Rocket CE for Power

To: Valued Open-Source Community and Users of RocketCE and Rocket AI Hub for Power We are writing to inform you that we have made the decision to end support for the following products, effective 31 December 2025.RocketCE:These are a set of conda packages optimized for the IBM Power Platform​. Rocket was building/distributing/supporting the packages for IBM Power Platform based on https://github.com/open-ce/open-ce. Packages have been available here: https://anaconda.org/rocketceRocket AI Hub for Power:Rocket AI Hub is an MLOps Platform Optimized for IBM Power. Rocket was building/distributing/supporting the packages based on

Hello everyone! First of all, I would like to thank the creators and maintainers of RocketCE! My organization had acquired an HPC cluster based on IBM Power9, and RocketCE is the main way we can use it for ML-based research. We have recently started to look into quantization, and plan to use the official torchao library. However, torchao only supports recent versions of PyTorch (at least 2.5 for the current torchao 0.13, at least 2.3 for the very first version, torchao 0.0.1!). However, RocketCE includes PyTorch 2.1. I was wondering whether there are any plans to update the PyTorch version included in RocketCE, or whether anyone has managed to run any version of torchao with PyTorch 2.1. Thanks a lot, best! Francesco------------------------------Francesco MalandrinoRocket Software Forum Member------------------------------

As part of this Rocket provides the AI/ML conda packages for Power machines.The RocketCE conda packages are build and supported for Power 9 and Power 10 with ppc64le architecture. These packages are installed into a conda environment so after installation is complete, the frameworks are ready for use. Presently these packages are available in RocketCE conda channel.This version of packages are available for Python 3.10 and 3.11 and for CUDA 11.8 and 12.2.Note: Access and availability to this data is subject to the community and can change at any time.We are releasing RocketCE packages based on OpenCE v1.11.2.To know more about RocketCE and the various packages rel

Overview  XGBoost is an optimized distributed gradient boosting library designed to be highly efficient, flexible, and portable. It implements machine learning algorithms under the Gradient Boosting framework.   This release of RocketCE is based on Release OpenCE v1.11.2 and it includes XGBoost 2.0.3.  These packages are supported for both Power 9 and Power 10

Overview  ONNX Runtime is a cross-platform inference and training machine-learning accelerator.  ONNX Runtime is an accelerator for machine learning models with multi-platform support and a flexible interface to integrate with hardware-specific libraries. ONNX Runtime can be used with models from PyTorch, Tensorflow/Keras, TFLite, scikit-learn, and other frameworks.  This release of RocketCE is based on Release OpenCE v1.11.2 and it includes ONNX Runtime v1.16.3.  These packages are supported for both Power 9 and Power 10 Sys

Overview  RocketCE Pytorch includes support for IBM's Distributed Deep Learning (DDL).   This version of Pytorch will have the support of matrix-multiply assist (MMA) which was introduced in Power 10.  This release of RocketCE is based on Release OpenCE v1.11.2 and it includes Pytorch 2.1.2. CPU version optimized for Power10 pytorch-cpu 2.1.2 of Python 3.11 pytorch-cpu 2.1.2 of Python 3.10 Cuda versions For Cuda 12.2 pytorch 2.1.2 of Python 3.11 pytorch 2.1.2 of Python 3.10 Cuda 11.8 pytorch 2.1.2 of Python 3.11 pytorch 2.1.2 of Python 3.10 These packages are su

Overview  TensorFlow is an open source library to help you develop and train machine learning models.   RocketCE includes several packages from the TensorFlow ecosystem and these packages are available in channel rocketce of anaconda.org.   First release of RocketCE is based on OpenCE Release v1.11.2. In this version of RocketCE, there is TensorFlow 2.14.1. 

