# Optimizing DiffEq Code

##### Chris Rackauckas

In this notebook we will walk through some of the main tools for optimizing your code in order to efficiently solve DifferentialEquations.jl. User-side optimizations are important because, for sufficiently difficult problems, most of the time will be spent inside of your f function, the function you are trying to solve. "Efficient" integrators are those that reduce the required number of f calls to hit the error tolerance. The main ideas for optimizing your DiffEq code, or any Julia function, are the following:

• Make it non-allocating

• Use StaticArrays for small arrays

• Make it type-stable

• Reduce redundant calculations

• Make use of BLAS calls

• Optimize algorithm choice

We'll discuss these strategies in the context of small and large systems. Let's start with small systems.

## Optimizing Small Systems (<100 DEs)

Let's take the classic Lorenz system from before. Let's start by naively writing the system in its out-of-place form:

function lorenz(u,p,t)
dx = 10.0*(u-u)
dy = u*(28.0-u) - u
dz = u*u - (8/3)*u
[dx,dy,dz]
end

lorenz (generic function with 1 method)


Here, lorenz returns an object, [dx,dy,dz], which is created within the body of lorenz.

This is a common code pattern from high-level languages like MATLAB, SciPy, or R's deSolve. However, the issue with this form is that it allocates a vector, [dx,dy,dz], at each step. Let's benchmark the solution process with this choice of function:

using DifferentialEquations, BenchmarkTools
u0 = [1.0;0.0;0.0]
tspan = (0.0,100.0)
prob = ODEProblem(lorenz,u0,tspan)
@benchmark solve(prob,Tsit5())

BenchmarkTools.Trial:
memory estimate:  10.81 MiB
allocs estimate:  100152
--------------
minimum time:     3.613 ms (0.00% GC)
median time:      3.744 ms (0.00% GC)
mean time:        4.838 ms (12.95% GC)
maximum time:     13.302 ms (67.82% GC)
--------------
samples:          1031
evals/sample:     1


The BenchmarkTools.jl package's @benchmark runs the code multiple times to get an accurate measurement. The minimum time is the time it takes when your OS and other background processes aren't getting in the way. Notice that in this case it takes about 5ms to solve and allocates around 11.11 MiB. However, if we were to use this inside of a real user code we'd see a lot of time spent doing garbage collection (GC) to clean up all of the arrays we made. Even if we turn off saving we have these allocations.

@benchmark solve(prob,Tsit5(),save_everystep=false)

BenchmarkTools.Trial:
memory estimate:  9.47 MiB
allocs estimate:  88645
--------------
minimum time:     3.233 ms (0.00% GC)
median time:      3.348 ms (0.00% GC)
mean time:        4.425 ms (11.89% GC)
maximum time:     11.764 ms (51.89% GC)
--------------
samples:          1127
evals/sample:     1


The problem of course is that arrays are created every time our derivative function is called. This function is called multiple times per step and is thus the main source of memory usage. To fix this, we can use the in-place form to ***make our code non-allocating***:

function lorenz!(du,u,p,t)
du = 10.0*(u-u)
du = u*(28.0-u) - u
du = u*u - (8/3)*u
end

lorenz! (generic function with 1 method)


Here, instead of creating an array each time, we utilized the cache array du. When the inplace form is used, DifferentialEquations.jl takes a different internal route that minimizes the internal allocations as well. When we benchmark this function, we will see quite a difference.

u0 = [1.0;0.0;0.0]
tspan = (0.0,100.0)
prob = ODEProblem(lorenz!,u0,tspan)
@benchmark solve(prob,Tsit5())

BenchmarkTools.Trial:
memory estimate:  1.37 MiB
allocs estimate:  11752
--------------
minimum time:     770.421 μs (0.00% GC)
median time:      792.791 μs (0.00% GC)
mean time:        936.532 μs (8.20% GC)
maximum time:     7.417 ms (84.98% GC)
--------------
samples:          5308
evals/sample:     1

@benchmark solve(prob,Tsit5(),save_everystep=false)

BenchmarkTools.Trial:
memory estimate:  6.70 KiB
allocs estimate:  47
--------------
minimum time:     345.686 μs (0.00% GC)
median time:      350.476 μs (0.00% GC)
mean time:        351.953 μs (0.12% GC)
maximum time:     4.719 ms (91.27% GC)
--------------
samples:          10000
evals/sample:     1


There is a 4x time difference just from that change! Notice there are still some allocations and this is due to the construction of the integration cache. But this doesn't scale with the problem size:

tspan = (0.0,500.0) # 5x longer than before
prob = ODEProblem(lorenz!,u0,tspan)
@benchmark solve(prob,Tsit5(),save_everystep=false)

BenchmarkTools.Trial:
memory estimate:  6.70 KiB
allocs estimate:  47
--------------
minimum time:     1.729 ms (0.00% GC)
median time:      1.749 ms (0.00% GC)
mean time:        1.753 ms (0.00% GC)
maximum time:     5.282 ms (0.00% GC)
--------------
samples:          2848
evals/sample:     1


since that's all just setup allocations.

#### But if the system is small we can optimize even more.

Allocations are only expensive if they are "heap allocations". For a more in-depth definition of heap allocations, there are a lot of sources online. But a good working definition is that heap allocations are variable-sized slabs of memory which have to be pointed to, and this pointer indirection costs time. Additionally, the heap has to be managed and the garbage controllers has to actively keep track of what's on the heap.

However, there's an alternative to heap allocations, known as stack allocations. The stack is statically-sized (known at compile time) and thus its accesses are quick. Additionally, the exact block of memory is known in advance by the compiler, and thus re-using the memory is cheap. This means that allocating on the stack has essentially no cost!

Arrays have to be heap allocated because their size (and thus the amount of memory they take up) is determined at runtime. But there are structures in Julia which are stack-allocated. structs for example are stack-allocated "value-type"s. Tuples are a stack-allocated collection. The most useful data structure for DiffEq though is the StaticArray from the package StaticArrays.jl. These arrays have their length determined at compile-time. They are created using macros attached to normal array expressions, for example:

using StaticArrays
A = @SVector [2.0,3.0,5.0]

3-element StaticArrays.SVector{3, Float64} with indices SOneTo(3):
2.0
3.0
5.0


Notice that the 3 after SVector gives the size of the SVector. It cannot be changed. Additionally, SVectors are immutable, so we have to create a new SVector to change values. But remember, we don't have to worry about allocations because this data structure is stack-allocated. SArrays have a lot of extra optimizations as well: they have fast matrix multiplication, fast QR factorizations, etc. which directly make use of the information about the size of the array. Thus, when possible they should be used.

