia module structure-

Code-
function reserves_calc(projection::Projection)
max_t = policy_term(projection)* 12
annual_disc_rate = 0.08
#monthly_disc_rate = maximum(BasicTerm_ME.disc_rate(t,projection)) #0.0064

    res_l = zeros(max_t + 1)  # Preallocate array for inc_in_res

    for t in max_t:-1:0
        if t == max_t
            res_l[t+1] = 0#BasicTerm_ME.reserves(t, projection)
        elseif t > proj_len(projection)[] 
            res_l[t+1] = 0.0
        else
            res_l[t+1] = max(-10000000,(val_net_cashflow(t+1, projection) .+ res_l[t+2]) / (1 .+ maximum(disc_rate(t,projection))))#BasicTerm_ME.reserves(t, projection) #- BasicTerm_ME.reserves((t-1), projection)
        end
    end
    return res_l
end


function reserves(t,projection)#, projection::Projection)
    x = reserves_calc(projection)
    return x[t]
end    

I want to use the results of res_l array in reserves function without running reserves_calc function each time.

Start by reading the reserves_calc and reserves functions shown in the issue, focusing on how res_l is created and accessed across calls. Check the Julia module scope and global-variable behavior relevant to this calculation. Done means reserves can reuse the computed res_l results without rerunning reserves_calc for every call, with the chosen state behavior demonstrated.