New GPU binaries with cuQuantum support #1854

doichanj · 2023-06-26T09:05:54Z

Summary

Change supported CUDA version from 10 to 11.2 or higher and support cuQuantum SDK.

Details and comments

Now there are separate binaries for CUDA 11 and CUDA 12:

qiskit-aer-gpu for CUDA 12 and
qiskit-aer-gpu-cu11 for CUDA 11 for backward compatibility

required libraries are automatically installed via pip for each CUDA version.
Thanks to the demo https://github.com/NVIDIA/cuQuantum/tree/main/extra/demo_build_with_wheels

hhorii · 2023-07-04T12:55:50Z

Now there are separate binaries for CUDA 11 and CUDA 12:

qiskit-aer-gpu-cu11 for CUDA 11, and
qiskit-aer-gpu-cu12 for CUDA 12

required libraries are automatically installed via pip for each CUDA version.

Will we deprecate qiskit-aer-gpu?

…cu11 for 11

mtreinish

I'm fine with that as a general approach, but I think the most important thing here is what the upgrade path is for existing users of qiskit-aer-gpu. I think we need a release note and more documentation on how an existing qiskit-aer-gpu user will upgrade after this change. I think the answer is as easy as pip uninstall qiskit-aer-gpu && pip install -U qiskit-aer-gpu-cu11, but we need to test this and make it very clear what manual steps are needed.

README.md

mtreinish · 2023-07-06T13:12:34Z

releasenotes/notes/renew_gpu_binaries-2cf3eba0853b8407.yaml

+    Changed supported CUDA version for GPU simulator to 11.2 or higher.
+    Now GPU binaries are including cuQuantum support. There are separate
+    binaries for CUDA 11 and 12. qiskit-aer-gpu-cu11 for CUDA 11 and
+    qiskit-aer-gpu-cu12 for CUDA 12.


I think it's critical we document the upgrade path for existing users of qiskit-aer-gpu here. Users already have enough difficulty navigating the gpu support in aer for whatever reason, so I think we need to explicitly document in this release note how you go from qiskit-aer-gpu to qiskit-aer-gpu-cu11 or qiskit-aer-gpu-cu12 and also put this in the README too.

mtreinish · 2023-07-06T13:14:37Z

.github/workflows/deploy.yml

    runs-on: ubuntu-latest
    steps:
+      - name: Maximize build space
+        uses: easimon/maximize-build-space@master


Are we running out of space with the newer cuda builds?

yes I observed out of space error without this setting. I tested with some parameters here and I set the safety reserved space

mtreinish · 2023-07-06T13:15:41Z

.github/workflows/deploy.yml

-          CIBW_SKIP: "*-manylinux_i686 cp310* pp* cp36* *musllinux*"
-          CIBW_TEST_SKIP: "cp*"
-          CIBW_ENVIRONMENT: QISKIT_AER_PACKAGE_NAME=qiskit-aer-gpu AER_THRUST_BACKEND=CUDA CUDACXX=/usr/local/cuda/bin/nvcc
+          CIBW_BEFORE_ALL: "yum install -y yum-utils wget && wget -q https://developer.download.nvidia.com/compute/cuda/11.8.0/local_installers/cuda-repo-rhel7-11-8-local-11.8.0_520.61.05-1.x86_64.rpm && rpm -i cuda-repo-rhel7-11-8-local-11.8.0_520.61.05-1.x86_64.rpm && yum clean all && yum -y install cuda && yum -y install openblas-devel && yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/cuda-rhel7.repo && yum clean all"


Have you done a test to validate this works in the buildwheel env? We've had a lot of issues in the past with the nvidia packages not actually setting up the compilers correctly.

Here is repository I used for tests. https://github.com/doichanj/qiskit-aer/tree/lite_GPU_package
And I tested installing and running qiskit-aer-gpu from testpypi uploaded by this one

mtreinish · 2023-07-06T13:17:54Z

setup.py

+            "nvidia-cuda-runtime-cu11>=11.8.89",
+            "nvidia-cublas-cu11>=11.11.3.6",
+            "nvidia-cusolver-cu11>=11.4.1.48",
+            "nvidia-cusparse-cu11>=11.7.5.86",
+            "cuquantum-cu11>=23.3.0",


So these nvidia packages include the dynamic library files for all the cuda symbols we need which is why you exclude them from the repair wheel command?

Yes this is the trick to decrease the binary size

releasenotes/notes/renew_gpu_binaries-2cf3eba0853b8407.yaml

Co-authored-by: Matthew Treinish <mtreinish@kortar.org>

mtreinish · 2023-07-11T14:25:14Z

releasenotes/notes/renew_gpu_binaries-2cf3eba0853b8407.yaml

+  - |
+    Changed supported CUDA version for GPU simulator to 11.2 or higher.
+    Now GPU binaries are including cuQuantum support. There are separate
+    binaries for CUDA 11 and 12. qiskit-aer-gpu-cu11 for CUDA 11 and
+    qiskit-aer-gpu-cu12 for CUDA 12.