Unfortunately static arrays can only be used for sufficiently small arrays. After a certain size, they are forced to heap allocate after some instructions and their compile time balloons. Thus static arrays shouldn't be used if your system has more than 100 variables. Additionally, only the native Julia algorithms can fully utilize static arrays.

Let's ***optimize lorenz using static arrays***. Note that in this case, we want to use the out-of-place allocating form, but this time we want to output a static array:

function lorenz_static(u,p,t)
dx = 10.0*(u-u)
dy = u*(28.0-u) - u
dz = u*u - (8/3)*u
@SVector [dx,dy,dz]
end

lorenz_static (generic function with 1 method)


To make the solver internally use static arrays, we simply give it a static array as the initial condition:

u0 = @SVector [1.0,0.0,0.0]
tspan = (0.0,100.0)
prob = ODEProblem(lorenz_static,u0,tspan)
@benchmark solve(prob,Tsit5())

BenchmarkTools.Trial:
memory estimate:  446.73 KiB
allocs estimate:  1314
--------------
minimum time:     313.937 μs (0.00% GC)
median time:      320.696 μs (0.00% GC)
mean time:        348.313 μs (4.60% GC)
maximum time:     5.044 ms (90.94% GC)
--------------
samples:          10000
evals/sample:     1

@benchmark solve(prob,Tsit5(),save_everystep=false)

BenchmarkTools.Trial:
memory estimate:  3.69 KiB
allocs estimate:  22
--------------
minimum time:     194.038 μs (0.00% GC)
median time:      196.887 μs (0.00% GC)
mean time:        200.723 μs (0.00% GC)
maximum time:     3.633 ms (0.00% GC)
--------------
samples:          10000
evals/sample:     1


And that's pretty much all there is to it. With static arrays you don't have to worry about allocating, so use operations like * and don't worry about fusing operations (discussed in the next section). Do "the vectorized code" of R/MATLAB/Python and your code in this case will be fast, or directly use the numbers/values.

#### Exercise 1

Implement the out-of-place array, in-place array, and out-of-place static array forms for the Henon-Heiles System and time the results.

## Optimizing Large Systems

### Interlude: Managing Allocations with Broadcast Fusion

When your system is sufficiently large, or you have to make use of a non-native Julia algorithm, you have to make use of Arrays. In order to use arrays in the most efficient manner, you need to be careful about temporary allocations. Vectorized calculations naturally have plenty of temporary array allocations. This is because a vectorized calculation outputs a vector. Thus:

A = rand(1000,1000); B = rand(1000,1000); C = rand(1000,1000)
test(A,B,C) = A + B + C
@benchmark test(A,B,C)

BenchmarkTools.Trial:
memory estimate:  7.63 MiB
allocs estimate:  2
--------------
minimum time:     1.176 ms (0.00% GC)
median time:      1.240 ms (0.00% GC)
mean time:        1.558 ms (10.94% GC)
maximum time:     4.484 ms (34.51% GC)
--------------
samples:          3142
evals/sample:     1


That expression A + B + C creates 2 arrays. It first creates one for the output of A + B, then uses that result array to + C to get the final result. 2 arrays! We don't want that! The first thing to do to fix this is to use broadcast fusion. Broadcast fusion puts expressions together. For example, instead of doing the + operations separately, if we were to add them all at the same time, then we would only have a single array that's created. For example:

test2(A,B,C) = map((a,b,c)->a+b+c,A,B,C)
@benchmark test2(A,B,C)

BenchmarkTools.Trial:
memory estimate:  7.63 MiB
allocs estimate:  2
--------------
minimum time:     1.104 ms (0.00% GC)
median time:      1.233 ms (0.00% GC)
mean time:        1.570 ms (10.76% GC)
maximum time:     4.938 ms (31.12% GC)
--------------
samples:          3118
evals/sample:     1


Puts the whole expression into a single function call, and thus only one array is required to store output. This is the same as writing the loop:

function test3(A,B,C)
D = similar(A)
@inbounds for i in eachindex(A)
D[i] = A[i] + B[i] + C[i]
end
D
end
@benchmark test3(A,B,C)

BenchmarkTools.Trial:
memory estimate:  7.63 MiB
allocs estimate:  2
--------------
minimum time:     1.170 ms (0.00% GC)
median time:      1.232 ms (0.00% GC)
mean time:        1.552 ms (11.03% GC)
maximum time:     11.761 ms (46.68% GC)
--------------
samples:          3155
evals/sample:     1


However, Julia's broadcast is syntactic sugar for this. If multiple expressions have a ., then it will put those vectorized operations together. Thus:

test4(A,B,C) = A .+ B .+ C
@benchmark test4(A,B,C)

BenchmarkTools.Trial:
memory estimate:  7.63 MiB
allocs estimate:  2
--------------
minimum time:     1.180 ms (0.00% GC)
median time:      1.253 ms (0.00% GC)
mean time:        1.564 ms (10.94% GC)
maximum time:     4.450 ms (35.00% GC)
--------------
samples:          3131
evals/sample:     1


is a version with only 1 array created (the output). Note that .s can be used with function calls as well:

sin.(A) .+ sin.(B)

1000×1000 Matrix{Float64}:
0.720069  0.182457  0.763895  1.04962   …  0.41588    0.508065  0.361555
0.896943  1.41017   0.409182  0.825458     0.812476   0.785988  0.881462
1.36059   0.394225  1.11242   0.538726     0.714643   1.38476   1.34051
0.540476  0.817361  0.600552  0.446012     1.16538    0.786801  0.927957
1.58465   0.709142  0.872294  0.615889     0.834811   0.323699  0.773447
0.911922  1.34667   0.887216  1.35896   …  0.677299   1.02647   1.43337
1.27911   0.269067  0.829162  0.930278     0.273203   0.904372  0.682316
0.927549  0.788235  0.469908  1.18655      0.264846   0.634813  1.18439
0.781568  0.695038  1.13002   0.837556     1.10733    0.840558  1.10162
1.15982   1.01843   1.02094   0.590594     1.09131    0.489211  0.643781
⋮                                       ⋱
1.15489   1.13873   1.25706   1.09846      0.836945   1.45453   0.668872
1.50925   0.339285  0.63122   1.31708      0.0284809  0.675261  0.414872
1.21772   0.933767  1.1721    1.125        0.539415   0.336139  0.796335
0.62414   0.907673  0.24076   1.06522      1.28061    0.663797  1.10831
0.420111  1.07505   0.910339  1.5706    …  0.879288   0.96245   0.602083
0.60329   1.41653   1.03081   0.840123     0.344945   1.26708   0.810059
0.430383  0.407107  0.242699  0.941784     1.09034    0.614127  0.883191
0.942736  0.90306   0.415315  0.892401     1.07394    0.934953  0.346971
0.441236  1.33947   1.39745   0.718495     0.417654   0.733623  1.45458


