diff --git a/docs/access/index.md b/docs/access/index.md index 339763b69..666b6500d 100644 --- a/docs/access/index.md +++ b/docs/access/index.md @@ -32,7 +32,6 @@ Remote SSH access to Confidential Data Workspace VMs is **disabled** for securit Includes access to the following services: -* [Ultra2](../services/ultra2/index.md) * [EIDF Interim and Experimental Systems](../services/experimental/index.md) To login to most command-line services with `ssh` you should use the username and password diff --git a/docs/services/index.md b/docs/services/index.md index 2a09e1039..02186cb12 100644 --- a/docs/services/index.md +++ b/docs/services/index.md @@ -8,8 +8,6 @@ [Notebooks](./jhub/index.md) -[Ultra2](./ultra2/index.md) - [GitLab Code Storage](./gitlab/index.md) ## Data Services diff --git a/docs/services/ultra2/access.md b/docs/services/ultra2/access.md deleted file mode 100644 index 690886966..000000000 --- a/docs/services/ultra2/access.md +++ /dev/null @@ -1,11 +0,0 @@ -# Overview - -Ultra2 is a single logical CPU system based at EPCC. It is suitable for running jobs which require large volumes of non-distributed memory (as opposed to a cluster). - -## Specifications - -The system is a HPE SuperDome Flex containing 576 individual cores in a SMT-1 arrangement (1 thread per core). The system has 18TB of memory available to users. Home directories are network mounted from the EIDF e1000 Lustre filesystem, although some local NVMe storage is available for temporary file storage during runs. - -## Getting Access - -Access to Ultra2 is currently by arrangement with EIDF. Please apply for a project on the [EIDF Portal](https://portal.eidf.ac.uk/). diff --git a/docs/services/ultra2/connect.md b/docs/services/ultra2/connect.md deleted file mode 100644 index 90791edc7..000000000 --- a/docs/services/ultra2/connect.md +++ /dev/null @@ -1,20 +0,0 @@ -# Login - -The hostname for SSH access to the system is `ultra2.eidf.ac.uk` - -## Access credentials - -To access Ultra2, you need to use two credentials: your SSH key pair protected by a passphrase **and** a Time-based one-time password (TOTP). - -### SSH Key - -You must upload the public part of your SSH key pair to the SAFE by following the [instructions from the SAFE documentation](https://epcced.github.io/safe-docs/safe-for-users/#how-to-add-an-ssh-public-key-to-your-account) - -### Time-based one-time password (TOTP) - -You must set up your TOTP token by following the [instructions from the SAFE documentation](https://epcced.github.io/safe-docs/safe-for-users/#how-to-turn-on-mfa-on-your-machine-account) - -### SSH Login example - -To login to Ultra2, you will need to use the SSH Key and TOTP token as noted above. -With the appropriate key loaded
`ssh @ultra2.eidf.ac.uk` will then prompt you, roughly once per day, for your TOTP code. diff --git a/docs/services/ultra2/index.md b/docs/services/ultra2/index.md deleted file mode 100644 index 6c1e848d6..000000000 --- a/docs/services/ultra2/index.md +++ /dev/null @@ -1,7 +0,0 @@ -# Ultra2 Large Memory System - -[Overview](./access.md) - -[Connect](./connect.md) - -[Running jobs](./run.md) diff --git a/docs/services/ultra2/run.md b/docs/services/ultra2/run.md deleted file mode 100644 index b93370931..000000000 --- a/docs/services/ultra2/run.md +++ /dev/null @@ -1,50 +0,0 @@ -# Running jobs - -## Software - -### OneAPI - -The primary HPC software provided is Intel's OneAPI suite containing mpi compilers and runtimes, debuggers and the vTune performance analyser. Standard GNU compilers are also available. -The OneAPI suite can be loaded by sourcing the shell script: - -```bash -source /opt/intel/oneapi/setvars.sh -``` - -## Queue system - -All jobs must be run via SLURM to avoid inconveniencing other users of the system. Users should not run jobs directly. Note that the system has one logical processor with a large number of threads and thus appears to SLURM as a single node. This is intentional. - -### Queue limits - -We kindly request that users limit their maximum total running job size to 288 cores and 4TB of memory, whether that be a divided into a single job, or a number of jobs. -This may be enforced via SLURM in the future. - -### Example MPI job - -An example script to run a multi-process MPI "Hello world" example is shown. - -```bash -#!