Suggested change

- |

Changed supported CUDA version for GPU simulator to 11.2 or higher.

Now GPU binaries are including cuQuantum support. There are separate

binaries for CUDA 11 and 12. qiskit-aer-gpu-cu11 for CUDA 11 and

qiskit-aer-gpu-cu12 for CUDA 12.

- |

Qiskit Aer now requires CUDA version for GPU simulator to 11.2 or

higher. Previously, CUDA 10.1 was the minimum supported version.

This change was necessary because of changes in the upstream CUDA

ecosystem, including cuQuantum support. To support users running

with different versions of CUDA there is now a separate package available

for running with CUDA 11: ``qiskit-aer-gpu-cu11`` and using the

``qiskit-aer-gpu`` package now requires CUDA 12. If you're an existing

user of the ``qiskit-aer-gpu`` package and want to use CUDA 11 you will need to run::

pip uninstall qiskit-aer-gpu && pip install -U qiskit-aer-gpu-cu11

to go from the previously CUDA 10.x compatible ``qiskit-aer-gpu``

package's releases to upgrade to the new CUDA 11 compatible

package. If you're running CUDA 12 locally already you can upgrade

the ``qiskit-aer-gpu`` package as normal.

I updated release note

* New GPU binaries with cuQuantum support * format setup.py * remove testpipy settings * Update setup.py to use numpy with version lower than 1.25 * changed package names, qiskit-aer-gpu for CUDA12, and qiskit-aer-gpu-cu11 for 11 * format setup.py * Update README.md Co-authored-by: Matthew Treinish <mtreinish@kortar.org> * Update releasenotes/notes/renew_gpu_binaries-2cf3eba0853b8407.yaml Co-authored-by: Matthew Treinish <mtreinish@kortar.org> * update release note --------- Co-authored-by: Hiroshi Horii <hhorii@users.noreply.github.com> Co-authored-by: Matthew Treinish <mtreinish@kortar.org>

* bump version strings and prepare for release * Remove `AerCircuit` from result of `backend.run()` (#1845) * take AerCircuit from _metadata in experiment result * add reno * Fix rST formatting --------- Co-authored-by: Jake Lishman <jake@binhbar.com> * pin numpy version lower than 1.25 (#1859) * Fix param positions for circuits with conditionals (#1851) * fix parameter indexing issues. * fix issues in parameterization of global phase * add reno * fix lint * fix release note * fix a bug to call _convert_circuit_binds * add test * Support save_statevector for QuantumCircuit from QASM3 string (#1846) * Support save_statevector for QuantumCircuit from QASM3 string * add qiskit_qasm3_import to requirements-dev.txt * simplify default_qubits() * fix lint error * skip qasm3 circuit test in 3.7 * New GPU binaries with cuQuantum support (#1854) * New GPU binaries with cuQuantum support * format setup.py * remove testpipy settings * Update setup.py to use numpy with version lower than 1.25 * changed package names, qiskit-aer-gpu for CUDA12, and qiskit-aer-gpu-cu11 for 11 * format setup.py * Update README.md Co-authored-by: Matthew Treinish <mtreinish@kortar.org> * Update releasenotes/notes/renew_gpu_binaries-2cf3eba0853b8407.yaml Co-authored-by: Matthew Treinish <mtreinish@kortar.org> * update release note --------- Co-authored-by: Hiroshi Horii <hhorii@users.noreply.github.com> Co-authored-by: Matthew Treinish <mtreinish@kortar.org> * bump up python version from 3.7 to 3.8 to run cibuildwheel (#1867) * bump up python version from 3.7 to 3.8 to run cibuildwheel * bump up all python version in deploy workflow * add reno * set numpy version constraint only in dev (#1871) * add numpy version constraint only in dev * set numpy version constraint in contraints.txt * bump up version in docs --------- Co-authored-by: Jake Lishman <jake@binhbar.com> Co-authored-by: Jun Doi <doichan@jp.ibm.com> Co-authored-by: Matthew Treinish <mtreinish@kortar.org>