Also, the @. macro applys a dot to every operator:

test5(A,B,C) = @. A + B + C #only one array allocated
@benchmark test5(A,B,C)

BenchmarkTools.Trial:
memory estimate:  7.63 MiB
allocs estimate:  2
--------------
minimum time:     1.171 ms (0.00% GC)
median time:      1.266 ms (0.00% GC)
mean time:        1.564 ms (11.02% GC)
maximum time:     6.052 ms (79.63% GC)
--------------
samples:          3130
evals/sample:     1


Using these tools we can get rid of our intermediate array allocations for many vectorized function calls. But we are still allocating the output array. To get rid of that allocation, we can instead use mutation. Mutating broadcast is done via .=. For example, if we pre-allocate the output:

D = zeros(1000,1000);


Then we can keep re-using this cache for subsequent calculations. The mutating broadcasting form is:

test6!(D,A,B,C) = D .= A .+ B .+ C #only one array allocated
@benchmark test6!(D,A,B,C)

BenchmarkTools.Trial:
memory estimate:  0 bytes
allocs estimate:  0
--------------
minimum time:     1.197 ms (0.00% GC)
median time:      1.210 ms (0.00% GC)
mean time:        1.220 ms (0.00% GC)
maximum time:     1.425 ms (0.00% GC)
--------------
samples:          3991
evals/sample:     1


If we use @. before the =, then it will turn it into .=:

test7!(D,A,B,C) = @. D = A + B + C #only one array allocated
@benchmark test7!(D,A,B,C)

BenchmarkTools.Trial:
memory estimate:  0 bytes
allocs estimate:  0
--------------
minimum time:     1.190 ms (0.00% GC)
median time:      1.204 ms (0.00% GC)
mean time:        1.213 ms (0.00% GC)
maximum time:     1.325 ms (0.00% GC)
--------------
samples:          4013
evals/sample:     1


Notice that in this case, there is no "output", and instead the values inside of D are what are changed (like with the DiffEq inplace function). Many Julia functions have a mutating form which is denoted with a !. For example, the mutating form of the map is map!:

test8!(D,A,B,C) = map!((a,b,c)->a+b+c,D,A,B,C)
@benchmark test8!(D,A,B,C)

BenchmarkTools.Trial:
memory estimate:  0 bytes
allocs estimate:  0
--------------
minimum time:     2.340 ms (0.00% GC)
median time:      2.371 ms (0.00% GC)
mean time:        2.374 ms (0.00% GC)
maximum time:     4.860 ms (0.00% GC)
--------------
samples:          2078
evals/sample:     1


Some operations require using an alternate mutating form in order to be fast. For example, matrix multiplication via * allocates a temporary:

@benchmark A*B

BenchmarkTools.Trial:
memory estimate:  7.63 MiB
allocs estimate:  2
--------------
minimum time:     9.142 ms (0.00% GC)
median time:      9.285 ms (0.00% GC)
mean time:        10.756 ms (5.28% GC)
maximum time:     59.078 ms (75.02% GC)
--------------
samples:          465
evals/sample:     1


Instead, we can use the mutating form mul! into a cache array to avoid allocating the output:

using LinearAlgebra
@benchmark mul!(D,A,B) # same as D = A * B

BenchmarkTools.Trial:
memory estimate:  0 bytes
allocs estimate:  0
--------------
minimum time:     9.368 ms (0.00% GC)
median time:      9.445 ms (0.00% GC)
mean time:        9.496 ms (0.00% GC)
maximum time:     15.280 ms (0.00% GC)
--------------
samples:          527
evals/sample:     1


For repeated calculations this reduced allocation can stop GC cycles and thus lead to more efficient code. Additionally, ***we can fuse together higher level linear algebra operations using BLAS***. The package SugarBLAS.jl makes it easy to write higher level operations like alpha*B*A + beta*C as mutating BLAS calls.

### Example Optimization: Gierer-Meinhardt Reaction-Diffusion PDE Discretization

Let's optimize the solution of a Reaction-Diffusion PDE's discretization. In its discretized form, this is the ODE:

\begin{align} du &= D_1 (A_y u + u A_x) + \frac{au^2}{v} + \bar{u} - \alpha u\\ dv &= D_2 (A_y v + v A_x) + a u^2 + \beta v \end{align}

where $u$, $v$, and $A$ are matrices. Here, we will use the simplified version where $A$ is the tridiagonal stencil $[1,-2,1]$, i.e. it's the 2D discretization of the LaPlacian. The native code would be something along the lines of:

# Generate the constants
p = (1.0,1.0,1.0,10.0,0.001,100.0) # a,α,ubar,β,D1,D2
N = 100
Ax = Array(Tridiagonal([1.0 for i in 1:N-1],[-2.0 for i in 1:N],[1.0 for i in 1:N-1]))
Ay = copy(Ax)
Ax[2,1] = 2.0
Ax[end-1,end] = 2.0
Ay[1,2] = 2.0
Ay[end,end-1] = 2.0

function basic_version!(dr,r,p,t)
a,α,ubar,β,D1,D2 = p
u = r[:,:,1]
v = r[:,:,2]
Du = D1*(Ay*u + u*Ax)
Dv = D2*(Ay*v + v*Ax)
dr[:,:,1] = Du .+ a.*u.*u./v .+ ubar .- α*u
dr[:,:,2] = Dv .+ a.*u.*u .- β*v
end

a,α,ubar,β,D1,D2 = p
uss = (ubar+β)/α
vss = (a/β)*uss^2
r0 = zeros(100,100,2)
r0[:,:,1] .= uss.+0.1.*rand.()
r0[:,:,2] .= vss

prob = ODEProblem(basic_version!,r0,(0.0,0.1),p)

