All public logs
Jump to navigation
Jump to search
Combined display of all available logs of Define Wiki. You can narrow down the view by selecting a log type, the username (case-sensitive), or the affected page (also case-sensitive).
(newest | oldest) View (newer 50 | older 50) (20 | 50 | 100 | 250 | 500)- 16:05, 16 September 2024 David talk contribs created page Dockerfiles for setting up pytorch for AMD RoCM (Created page with "== ROCM install script == <pre> #!/bin/bash # ROCm 6.1.3 Setup for Ubuntu 20.04.6 and Ubuntu 22.04.4 DT and Server build # ===================================================...")
- 15:48, 3 September 2024 David talk contribs created page Benchmarking: HS23 HEPSPEC 23 (Created page with "<pre> # on rhel 94 # following https://it-procurement-tenders.docs.cern.ch/tenders/it-5008/rhel-install/ subscription-manager repos --enable codeready-builder-for-rhel-9-$(ar...")
- 08:34, 24 June 2024 David talk contribs created page OSU Benchmarking with Python and Pytorch (Created page with "Note; this is only for CPU benchmarks but still tests the python libs <pre> python3.9 -m venv ~/venvs/pytorch-osu-testing source ~/venvs/pytorch-osu-testing/bin/activate pip...")
- 14:40, 20 June 2024 David talk contribs created page Short MPI Tests using Pytorch example code (Created page with "<pre> [antony@gpu5 pytorch-nccl-test]$ cat run_dp_mpi.sh #!/bin/bash # mpirun --mca pml ucx -np 8 -H gpu0:4,gpu5:4 ./run_dp_mpi.sh source ~/dp-build/torch-rocm61/bin/activat...")
- 10:46, 23 April 2024 David talk contribs created page Broadcom install for RoCE on ROCm devices (Created page with "<pre> ❯ cat broadcom-roce-setup.txt # ubuntu 22.04 apt install ibverbs-utils libibverbs1 rdmacm-utils librdmacm1 librdmacm-dev infiniband-diags perftest hwloc ibutils numa...")
- 13:22, 12 April 2024 David talk contribs created page OpenMPI with ROCm Support (Created page with "<pre> ❯ cat rocm-mpi-install.txt # ubuntu 22.04 LTS apt update apt install git vim libtool flex apt install build-essential # make sure user ubuntu is part of render group...")
- 22:45, 3 April 2024 David talk contribs created page ROCm Installation and Setup, tooling and info (Created page with "== Show theoretical bandwidth between gpus == <pre> root@gpu1:~# rocm-smi --shownodesbw ============================ ROCm System Management Interface =======================...")
- 21:50, 3 April 2024 David talk contribs created page CoralGemm - Matrix multiply stress test for AMD ROCm (Created page with "<pre> # rocm 6.0.2, mi210, ubuntu 2204 git clone https://github.com/AMD-HPC/CoralGemm.git cd CoralGemm/ cd src/ make # execute # dgemm mi210 64GB device ./gemm R_64F...")
- 21:04, 3 April 2024 David talk contribs created page Transferbench - PCIe and XGMI Bandwidth for AMD ROCm (Created page with "<pre> # rocm 6.0.2 mi210 on ubuntu 2204 sudo apt install libnuma-dev git clone https://github.com/ROCm/TransferBench.git cd TransferBench make # Execution for performance :...")
- 20:48, 3 April 2024 David talk contribs created page BabelStream GPU Memory Bandwidth - AMD ROCm (Created page with "<pre> # rocm 6.0.2 on ubuntu 22.04 apt install build-essential cmake libstdc++-12-dev git clone https://github.com/UoB-HPC/BabelStream.git cd BabelStream cmake -Bbuild -H. -D...")
