# The performance difference of transferring (SubArray, ReshapedArray) Array to GPU

**URL:** <https://discourse.julialang.org/t/the-performance-difference-of-transferring-subarray-reshapedarray-array-to-gpu/31303>\
**Category:** GPU\
**Tags:** flux\
**Created:** [November 20, 2019, 8:29am UTC](https://discourse.julialang.org/t/the-performance-difference-of-transferring-subarray-reshapedarray-array-to-gpu/31303 "2019-11-20T08:29:25Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![findmyway](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/findmyway/32/4946_2.png) [@findmyway](https://discourse.julialang.org/u/findmyway)\
**Post date:** [November 20, 2019, 8:29am UTC](https://discourse.julialang.org/t/the-performance-difference-of-transferring-subarray-reshapedarray-array-to-gpu/31303/1 "2019-11-20T08:29:25Z")

</div>

I encounter some performance differences when trying to move data from cpu to gpu. The result is listed in the table bellow:

| methods | cpu(x) | gpu(x) | CuArray{Float32}(x) |
| --- | --- | --- | --- |
| a = rand(Float32, 100, 100, 32); | 62.947 ns (2 allocations: 368 bytes) | 220.253 μs (12 allocations: 1.22 MiB) | 325.540 μs (12 allocations: 2.44 MiB) |
| inds = rand(1:32, 32); b = @view a[:, :, inds]; | 1.816 μs (12 allocations: 2.30 KiB) | 311.065 μs (31 allocations: 1.22 MiB) | 762.167 μs (12 allocations: 2.44 MiB) |
| c = reshape(b, 100, 100, 1, 32); | 3.110 μs (15 allocations: 2.45 KiB) | 316.978 μs (34 allocations: 1.22 MiB) | 2.144 ms (14 allocations: 2.44 MiB) |

My questions are that:

1. Why is `cpu(c)` much slower compared to `cpu(b)`?  
I thought that they are both copying the same data and shouldn’t have so much difference?
2. Why is `CuArray{Float32}(a)` slower than `gpu(a)`?
3. Why is `gpu(c)` is similar to `gpu(b)` but `CuArray{Float32}(c)` is much slower compared to `CuArray{Float32}(b)`

By the way, `gpu(c)` will trigger an error:

```julia
gpu(c)

ArgumentError: invalid index: 23.0f0 of type Float32

```

Where should I report this issue? (In which repo?)

* * *
Code to reproduce the result in the table above:

```julia
using CuArrays, BenchmarkTools, Flux
a = rand(Float32, 100, 100, 32);
inds = rand(1:32, 32)
b = @view a[:, :, inds];
c = reshape(b, 100, 100, 1, 32);

for x in [a,b,c]
@btime cpu($x);
@btime gpu($x);
@btime CuArray{Float32}($x);
end

```

---

<div class="post-metadata">

**Author:** ![maleadt](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/maleadt/32/10097_2.png) [@maleadt](https://discourse.julialang.org/u/maleadt)\
**Post date:** [November 20, 2019, 8:44am UTC](https://discourse.julialang.org/t/the-performance-difference-of-transferring-subarray-reshapedarray-array-to-gpu/31303/2 "2019-11-20T08:44:26Z")

</div>

> [@findmyway](#):
>
> By the way, `gpu(c)` will trigger an error:

Only displaying does, you actually get a proper object. You can report this at the CuArrays repo.

FWIW, `Flux.gpu` == `CuArrays.cu`. The difference between `cu` and calling the CuArray constructor is that the former adapts the object, while the latter creates a CuArray. The difference is important:

```julia
julia> typeof(b)
SubArray{Float32,3,Array{Float32,3},Tuple{Base.Slice{Base.OneTo{Int64}},Base.Slice{Base.OneTo{Int64}},Array{Int64,1}},false}

julia> typeof(cu(b))
SubArray{Float32,3,CuArray{Float32,3,Nothing},Tuple{Base.Slice{Base.OneTo{Int64}},Base.Slice{Base.OneTo{Int64}},CuArray{Float32,1,Nothing}},false}

julia> typeof(cu(CuArray{Float32}(b)))
CuArray{Float32,3,Nothing}

```

---

<div class="post-metadata">

**Author:** ![findmyway](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/findmyway/32/4946_2.png) [@findmyway](https://discourse.julialang.org/u/findmyway)\
**Post date:** [November 20, 2019, 8:50am UTC](https://discourse.julialang.org/t/the-performance-difference-of-transferring-subarray-reshapedarray-array-to-gpu/31303/3 "2019-11-20T08:50:29Z")

</div>

Now I see. Thanks for your swift reply.

[https://github.com/JuliaGPU/CuArrays.jl/issues/506](https://github.com/JuliaGPU/CuArrays.jl/issues/506)