ODEProblem with uType Array{Float64, 3} and tType Float64. In-place: true
timespan: (0.0, 0.1)
u0: 100×100×2 Array{Float64, 3}:
[:, :, 1] =
11.0535  11.0553  11.0402  11.0053  …  11.0975  11.0065  11.0951  11.0041
11.0283  11.0561  11.0491  11.0346     11.0782  11.0436  11.0835  11.0784
11.0295  11.0513  11.0118  11.094      11.0985  11.0237  11.0473  11.0772
11.0145  11.0466  11.0063  11.0786     11.0292  11.0336  11.0165  11.0419
11.0513  11.0845  11.0594  11.0967     11.0168  11.0135  11.0916  11.0869
11.0839  11.0767  11.071   11.0097  …  11.0897  11.083   11.006   11.016
11.0555  11.0919  11.0273  11.0584     11.0101  11.0598  11.091   11.0167
11.0541  11.0224  11.0639  11.0231     11.0663  11.0907  11.0749  11.0622
11.0221  11.0687  11.0463  11.0196     11.0641  11.0938  11.0106  11.0371
11.0166  11.0454  11.0653  11.0541     11.0252  11.0313  11.0899  11.0836
⋮                                  ⋱
11.0299  11.0717  11.0751  11.0816     11.09    11.0345  11.0006  11.0903
11.0981  11.0425  11.0727  11.0055     11.0366  11.0544  11.0733  11.0173
11.0215  11.0698  11.0698  11.0454     11.051   11.0632  11.0058  11.0155
11.0634  11.0611  11.0109  11.0358     11.0083  11.0335  11.0634  11.0081
11.0737  11.0231  11.047   11.0946  …  11.0005  11.0105  11.0803  11.0531
11.0879  11.002   11.0728  11.0653     11.093   11.0359  11.0358  11.0107
11.0922  11.0321  11.0863  11.0364     11.0522  11.052   11.0148  11.0803
11.0329  11.0417  11.0288  11.0093     11.0713  11.0173  11.0276  11.0738
11.0884  11.0505  11.0854  11.0922     11.0139  11.0249  11.031   11.0494

[:, :, 2] =
12.1  12.1  12.1  12.1  12.1  12.1  …  12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1  …  12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
⋮                             ⋮    ⋱         ⋮
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1  …  12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1
12.1  12.1  12.1  12.1  12.1  12.1     12.1  12.1  12.1  12.1  12.1  12.1


In this version we have encoded our initial condition to be a 3-dimensional array, with u[:,:,1] being the A part and u[:,:,2] being the B part.

@benchmark solve(prob,Tsit5())

BenchmarkTools.Trial:
memory estimate:  186.90 MiB
allocs estimate:  7348
--------------
minimum time:     71.670 ms (3.61% GC)
median time:      74.963 ms (6.90% GC)
mean time:        76.086 ms (6.73% GC)
maximum time:     114.990 ms (10.78% GC)
--------------
samples:          66
evals/sample:     1


While this version isn't very efficient,

#### We recommend writing the "high-level" code first, and iteratively optimizing it!

The first thing that we can do is get rid of the slicing allocations. The operation r[:,:,1] creates a temporary array instead of a "view", i.e. a pointer to the already existing memory. To make it a view, add @view. Note that we have to be careful with views because they point to the same memory, and thus changing a view changes the original values:

A = rand(4)
@show A
B = @view A[1:3]
B = 2
@show A

A = [0.08839464722234869, 0.4314571825499718, 0.4874308762081436, 0.8265892
248309634]
A = [0.08839464722234869, 2.0, 0.4874308762081436, 0.8265892248309634]
4-element Vector{Float64}:
0.08839464722234869
2.0
0.4874308762081436
0.8265892248309634


Notice that changing B changed A. This is something to be careful of, but at the same time we want to use this since we want to modify the output dr. Additionally, the last statement is a purely element-wise operation, and thus we can make use of broadcast fusion there. Let's rewrite basic_version! to ***avoid slicing allocations*** and to ***use broadcast fusion***:

function gm2!(dr,r,p,t)
a,α,ubar,β,D1,D2 = p
u = @view r[:,:,1]
v = @view r[:,:,2]
du = @view dr[:,:,1]
dv = @view dr[:,:,2]
Du = D1*(Ay*u + u*Ax)
Dv = D2*(Ay*v + v*Ax)
@. du = Du + a.*u.*u./v + ubar - α*u
@. dv = Dv + a.*u.*u - β*v
end
prob = ODEProblem(gm2!,r0,(0.0,0.1),p)
@benchmark solve(prob,Tsit5())

BenchmarkTools.Trial:
memory estimate:  119.76 MiB
allocs estimate:  5878
--------------
minimum time:     49.146 ms (5.53% GC)
median time:      62.558 ms (4.49% GC)
mean time:        61.436 ms (6.35% GC)
maximum time:     94.152 ms (13.61% GC)
--------------
samples:          82
evals/sample:     1


Now, most of the allocations are taking place in Du = D1*(Ay*u + u*Ax) since those operations are vectorized and not mutating. We should instead replace the matrix multiplications with mul!. When doing so, we will need to have cache variables to write into. This looks like:

Ayu = zeros(N,N)
uAx = zeros(N,N)
Du = zeros(N,N)
Ayv = zeros(N,N)
vAx = zeros(N,N)
Dv = zeros(N,N)
function gm3!(dr,r,p,t)
a,α,ubar,β,D1,D2 = p
u = @view r[:,:,1]
v = @view r[:,:,2]
du = @view dr[:,:,1]
dv = @view dr[:,:,2]
mul!(Ayu,Ay,u)
mul!(uAx,u,Ax)
mul!(Ayv,Ay,v)
mul!(vAx,v,Ax)
@. Du = D1*(Ayu + uAx)
@. Dv = D2*(Ayv + vAx)
@. du = Du + a*u*u./v + ubar - α*u
@. dv = Dv + a*u*u - β*v
end
prob = ODEProblem(gm3!,r0,(0.0,0.1),p)
@benchmark solve(prob,Tsit5())

BenchmarkTools.Trial:
memory estimate:  29.98 MiB
allocs estimate:  4702
--------------
minimum time:     50.864 ms (0.00% GC)
median time:      52.639 ms (0.00% GC)
mean time:        53.631 ms (1.66% GC)
maximum time:     57.861 ms (7.23% GC)
--------------
samples:          94
evals/sample:     1


But our temporary variables are global variables. We need to either declare the caches as const or localize them. We can localize them by adding them to the parameters, p. It's easier for the compiler to reason about local variables than global variables. ***Localizing variables helps to ensure type stability***.