Hi, does anyone here have experience with https://github.com/openai/whisper, I tried to transcribe mp3 length 13:35 min  on different  hardware  with following results  : i used LARGE MODEL :  1)  AC922 GPU 1xV100 utilization (U)=30 % ,  transcription time  (TS) = 5:38 min 2) AC922 32xP9 CPU   U=90 + %, TS= 80 min 3) 1050 p311  pytorch-cpu 2.1.1  8CPU  TS=28 min   U = more less 4.5 cores only few threads work 4) 1050 p311  pytorch-cpu 2.1.1  12CPU   TS=25 min in 4)  i used  export OPENBLAS_NUM_THREADS=12export GOTO_NUM_THREADS=12export OMP_NUM_THREADS=12 I found out that by incrasing the number of threads (export OPENBLAS....)  I can increase CPU utilization but this does not shorten the transition time , but rather increases it Is it possible to get clos

Overview  RocketCE Pytorch includes support for IBM's Distributed Deep Learning (DDL).   This version of Pytorch will have the support of matrix-multiply assist (MMA) which was introduced in Power 10.  This release of RocketCE is based on Release OpenCE v1.9.1 and it includes Pytorch 2.0.1.  These packages are supported for both Power 9 and Power 10 Systems of IBM with linux-ppc64le.     CPU Support  The CPU-only variant of Pytorch is built without GPU and CUDA support. It has a s

Hi，I try to run a model training task on a power8 CPU + GPU server. I used provided env yaml to create my conda enviroment, the installation is successful, but when I attempt to run Python or any Python-related package, I encounter illegal instruction errors. I check the python-config, the CFLAG for python3.9 is -mcpu=power9 -mtune=power10 -mcpu=power9 -mtune=power10, however my cpu is power 8, this might be the problem.  I checked conda package provided by RocketCE， there is only one python3.9 package, which is the one i'm using.  Here are some details of my problem.

Starting to build this env. to support RocketCE for Power and got the following ?'s  Downloaded -- Anaconda3-2023.09-0-Linux-ppc64le.sh   REHL 9.3 LE   On install of Anaconda -- this is how we will be using    -- we will have 3-4 users using wither gitlab runners to deploy their code or signon on directly to the lpar using ssh and run their code  so ..       -- For other WML CE users, reply "yes" to allow the installer to update the .bashrc file. In this case, if multiple users are using the same system, each user should install Anaconda individually.   -- so just not sure which way to go with this option.. can you let me know..  Thanks ------------------------------Mike BohlenIBM Power Systems Adm

Hello  Running an inferencing script on x86 with the following  stack works ok# conda list | grep torchintel-extension-for-pytorch 2.1.100                  pypi_0    pypitorch                     2.3.0.dev20231219+cpu          pypi_0    pypitorchaudio                2.2.0.dev20231219+cpu          pypi_0    pypitorchvision               0.18.0.dev20231219+cpu          pypi_0    pypiHowever on P10 with the following conda environment  conda list | grep torch _pytorch_select 1.0 cpu_2 rocketce pytorch-base 2.0.1 cpu_py310_pb4.21.12_1 rocketce pytorch-cpu 2.0.1 py310_1 rocketce torchdata 0.6.0 py310_2 rocketce torchtext-base 0.15.2 cpu_py310_1 rocketce torchtext-cpu 0.15.2 py310_1 rocketce torchvision-base 0.15.2

As part of this Rocket provides the AI/ML conda packages for Power machines.The RocketCE conda packages are build and supported for Power 9 and Power 10 with ppc64le architecture. These packages are installed into a conda environment so after installation is complete, the frameworks are ready for use. Presently these packages are available in RocketCE conda channel.This version of packages are available for Python 3.9, 3.10 and 3.11 and for CUDA 11.8 and 12.2.Note: Access and availability to this data is subject to the community and can change at any time.We are releasing RocketCE packages based on OpenCE v1.10.0.To know more about RocketCE and the various package

Overview  XGBoost is an optimized distributed gradient boosting library designed to be highly efficient, flexible, and portable. It implements machine learning algorithms under the Gradient Boosting framework.   This release of RocketCE is based on Release OpenCE v1.10.0 and it includes XGBoost 1.7.6.  These packages are supported for both Power 9 and Power 10