- 20:01, 3 April 2024 David talk contribs created page AMD: AMD Landing Page (Created page with "== Benchmarking GPUs == * BabelStream GPU Memory Bandwidth - AMD ROCm")
- 17:02, 14 March 2024 User account Paddy talk contribs was created by David talk contribs
- 17:05, 30 December 2023 David talk contribs created page OpenFOAM 11 on Rocky 9 (Created page with "Headnode Rocky 9 and OpenHPC 3 - with development tools and standard openhpc packages already installed == Pull down and extract OpenFOAM == <pre> wget -O - http://dl.openfo...")
- 16:04, 18 December 2023 David talk contribs created page Installation notes on OpenHPC 3 and Rocky 9 (Created page with "== Headnode Installation == * Rocky 9.2 (this guide based on a VM so no OOB), 2 ports, port security disabled on internal port * Install guide from https://github.com/openhpc...")
- 15:21, 23 November 2023 David talk contribs created page Install JedAI - Ubuntu 22.04 with MaaS (Created page with "* Current setup - one deploy node with 2204 pre-installed, access to the ipmi on 3 other systems. == Setup Networking on the Deploy Node == <pre> # Assuming the following n...")
- 23:09, 7 November 2023 David talk contribs created page OpenOnDemand Demo using Jupyter and Machine Learning with Python and venv (Created page with "# example for jupyter demos git clone https://github.com/tirthajyoti/Machine-Learning-with-Python.git python3 -m venv ml-venv source ml-venv/bin/activate pip install -U pip...")
- 00:24, 7 November 2023 David talk contribs created page OpenOnDemand RStudio on OOD 3.x (Created page with "This was on Rocky 8.8 <pre> mkdir /var/www/ood/apps/sys/ cd /var/www/ood/apps/sys git clone https://github.com/OSC/bc_example_rstudio.git mv bc_example_rstudio rstudio cd rs...")
- 00:20, 7 November 2023 David talk contribs created page OpenOnDemand Landing Page (Created page with "* Open OnDemand Portal and Visualization *OpenOnDemand RStudio on OOD 3.x")
- 10:48, 15 October 2023 David talk contribs created page Finding the volumes clients with high IOPS consumption - who is the little hog (Created page with "== Check the perf stats from rbd == <pre> [root@deploy ~]# rbd perf image iostat rbd: waiting for initial image stats NAME...")
- 22:58, 6 October 2023 David talk contribs created page Upgrading from Centos 8 Stream to Centos 9 Stream to Rocky 9 (Created page with "What a lovely setup - we've ceph running via podman and docker running the openstack containers... what could go wrong.. == Docker and Podman == If pod man and docker are in...")
- 12:56, 4 October 2023 David talk contribs created page Benchmarking: CERN Systems for Amari IT-4776 (Created page with "<pre> ❯ cat ~/notes/cern-amari-testing.txt NodeA – 172.16.40.200 | WFTYMOKZQQ | ssh root@172.16.40.205 NodeB – 172.16.40.201 | EHIYEYXZDY | ssh root@172.16.40.207 NodeC...")
- 20:23, 24 September 2023 David talk contribs created page Starting / Stopping Ceph at Redmond (Created page with "<pre> root@str-237:~# systemctl start ceph.target root@str-238:~# systemctl start ceph.target WAIT at LEAST 60S root@str-238:~# ceph -s cluster: id: 75b74760-4...")
- 22:40, 23 August 2023 David talk contribs created page IPA FreeIPA Authentication (Created page with "Notes from CIX system == Reset a user password == <pre> nuig-crt 23:29:56 [root@head ~]# ipa user-mod mwalsh --random ipa: ERROR: Ticket expired nuig-crt 23:30:15 [root@head...")
- 13:06, 18 May 2023 David talk contribs created page Kolla: Upgrade from Wallaby to Xena on CentOS 8 Rocky 8 (Created page with "<pre> ### ubuntu xena notes sudo apt update sudo apt install git python3-dev libffi-dev gcc libssl-dev sudo apt install docker.io sudo systemctl enable docker sudo systemctl s...")