p = (1.0,1.0,1.0,10.0,0.001,100.0,Ayu,uAx,Du,Ayv,vAx,Dv) # a,α,ubar,β,D1,D2
function gm4!(dr,r,p,t)
a,α,ubar,β,D1,D2,Ayu,uAx,Du,Ayv,vAx,Dv = p
u = @view r[:,:,1]
v = @view r[:,:,2]
du = @view dr[:,:,1]
dv = @view dr[:,:,2]
mul!(Ayu,Ay,u)
mul!(uAx,u,Ax)
mul!(Ayv,Ay,v)
mul!(vAx,v,Ax)
@. Du = D1*(Ayu + uAx)
@. Dv = D2*(Ayv + vAx)
@. du = Du + a*u*u./v + ubar - α*u
@. dv = Dv + a*u*u - β*v
end
prob = ODEProblem(gm4!,r0,(0.0,0.1),p)
@benchmark solve(prob,Tsit5())

BenchmarkTools.Trial:
memory estimate:  29.66 MiB
allocs estimate:  1027
--------------
minimum time:     46.163 ms (0.00% GC)
median time:      50.698 ms (0.00% GC)
mean time:        51.970 ms (1.86% GC)
maximum time:     80.866 ms (11.54% GC)
--------------
samples:          97
evals/sample:     1


We could then use the BLAS gemmv to optimize the matrix multiplications some more, but instead let's devectorize the stencil.

p = (1.0,1.0,1.0,10.0,0.001,100.0,N)
function fast_gm!(du,u,p,t)
a,α,ubar,β,D1,D2,N = p

@inbounds for j in 2:N-1, i in 2:N-1
du[i,j,1] = D1*(u[i-1,j,1] + u[i+1,j,1] + u[i,j+1,1] + u[i,j-1,1] - 4u[i,j,1]) +
a*u[i,j,1]^2/u[i,j,2] + ubar - α*u[i,j,1]
end

@inbounds for j in 2:N-1, i in 2:N-1
du[i,j,2] = D2*(u[i-1,j,2] + u[i+1,j,2] + u[i,j+1,2] + u[i,j-1,2] - 4u[i,j,2]) +
a*u[i,j,1]^2 - β*u[i,j,2]
end

@inbounds for j in 2:N-1
i = 1
du[1,j,1] = D1*(2u[i+1,j,1] + u[i,j+1,1] + u[i,j-1,1] - 4u[i,j,1]) +
a*u[i,j,1]^2/u[i,j,2] + ubar - α*u[i,j,1]
end
@inbounds for j in 2:N-1
i = 1
du[1,j,2] = D2*(2u[i+1,j,2] + u[i,j+1,2] + u[i,j-1,2] - 4u[i,j,2]) +
a*u[i,j,1]^2 - β*u[i,j,2]
end
@inbounds for j in 2:N-1
i = N
du[end,j,1] = D1*(2u[i-1,j,1] + u[i,j+1,1] + u[i,j-1,1] - 4u[i,j,1]) +
a*u[i,j,1]^2/u[i,j,2] + ubar - α*u[i,j,1]
end
@inbounds for j in 2:N-1
i = N
du[end,j,2] = D2*(2u[i-1,j,2] + u[i,j+1,2] + u[i,j-1,2] - 4u[i,j,2]) +
a*u[i,j,1]^2 - β*u[i,j,2]
end

@inbounds for i in 2:N-1
j = 1
du[i,1,1] = D1*(u[i-1,j,1] + u[i+1,j,1] + 2u[i,j+1,1] - 4u[i,j,1]) +
a*u[i,j,1]^2/u[i,j,2] + ubar - α*u[i,j,1]
end
@inbounds for i in 2:N-1
j = 1
du[i,1,2] = D2*(u[i-1,j,2] + u[i+1,j,2] + 2u[i,j+1,2] - 4u[i,j,2]) +
a*u[i,j,1]^2 - β*u[i,j,2]
end
@inbounds for i in 2:N-1
j = N
du[i,end,1] = D1*(u[i-1,j,1] + u[i+1,j,1] + 2u[i,j-1,1] - 4u[i,j,1]) +
a*u[i,j,1]^2/u[i,j,2] + ubar - α*u[i,j,1]
end
@inbounds for i in 2:N-1
j = N
du[i,end,2] = D2*(u[i-1,j,2] + u[i+1,j,2] + 2u[i,j-1,2] - 4u[i,j,2]) +
a*u[i,j,1]^2 - β*u[i,j,2]
end

@inbounds begin
i = 1; j = 1
du[1,1,1] = D1*(2u[i+1,j,1] + 2u[i,j+1,1] - 4u[i,j,1]) +
a*u[i,j,1]^2/u[i,j,2] + ubar - α*u[i,j,1]
du[1,1,2] = D2*(2u[i+1,j,2] + 2u[i,j+1,2] - 4u[i,j,2]) +
a*u[i,j,1]^2 - β*u[i,j,2]

i = 1; j = N
du[1,N,1] = D1*(2u[i+1,j,1] + 2u[i,j-1,1] - 4u[i,j,1]) +
a*u[i,j,1]^2/u[i,j,2] + ubar - α*u[i,j,1]
du[1,N,2] = D2*(2u[i+1,j,2] + 2u[i,j-1,2] - 4u[i,j,2]) +
a*u[i,j,1]^2 - β*u[i,j,2]

i = N; j = 1
du[N,1,1] = D1*(2u[i-1,j,1] + 2u[i,j+1,1] - 4u[i,j,1]) +
a*u[i,j,1]^2/u[i,j,2] + ubar - α*u[i,j,1]
du[N,1,2] = D2*(2u[i-1,j,2] + 2u[i,j+1,2] - 4u[i,j,2]) +
a*u[i,j,1]^2 - β*u[i,j,2]

i = N; j = N
du[end,end,1] = D1*(2u[i-1,j,1] + 2u[i,j-1,1] - 4u[i,j,1]) +
a*u[i,j,1]^2/u[i,j,2] + ubar - α*u[i,j,1]
du[end,end,2] = D2*(2u[i-1,j,2] + 2u[i,j-1,2] - 4u[i,j,2]) +
a*u[i,j,1]^2 - β*u[i,j,2]
end
end
prob = ODEProblem(fast_gm!,r0,(0.0,0.1),p)
@benchmark solve(prob,Tsit5())

BenchmarkTools.Trial:
memory estimate:  29.63 MiB
allocs estimate:  439
--------------
minimum time:     6.669 ms (0.00% GC)
median time:      6.818 ms (0.00% GC)
mean time:        8.679 ms (12.16% GC)
maximum time:     18.755 ms (34.68% GC)
--------------
samples:          574
evals/sample:     1


Lastly, we can do other things like multithread the main loops, but these optimizations get the last 2x-3x out. The main optimizations which apply everywhere are the ones we just performed (though the last one only works if your matrix is a stencil. This is known as a matrix-free implementation of the PDE discretization).