Overview  ONNX Runtime is a cross-platform inference and training machine-learning accelerator.  ONNX Runtime is an accelerator for machine learning models with multi-platform support and a flexible interface to integrate with hardware-specific libraries. ONNX Runtime can be used with models from PyTorch, Tensorflow/Keras, TFLite, scikit-learn, and other frameworks.  This release of RocketCE is based on Release OpenCE v1.10.0 and it includes ONNX Runtime v1.15.1.  These packages are supported for both Power 9 and 

Overview  RocketCE Pytorch includes support for IBM's Distributed Deep Learning (DDL).   This version of Pytorch will have the support of matrix-multiply assist (MMA) which was introduced in Power 10.  This release of RocketCE is based on Release OpenCE v1.10.0 and it includes Pytorch 2.0.1. CPU version optimized for Power10 pytorch-cpu 2.0.1 of Python 3.11 Cuda versions For Cuda 12.2 pytorch 2.0.1 of Python 3.11 pytorch 2.0.1 of Python 3.10 pytorch 2.0.1 of Python 3.9 Cuda 11.8 pytorch 2.0.1 of Python 3.11 These packages are supported for both 

Overview  TensorFlow is an open source library to help you develop and train machine learning models.   RocketCE includes several packages from the TensorFlow ecosystem and these packages are available in channel rocketce of anaconda.org.   First release of RocketCE is based on OpenCE Release v1.10.0. In this version of RocketCE, there is TensorFlow 2.13.0. 

RocketCE conda packages are installed in Power Systems. Following are helpful to setup the initial system requirements for RocketCE.Operating SystemRed Hat Enterprise Linux 8.4Setting up Red Hat Enterprise LinuxRHEL 8 can be installed by following Install RHEL 8Installing AnacondaAnaconda on Power machines can be installed by the following steps Download Anaconda3-2021.05-Linux-ppc64le.sh  file for machine with Power architecture Run the following command to install Anaconda bash <<path to file>>/Anaconda3-2020.02-Linux

Would like to know the availability of this Rocket AI Hub for IBM Power Support v1.7.0 packages -  If so what is the Link to Download this  Is there an Enterprise Support and a Free Version also ? ------------------------------Raghuvira ReddysMrRocket Forum Shared Account------------------------------

Will this announcement drive more clients to move to RocketCE?https://www.ibm.com/docs/en/announcements/db2-webquery-eom------------------------------David ChizekPower SpecialistRocket Forum Shared Account------------------------------

Is Open Source Kubernetes available on Power10?  ------------------------------Pradip BhattIBM Principal Sales SpecialistRocket Forum Shared Account------------------------------

Is VectorDB available on IBM Power10 server?  VectorDB Vectordb remove preview VectorDB VectorDB is a lightweight Python package for storing and retrieving text using chunking, embedding, and vector search techniques. It provides an easy-to-use interface for

Good afternoon, I've looked through a few of the posts here, but I can't quite find if the rocketce packages are intended to run with openMPI or not.  We use SpectrumMPI and I was hoping to save some time by getting them from the rocketce channel instead of building them via the open-ce-builder which is proving difficult. Jamie------------------------------Jamie FinneyRocket Forum Shared Account------------------------------

As part of this Rocket provides the AI/ML conda packages and GPU Operator for Power machines.The RocketCE conda packages are build and supported for Power 9 and Power 10 with ppc64le architecture. These packages are installed into a conda environment so after installation is complete, the frameworks are ready for use. Presently these packages are available in RocketCE conda channel.This version of packages are available for Python 3.9 and 3.10 and for CUDA 11.2 and 11.8.Note: Access and availability to this data is subject to the community and can change at any time.We are releasing RocketCE packages based on OpenCE v1.9.1.To know more about RocketCE and the variou

Overview  XGBoost is an optimized distributed gradient boosting library designed to be highly efficient, flexible, and portable. It implements machine learning algorithms under the Gradient Boosting framework.   This release of RocketCE is based on Release OpenCE v1.9.1 and it includes XGBoost 1.7.6.  These packages are supported for both Power 9 and Power 10