- 09:53, 16 May 2023 David talk contribs created page Ceph: Commands and Cheatsheet (Created page with "== Working with Ceph pools == Check the pools <pre> [root@deploy-ext kolla]# ceph osd pool ls device_health_metrics images volumes backups manila_data manila_metadata .rgw.ro...")
- 09:36, 19 April 2023 David talk contribs created page Creating an upload service using FastAPI - cern automation (Created page with "== Setup VM prereqs == Ubuntu 2204 VM starting point <pre> sudo apt update sudo apt install python3-pip sudo apt install python3-venv python3 -m venv fastapi-venv source fa...")
- 15:36, 30 March 2023 David talk contribs created page Adding Storpool Storage to OpenStack - Zed release (Created page with "== Get the Storpool containers == <pre> # storpool images downloaded define@kolla-deploy:~$ docker images | grep storpool cts.storpool.com/kolla/ubuntu-binary-nova-compute...")
- 16:41, 13 March 2023 David talk contribs created page NCCL Testing NVLINK NVIDIA Collectives Library (Created page with "== Install NCCL == <pre> git clone https://github.com/NVIDIA/nccl.git cd nccl make -j 64 make install </pre> == Install NCCL Tests == <pre> git clone https://github.com/NVI...")
- 15:42, 9 March 2023 David talk contribs created page Using Wazuh for SIEM and vulnerability assessment (Created page with " == Install Wazuh == <pre> curl -sO https://packages.wazuh.com/4.3/wazuh-install.sh && sudo bash ./wazuh-install.sh -a </pre> == Install Agents == * Log into portal * Add...")
- 18:41, 8 March 2023 David talk contribs created page Using metasploit for pen testing - on ubuntu 22.04 vm (Created page with "== Install == * From: https://docs.metasploit.com/docs/using-metasploit/getting-started/nightly-installers.html <pre> curl https://raw.githubusercontent.com/rapid7/metasploit...")
- 23:01, 21 February 2023 David talk contribs created page Setup Volcano and run some basic jobs (Created page with "== Install Volcano == <pre> kubectl create namespace volcano-system kubectl apply -f https://raw.githubusercontent.com/volcano-sh/volcano/master/installer/volcano-development...")
- 22:58, 21 February 2023 David talk contribs created page Add the GPU operator to a rancher deployed k8s environment (Created page with "== Setup helm == <pre> # add the gpu-operator (Mac osx) brew install helm </pre> == Add the Nvidia repo == <pre> helm repo add nvidia https://nvidia.github.io/gpu-operator h...")
- 18:02, 21 February 2023 David talk contribs created page Set up your k8s cluster to use cinder persistent volumes PV / PVC (Created page with "* At this stage we assume the rancher UI is up and running * There's a Kubernetes environment spun up and working and the correct $KUBECONFIG is setup")
- 17:45, 21 February 2023 David talk contribs created page Setting up rancher to work with OpenStack (Created page with "* Assumes the rancher UI is up and running - if not check here https://wiki.define-technology.com/mediawiki-1.35.0/index.php/Setting_up_the_rancher_UI == Enable OpenStack Nod...")
- 14:35, 17 February 2023 David talk contribs created page Setting up the rancher UI (Created page with "== Setting up the rancher UI == * Working with ubuntu 20.04 VM * m1.medium (but check rancher min requirements - I just went middle of the road) * verified on ubuntu 22.04 VM...")
- 14:24, 17 February 2023 David talk contribs created page Rancher Kubernetes on OpenStack (Created page with "== Rancher setup and basics == * Setting up the rancher UI * Setting up rancher to work with OpenStack")
- 23:28, 11 July 2022 David talk contribs created page Parsec on m60 remote cloud gaming (Created page with "this little nugget cost me a lot of time! * install the Tesla standard driver (no grid needed) * this is based on m60 in pci passthrough * setup the server on the windows vm...")