This gets us to about 8x faster than our original MATLAB/SciPy/R vectorized style code!

The last thing to do is then ***optimize our algorithm choice***. We have been using Tsit5() as our test algorithm, but in reality this problem is a stiff PDE discretization and thus one recommendation is to use CVODE_BDF(). However, instead of using the default dense Jacobian, we should make use of the sparse Jacobian afforded by the problem. The Jacobian is the matrix $\frac{df_i}{dr_j}$, where $r$ is read by the linear index (i.e. down columns). But since the $u$ variables depend on the $v$, the band size here is large, and thus this will not do well with a Banded Jacobian solver. Instead, we utilize sparse Jacobian algorithms. CVODE_BDF allows us to use a sparse Newton-Krylov solver by setting linear_solver = :GMRES (see the solver documentation, and thus we can solve this problem efficiently. Let's see how this scales as we increase the integration time.

prob = ODEProblem(fast_gm!,r0,(0.0,10.0),p)
@benchmark solve(prob,Tsit5())

BenchmarkTools.Trial:
memory estimate:  2.76 GiB
allocs estimate:  39336
--------------
minimum time:     1.033 s (37.02% GC)
median time:      1.294 s (43.35% GC)
mean time:        1.296 s (42.87% GC)
maximum time:     1.563 s (47.48% GC)
--------------
samples:          4
evals/sample:     1

using Sundials
@benchmark solve(prob,CVODE_BDF(linear_solver=:GMRES))

BenchmarkTools.Trial:
memory estimate:  120.29 MiB
allocs estimate:  20210
--------------
minimum time:     666.229 ms (0.48% GC)
median time:      668.269 ms (0.48% GC)
mean time:        668.412 ms (0.48% GC)
maximum time:     670.411 ms (0.47% GC)
--------------
samples:          8
evals/sample:     1

prob = ODEProblem(fast_gm!,r0,(0.0,100.0),p)
# Will go out of memory if we don't turn off save_everystep!
@benchmark solve(prob,Tsit5(),save_everystep=false)

BenchmarkTools.Trial:
memory estimate:  2.91 MiB
allocs estimate:  67
--------------
minimum time:     4.247 s (0.00% GC)
median time:      4.249 s (0.00% GC)
mean time:        4.249 s (0.00% GC)
maximum time:     4.250 s (0.00% GC)
--------------
samples:          2
evals/sample:     1

@benchmark solve(prob,CVODE_BDF(linear_solver=:GMRES))

BenchmarkTools.Trial:
memory estimate:  302.44 MiB
allocs estimate:  52631
--------------
minimum time:     1.805 s (1.83% GC)
median time:      1.807 s (1.82% GC)
mean time:        1.859 s (4.09% GC)
maximum time:     1.966 s (9.93% GC)
--------------
samples:          3
evals/sample:     1


Now let's check the allocation growth.

@benchmark solve(prob,CVODE_BDF(linear_solver=:GMRES),save_everystep=false)

BenchmarkTools.Trial:
memory estimate:  3.46 MiB
allocs estimate:  44125
--------------
minimum time:     1.738 s (0.00% GC)
median time:      1.744 s (0.00% GC)
mean time:        1.754 s (0.00% GC)
maximum time:     1.779 s (0.00% GC)
--------------
samples:          3
evals/sample:     1

prob = ODEProblem(fast_gm!,r0,(0.0,500.0),p)
@benchmark solve(prob,CVODE_BDF(linear_solver=:GMRES),save_everystep=false)

BenchmarkTools.Trial:
memory estimate:  4.44 MiB
allocs estimate:  61057
--------------
minimum time:     2.410 s (0.00% GC)
median time:      2.413 s (0.00% GC)
mean time:        2.412 s (0.00% GC)
maximum time:     2.413 s (0.00% GC)
--------------
samples:          3
evals/sample:     1


Notice that we've elimated almost all allocations, allowing the code to grow without hitting garbage collection and slowing down.

Why is CVODE_BDF doing well? What's happening is that, because the problem is stiff, the number of steps required by the explicit Runge-Kutta method grows rapidly, whereas CVODE_BDF is taking large steps. Additionally, the GMRES linear solver form is quite an efficient way to solve the implicit system in this case. This is problem-dependent, and in many cases using a Krylov method effectively requires a preconditioner, so you need to play around with testing other algorithms and linear solvers to find out what works best with your problem.

## Conclusion

Julia gives you the tools to optimize the solver "all the way", but you need to make use of it. The main thing to avoid is temporary allocations. For small systems, this is effectively done via static arrays. For large systems, this is done via in-place operations and cache arrays. Either way, the resulting solution can be immensely sped up over vectorized formulations by using these principles.

## Appendix

These tutorials are a part of the SciMLTutorials.jl repository, found at: https://github.com/SciML/SciMLTutorials.jl. For more information on high-performance scientific machine learning, check out the SciML Open Source Software Organization https://sciml.ai.

To locally run this tutorial, do the following commands:

using SciMLTutorials
SciMLTutorials.weave_file("tutorials/introduction","03-optimizing_diffeq_code.jmd")

Computer Information:

Julia Version 1.6.1
Commit 6aaedecc44 (2021-04-23 05:59 UTC)
Platform Info:
OS: Linux (x86_64-pc-linux-gnu)
CPU: AMD EPYC 7502 32-Core Processor
WORD_SIZE: 64
LIBM: libopenlibm
LLVM: libLLVM-11.0.1 (ORCJIT, znver2)
Environment:
JULIA_DEPOT_PATH = /root/.cache/julia-buildkite-plugin/depots/a6029d3a-f78b-41ea-bc97-28aa57c6c6ea


Package Information:

      Status /var/lib/buildkite-agent/builds/6-amdci4-julia-csail-mit-edu/julialang/scimltutorials/tutorials/introduction/Project.toml
[6e4b80f9] BenchmarkTools v1.0.0
[0c46a032] DifferentialEquations v6.17.1
[65888b18] ParameterizedFunctions v5.10.0
[91a5bcdd] Plots v1.15.2
[30cb0354] SciMLTutorials v0.9.0
[90137ffa] StaticArrays v1.2.0
[37e2e46d] LinearAlgebra

And the full manifest:

      Status /var/lib/buildkite-agent/builds/6-amdci4-julia-csail-mit-edu/julialang/scimltutorials/tutorials/introduction/Manifest.toml
[c3fe647b] AbstractAlgebra v0.16.0
[1520ce14] AbstractTrees v0.3.4
[ec485272] ArnoldiMethod v0.1.0
[4fba245c] ArrayInterface v3.1.15
[4c555306] ArrayLayouts v0.7.0
[aae01518] BandedMatrices v0.16.9
[6e4b80f9] BenchmarkTools v1.0.0
[764a87c0] BoundaryValueDiffEq v2.7.1
[fa961155] CEnum v0.4.1
[d360d2e6] ChainRulesCore v0.9.44
[35d6a980] ColorSchemes v3.12.1
[3da002f7] ColorTypes v0.11.0
[5ae59095] Colors v0.12.8
[861a8166] Combinatorics v1.0.2
[38540f10] CommonSolve v0.2.0
[bbf7d656] CommonSubexpressions v0.3.0
[34da2185] Compat v3.30.0
[8f4d0f93] Conda v1.5.2
[187b0558] ConstructionBase v1.2.1
[d38c429a] Contour v0.5.7
[9a962f9c] DataAPI v1.6.0
[864edb3b] DataStructures v0.18.9
[e2d170a0] DataValueInterfaces v1.0.0
[bcd4f6db] DelayDiffEq v5.31.0
[2b5f629d] DiffEqBase v6.62.2
[459566f4] DiffEqCallbacks v2.16.1
[5a0ffddc] DiffEqFinancial v2.4.0
[c894b116] DiffEqJump v6.14.2
[77a26b50] DiffEqNoiseProcess v5.7.3
[055956cb] DiffEqPhysics v3.9.0
[163ba53b] DiffResults v1.0.3
[b552c78f] DiffRules v1.0.2
[0c46a032] DifferentialEquations v6.17.1
[c619ae07] DimensionalPlotRecipes v1.2.0
[b4f34e82] Distances v0.10.3
[31c24e10] Distributions v0.24.18
[ffbed154] DocStringExtensions v0.8.4
[d4d017d3] ExponentialUtilities v1.8.4
[e2ba6199] ExprTools v0.1.3
[c87230d0] FFMPEG v0.4.0
[9aa1b823] FastClosures v0.3.2
[1a297f60] FillArrays v0.11.7
[6a86dc24] FiniteDiff v2.8.0
[53c48c17] FixedPointNumbers v0.8.4
 Formatting v0.4.2
[f6369f11] ForwardDiff v0.10.18
[069b7b12] FunctionWrappers v1.1.2
[28b8d3ca] GR v0.57.4
[5c1252a2] GeometryBasics v0.3.12
[42e2da0e] Grisu v1.0.2
[cd3eb016] HTTP v0.9.9
[eafb193a] Highlights v0.4.5
[0e44f5e4] Hwloc v2.0.0
[7073ff75] IJulia v1.23.2
[615f187c] IfElse v0.1.0
[d25df0c9] Inflate v0.1.2
[83e8ac13] IniFile v0.5.0
[c8e1da08] IterTools v1.3.0
[42fd0dbc] IterativeSolvers v0.9.1
 IteratorInterfaceExtensions v1.0.0
[692b3bcd] JLLWrappers v1.3.0
[682c06a0] JSON v0.21.1
[b964fa9f] LaTeXStrings v1.2.1
[2ee39098] LabelledArrays v1.6.1
[23fbe1c1] Latexify v0.15.5
[093fc24a] LightGraphs v1.3.5
[d3d80556] LineSearches v7.1.1
[2ab3a3ac] LogExpFunctions v0.2.4
[bdcacae8] LoopVectorization v0.12.23
[1914dd2f] MacroTools v0.5.6
[739be429] MbedTLS v1.0.3
[442fdcdd] Measures v0.3.1
[e1d29d7a] Missings v1.0.0
[961ee093] ModelingToolkit v5.16.0
[f9640e96] MultiScaleArrays v1.8.1
[ffc61752] Mustache v1.0.10
[d41bc354] NLSolversBase v7.8.0
[2774e3e8] NLsolve v4.5.1
[77ba4419] NaNMath v0.3.5
[8913a72c] NonlinearSolve v0.3.8
[6fe1bfb0] OffsetArrays v1.9.0
[429524aa] Optim v1.3.0
[bac558e1] OrderedCollections v1.4.1
[1dea7af3] OrdinaryDiffEq v5.56.0
[90014a1f] PDMats v0.11.0
[65888b18] ParameterizedFunctions v5.10.0
[d96e819e] Parameters v0.12.2
[69de0a69] Parsers v1.1.0
[995b91a9] PlotUtils v1.0.10
[91a5bcdd] Plots v1.15.2
[e409e4f3] PoissonRandom v0.4.0
[f517fe37] Polyester v0.3.1
[85a6dd25] PositiveFactorizations v0.2.4
[21216c6a] Preferences v1.2.2
 Random123 v1.3.1
[fb686558] RandomExtensions v0.4.3
[e6cf234a] RandomNumbers v1.4.0
[3cdcf5f2] RecipesBase v1.1.1
[01d81517] RecipesPipeline v0.3.2
[731186ca] RecursiveArrayTools v2.11.4
[f2c3362d] RecursiveFactorization v0.1.12
[189a3867] Reexport v1.0.0
[ae029012] Requires v1.1.3
[ae5879a3] ResettableStacks v1.1.0
[79098fc4] Rmath v0.7.0
[7e49a35a] RuntimeGeneratedFunctions v0.5.2
[476501e8] SLEEFPirates v0.6.20
[1bc83da4] SafeTestsets v0.0.1
[0bca4576] SciMLBase v1.13.4
[30cb0354] SciMLTutorials v0.9.0
[6c6a2e73] Scratch v1.0.3
[efcf1570] Setfield v0.7.0
[992d4aef] Showoff v1.0.3
[699a6c99] SimpleTraits v0.9.3
[b85f4697] SoftGlobalScope v1.1.0
[a2af1166] SortingAlgorithms v1.0.0
[47a9eef4] SparseDiffTools v1.13.2
[276daf66] SpecialFunctions v1.4.1
[aedffcd0] Static v0.2.4