* New GPU binaries with cuQuantum support * format setup.py * remove testpipy settings * Update setup.py to use numpy with version lower than 1.25 * changed package names, qiskit-aer-gpu for CUDA12, and qiskit-aer-gpu-cu11 for 11 * format setup.py * Update README.md Co-authored-by: Matthew Treinish <mtreinish@kortar.org> * Update releasenotes/notes/renew_gpu_binaries-2cf3eba0853b8407.yaml Co-authored-by: Matthew Treinish <mtreinish@kortar.org> * update release note --------- Co-authored-by: Hiroshi Horii <hhorii@users.noreply.github.com> Co-authored-by: Matthew Treinish <mtreinish@kortar.org>

doichanj added 3 commits June 26, 2023 18:00

New GPU binaries with cuQuantum support

f3e8884

format setup.py

1aafa5c

remove testpipy settings

ba06c61

doichanj requested a review from hhorii June 27, 2023 02:24

hhorii added the stable-backport-potential The issue or PR might be minimal and/or import enough to backport to stable label Jul 5, 2023

hhorii added this to the Aer 0.12.2 milestone Jul 5, 2023

hhorii added 2 commits July 5, 2023 23:27

Merge branch 'main' into renew_gpu_binaries

d328d26

Update setup.py to use numpy with version lower than 1.25

c6ae7ba

hhorii mentioned this pull request Jul 6, 2023

Update Qiskit-aer-gpu to 0.12 #1852

Closed

doichanj and others added 4 commits July 6, 2023 16:28

Merge remote-tracking branch 'upstream/main' into renew_gpu_binaries

2429be9

changed package names, qiskit-aer-gpu for CUDA12, and qiskit-aer-gpu-…

2dd3abb

…cu11 for 11

format setup.py

2fa0d45

Merge branch 'main' into renew_gpu_binaries

a847dc1

mtreinish reviewed Jul 6, 2023

View reviewed changes

doichanj and others added 2 commits July 7, 2023 10:23

Update README.md

cfb4c10

Co-authored-by: Matthew Treinish <mtreinish@kortar.org>

Update releasenotes/notes/renew_gpu_binaries-2cf3eba0853b8407.yaml

4a8a263

Co-authored-by: Matthew Treinish <mtreinish@kortar.org>

hhorii mentioned this pull request Jul 7, 2023

Release 0.12.2 #1864

Merged

mtreinish reviewed Jul 11, 2023

View reviewed changes

doichanj added 4 commits July 11, 2023 23:36

Merge branch 'main' into renew_gpu_binaries

49990f7

Merge remote-tracking branch 'upstream/main' into renew_gpu_binaries

302f9a0

update release note

ec15941

update release note

1fbe30a

hhorii approved these changes Jul 11, 2023

View reviewed changes

hhorii merged commit 3024b24 into Qiskit:main Jul 11, 2023
34 checks passed

wshanks mentioned this pull request Jul 11, 2023

Support CUDA build conda-forge/qiskit-aer-feedstock#5

Closed

wshanks mentioned this pull request Dec 12, 2023

Enable CUDA builds conda-forge/qiskit-aer-feedstock#19

Merged

5 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

New GPU binaries with cuQuantum support #1854

New GPU binaries with cuQuantum support #1854

doichanj commented Jun 26, 2023 •

edited

Loading

hhorii commented Jul 4, 2023

mtreinish left a comment

mtreinish Jul 6, 2023

mtreinish Jul 6, 2023

doichanj Jul 7, 2023

mtreinish Jul 6, 2023

doichanj Jul 7, 2023

mtreinish Jul 6, 2023

doichanj Jul 7, 2023

mtreinish Jul 11, 2023 •

edited by doichanj

Loading

doichanj Jul 11, 2023

-  - |
-    Changed supported CUDA version for GPU simulator to 11.2 or higher.
-    Now GPU binaries are including cuQuantum support. There are separate
-    binaries for CUDA 11 and 12. qiskit-aer-gpu-cu11 for CUDA 11 and
-    qiskit-aer-gpu-cu12 for CUDA 12.
+  - |
+    Qiskit Aer now requires CUDA version for GPU simulator to 11.2 or
+    higher.  Previously, CUDA 10.1 was the minimum supported version.
+    This change was necessary because of changes in the upstream CUDA
+    ecosystem, including cuQuantum support. To support users running
+    with different versions of CUDA there is now a separate package available
+    for running with CUDA 11:  ``qiskit-aer-gpu-cu11`` and using the
+    ``qiskit-aer-gpu`` package now requires CUDA 12. If you're an existing
+    user of the ``qiskit-aer-gpu`` package and want to use CUDA 11 you will need to run::
+        pip uninstall qiskit-aer-gpu && pip install -U qiskit-aer-gpu-cu11
+    to go from the previously CUDA 10.x compatible ``qiskit-aer-gpu``
+    package's releases to upgrade to the new CUDA 11 compatible
+    package. If you're running CUDA 12 locally already you can upgrade
+    the ``qiskit-aer-gpu`` package as normal.

New GPU binaries with cuQuantum support #1854

New GPU binaries with cuQuantum support #1854

Conversation

doichanj commented Jun 26, 2023 • edited Loading

Summary

Details and comments

hhorii commented Jul 4, 2023

mtreinish left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mtreinish Jul 11, 2023 • edited by doichanj Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

doichanj commented Jun 26, 2023 •

edited

Loading

mtreinish Jul 11, 2023 •

edited by doichanj

Loading