- 07:47, 17 April 2022 David talk contribs created page OpenStack:CLI Cheet Sheet (Created page with "== CLI Commands == From: https://docs.openstack.org/nova/latest/admin/services.html <pre> # list hypervisor details openstack hypervisor list --long # list VMs with availa...")
- 23:32, 15 March 2022 David talk contribs created page Setup DCGM GPU Telemetry on Kubernetes - single node centos 8 (Created page with "# install base is centos stream 8 node # setenforce 0 # make sure hosts file is setup to resolve locally [root@pcie-a100-server ~]# ping `hostname` PING pcie-a100-server.ec...")
- 10:10, 23 November 2021 David talk contribs created page Lightbits - cheatsheet and using lbcli tool (Created page with "== Get Cluster Status == <pre> [root@stu-prr-nmv1-mlnx ~]# lbcli get cluster UUID Subsystem NQN...")
- 13:00, 21 October 2021 David talk contribs created page Working with CRUSH for SSD and HDD pools (Created page with "<bash> == Crush Rules == In this scenario we have a few HCI nodes running the standard volumes for OpenStack. There's two additional servers, 1x HDD and 1x SSD server. We are...")
- 13:49, 28 September 2021 David talk contribs created page Removing the NVIDIA kernel module in Ubuntu 20.04 rmmod nvidia (Created page with "<pre> # uninstall the driver apt-get remove nvidia-driver-450-server lsof /dev/nvidia* systemctl stop nvidia-persistenced.service systemctl set-default multi-user.target roo...")
- 21:46, 22 September 2021 David talk contribs created page Working with pfsense (Created page with "== pfsense cli via ssh == <pre> vi /cf/conf/config.xml edit the config, rm /tmp/config.cache, then run /etc/rc.filter_configure </pre>")
- 15:15, 23 July 2021 David talk contribs created page Virtual Network Port Magic - openstack allow all-in-one to use standard network to access internet (Created page with "<pre> [david@Davids-MacBook-Pro-4 notes]$ cat openstack-MAGIC-port-forward-proxy-vm-with-no-internet.txt # had a hypervisor with only 1 link out to the internet - the neutron...")
- 16:27, 7 July 2021 David talk contribs created page Rebuild cinder with lightbits integration - train notes (Created page with "== Dump of all notes == <pre> [david@Davids-MacBook-Pro-4 notes]$ cat lightbits-notes.txt # config file [root@openstack _data]# cat ~/kolla/config/cinder/cinder-volume.conf...")
- 11:57, 25 June 2021 David talk contribs created page Build kolla Victoria containers on fresh ubuntu 20.04 build VM (Created page with "* VM with ubuntu 20.04 cloud image <pre> ## ubuntu victoria notes apt-get update apt-get install python3-dev libffi-dev gcc libssl-dev apt install docker.io systemctl enable...")
- 21:54, 21 June 2021 David talk contribs created page Force removal of a cinder volume from the DB (Created page with "== Force removal of a cinder Volume from the DB == Cinder Volume deletion is possible using commands, in case volume is not able to delete even force remove we have to act fr...")
- 11:29, 10 June 2021 David talk contribs created page Rebuild the DT / custom horizon image - train notes (Created page with "== Rebuild horizon with template override == <pre> # note: you can check how the standard container is build and see what sections are avaialble to inject in the template-ov...")
- 22:37, 8 June 2021 David talk contribs created page Allow instances access provider networking directly using RBAC (Created page with "<pre> root@kolla-deploy:/kolla# openstack network list +--------------------------------------+---------------------+--------------------------------------+ | ID...")
- 09:40, 28 May 2021 David talk contribs created page Setting up the registry build environment with Docker (Created page with "== Setup the initial environment == We're using a centos 7 minimal VM (Started with Stein builds and moved to Train) <pre> setenforce 0 sed -i 's/SELINUX=enforcing/SELINUX=d...")