[90137ffa] StaticArrays v1.2.0
[82ae8749] StatsAPI v1.0.0
[2913bbd2] StatsBase v0.33.8
[4c63d2b9] StatsFuns v0.9.8
[789caeaf] StochasticDiffEq v6.34.1
[7792a7ef] StrideArraysCore v0.1.11
[09ab397b] StructArrays v0.5.1
[d1185830] SymbolicUtils v0.11.2
[0c5d862f] Symbolics v0.1.25
[3783bdb8] TableTraits v1.0.1
[bd369af6] Tables v1.4.2
[a759f4b9] TimerOutputs v0.5.9
[a2a6695c] TreeViews v0.3.0
[5c2747f8] URIs v1.3.0
[3a884ed6] UnPack v1.0.2
[1986cc42] Unitful v1.7.0
[3d5dd08c] VectorizationBase v0.20.11
[81def892] VersionParsing v1.2.0
[19fa3120] VertexSafeGraphs v0.1.2
[44d3d7a6] Weave v0.10.8
[ddb6d928] YAML v0.4.6
[c2297ded] ZMQ v1.2.1
[700de1a5] ZygoteRules v0.2.1
[6e34b625] Bzip2_jll v1.0.6+5
[83423d85] Cairo_jll v1.16.0+6
[5ae413db] EarCut_jll v2.1.5+1
[2e619515] Expat_jll v2.2.10+0
[b22a6f82] FFMPEG_jll v4.3.1+4
[a3f928ae] Fontconfig_jll v2.13.1+14
[d7e528f0] FreeType2_jll v2.10.1+5
[559328eb] FriBidi_jll v1.0.5+6
[0656b61e] GLFW_jll v3.3.4+0
[d2c73de3] GR_jll v0.57.2+0
[78b55507] Gettext_jll v0.21.0+0
[7746bdde] Glib_jll v2.68.1+0
[e33a78d0] Hwloc_jll v2.4.1+0
[aacddb02] JpegTurbo_jll v2.0.1+3
[c1c5ebd0] LAME_jll v3.100.0+3
[dd4b983a] LZO_jll v2.10.1+0
[dd192d2f] LibVPX_jll v1.9.0+1
[e9f186c6] Libffi_jll v3.2.2+0
[d4300ac3] Libgcrypt_jll v1.8.7+0
[7e76a0d4] Libglvnd_jll v1.3.0+3
[94ce4f54] Libiconv_jll v1.16.1+0
[4b2f31a3] Libmount_jll v2.35.0+0
[89763e89] Libtiff_jll v4.1.0+2
[38a345b3] Libuuid_jll v2.36.0+0
[e7412a2a] Ogg_jll v1.3.4+2
[458c3c95] OpenSSL_jll v1.1.1+6
[efe28fd5] OpenSpecFun_jll v0.5.4+0
[91d4177d] Opus_jll v1.3.1+3
[2f80f16e] PCRE_jll v8.44.0+0
 Pixman_jll v0.40.1+0
[ea2cea3b] Qt5Base_jll v5.15.2+0
[f50d1b31] Rmath_jll v0.3.0+0
[fb77eaff] Sundials_jll v5.2.0+1
[a2964d1f] Wayland_jll v1.17.0+4
[2381bf8a] Wayland_protocols_jll v1.18.0+4
[02c8fc9c] XML2_jll v2.9.12+0
[aed1982a] XSLT_jll v1.1.34+0
[4f6342f7] Xorg_libX11_jll v1.6.9+4
[0c0b7dd1] Xorg_libXau_jll v1.0.9+4
[935fb764] Xorg_libXcursor_jll v1.2.0+4
[a3789734] Xorg_libXdmcp_jll v1.1.3+4
[1082639a] Xorg_libXext_jll v1.3.4+4
[d091e8ba] Xorg_libXfixes_jll v5.0.3+4
[a51aa0fd] Xorg_libXi_jll v1.7.10+4
[d1454406] Xorg_libXinerama_jll v1.1.4+4
[ec84b674] Xorg_libXrandr_jll v1.5.2+4
[ea2f1a96] Xorg_libXrender_jll v0.9.10+4
[c7cfdc94] Xorg_libxcb_jll v1.13.0+3
[cc61e674] Xorg_libxkbfile_jll v1.1.0+4
 Xorg_xcb_util_image_jll v0.4.0+1
[2def613f] Xorg_xcb_util_jll v0.4.0+1
[975044d2] Xorg_xcb_util_keysyms_jll v0.4.0+1
[0d47668e] Xorg_xcb_util_renderutil_jll v0.3.9+1
[c22f9ab0] Xorg_xcb_util_wm_jll v0.4.1+1
 Xorg_xkbcomp_jll v1.4.2+4
[33bec58e] Xorg_xkeyboard_config_jll v2.27.0+4
[c5fb5394] Xorg_xtrans_jll v1.4.0+3
[8f1865be] ZeroMQ_jll v4.3.2+6
[3161d3a3] Zstd_jll v1.5.0+0
[0ac62f75] libass_jll v0.14.0+4
[f638f0a6] libfdk_aac_jll v0.1.6+4
[b53b4c65] libpng_jll v1.6.38+0
[a9144af2] libsodium_jll v1.0.20+0
[f27f6e37] libvorbis_jll v1.3.6+6
[1270edf5] x264_jll v2020.7.14+2
[dfaa095f] x265_jll v3.0.0+3
[d8fb68d0] xkbcommon_jll v0.9.1+5
[56f22d72] Artifacts
[2a0f44e3] Base64
[8bb1440f] DelimitedFiles
[8ba89e20] Distributed
[7b1f6079] FileWatching
[9fa8497b] Future
[b77e0a4c] InteractiveUtils
[b27032c2] LibCURL
[76f85450] LibGit2
[8f399da3] Libdl
[37e2e46d] LinearAlgebra
[56ddb016] Logging
[d6f4376e] Markdown
[ca575930] NetworkOptions
[44cfe95a] Pkg
[de0858da] Printf
[3fa0cd96] REPL
[9a3f8284] Random
[ea8e919c] SHA
[9e88b42a] Serialization
[1a1011a3] SharedArrays
[6462fe0b] Sockets
[2f01184e] SparseArrays
[10745b16] Statistics
[4607b0f0] SuiteSparse
[fa267f1f] TOML
[a4e569a6] Tar
[8dfed614] Test
[cf7118a7] UUIDs
[4ec0a83e] Unicode
[e66e0078] CompilerSupportLibraries_jll
[deac9b47] LibCURL_jll
[29816b5a] LibSSH2_jll
[c8ffd9c3] MbedTLS_jll
[14a3606d] MozillaCACerts_jll
[4536629a] OpenBLAS_jll
[bea87d4a] SuiteSparse_jll
[83775a58] Zlib_jll
[8e850ede] nghttp2_jll
[3f19e933] p7zip_jll