/usr/bin/env bash -#SBATCH -J HelloWorld -#SBATCH --nodes=1 -#SBATCH --tasks-per-node=4 -#SBATCH --nodelist=sdf-cs1 -#SBATCH --partition=standard -##SBATCH --exclusive - - -echo "Running on host ${HOSTNAME}" -echo "Using ${SLURM_NTASKS_PER_NODE} tasks per node" -echo "Using ${SLURM_CPUS_PER_TASK} cpus per task" -let mpi_threads=${SLURM_NTASKS_PER_NODE}*${SLURM_CPUS_PER_TASK} -echo "Using ${mpi_threads} MPI threads" - -# Source oneAPI to ensure mpirun available -if [[ -z "${SETVARS_COMPLETED}" ]]; then -source /opt/intel/oneapi/setvars.sh -fi - -# mpirun invocation for Intel suite. -mpirun -n ${mpi_threads} ./helloworld.exe -``` diff --git a/docs/status/index.md b/docs/status/index.md index 306efae62..c2c17fe19 100644 --- a/docs/status/index.md +++ b/docs/status/index.md @@ -12,7 +12,6 @@ The table below represents the broad status of each EIDF service. | VM SSH Gateway | ![Custom badge](https://img.shields.io/endpoint?style=plastic&url=https://epcced.github.io/eidf-status/Data/sshgateway.json) | | VM VDI Gateway | ![Custom badge](https://img.shields.io/endpoint?style=plastic&url=https://epcced.github.io/eidf-status/Data/vdigateway.json) | | Virtual Desktops | ![Custom badge](https://img.shields.io/endpoint?style=plastic&url=https://epcced.github.io/eidf-status/Data/virtualmachines.json) | -| Ultra2 | ![Custom badge](https://img.shields.io/endpoint?style=plastic&url=https://epcced.github.io/eidf-status/Data/superdome.json) | ## EIDF Regular Maintenance Sessions diff --git a/docs/storage/faq.md b/docs/storage/faq.md index f91f26ea5..6f5a8f848 100644 --- a/docs/storage/faq.md +++ b/docs/storage/faq.md @@ -2,11 +2,7 @@ ## I have deleted my important file, what can I do? -The EIDF does not offer a backup service and only replicates Ultra2-local storage. - -Whilst EIDF Ultra2 storage is replicated to tape in case of system failure or data loss related to the running of the service, we do **not** offer or plan to offer the ability to restore files from this medium in the case of user error. - -Users are responsible for backing up their data against user error. +The EIDF does not offer a backup service. Users are responsible for backing up their data against user error. ## I want to access storage from across all EIDF services diff --git a/docs/storage/overview.md b/docs/storage/overview.md index dfc2663e7..08fe3819d 100644 --- a/docs/storage/overview.md +++ b/docs/storage/overview.md @@ -4,7 +4,6 @@ The EIDF offers the following storage options: - [Virtual Machine (VM) local storage](#virtual-machine-vm-local) - [GPU local storage](#gpu-service-local) -- [Ultra2 local storage](#ultra2-local) - [S3](#s3) - [Shared Filesystem (CephFS)](#shared-filesystem-cephfs) - [Data Publishing Service](#data-publishing-service) @@ -13,13 +12,12 @@ These storage options are backed by the hardware described on this [page](https: ## Summary of Storage Options and Access -| Service | VM Local | GPU Local | Ultra2 Local | S3 | Shared Filesystem (CephFS) | Data Publishing Service | -| :---------------------: | :------: | :-------: | :----------: | :-: | :-------------------: | :---------------------: | -| VMs | Yes | No | No | Yes | Yes | Yes | -| GPU Service | No | Yes | No | Yes | Yes | Yes | -| Ultra2 | No | No | Yes | Yes | No | Yes | -| Cerebras | No | No | No | No | Yes | No | -| Notebooks | No | No | No | Yes | Yes | Yes | +| Service | VM Local | GPU Local | S3 | Shared Filesystem (CephFS) | Data Publishing Service | +| :---------------------: | :------: | :-------: | :-: | :-------------------: | :---------------------: | +| VMs | Yes | No | Yes | Yes | Yes | +| GPU Service | No | Yes | Yes | Yes | Yes | +| Cerebras | No | No | No | Yes | No | +| Notebooks | No | No | Yes | Yes | Yes | ## Persistent Volumes attached to specific services @@ -40,12 +38,6 @@ Currently GPU Service persistent volumes are included in the GPU Service costs. See more details about the [GPU service](../services/gpuservice/index.md). -### Ultra2 local - -Ultra2 uses our e1000 Lustre parallel filesystem; this storage is not accessible from anywhere else on the EIDF. - -See more general details about [Ultra2](../services/ultra2/access.md). - ## Storage available for use on multiple EIDF Services ### S3 @@ -78,7 +70,4 @@ See more information about the [Data Publishing service](../services/datapublish ## Disaster Recovery -The Ultra2-local storage is replicated onto tape for recovery from disaster. Disaster recovery is triggered by disk hardware failure, not user error. Restoring data due to user actions, such as accidental file deletion, is **not** supported or planned as a service. - -There is no user interface for the Disaster Recovery and any action will be taken by the EIDF team in response to issues. If you suspect disk failures on Ultra2, please raise a ticket through the EIDF Portal facility. -The EIDF Disaster Recovery is implemented through the [HPE Data Management Framework](https://www.hpe.com/us/en/collaterals/collateral.a00022795enw.html). +There is no Disaster Recovery available for EIDF services. You are strongly encouraged to keep sufficient copies of data elsewhere. diff --git a/mkdocs.yml b/mkdocs.yml index 081816b5e..1825f0f87 100644 --- a/mkdocs.yml +++ b/mkdocs.yml @@ -68,10 +68,6 @@ nav: - "Policies": services/confidentialdataworkspace/policies.md - "Storage": services/confidentialdataworkspace/storage.md - "FAQs": services/confidentialdataworkspace/faq.md - - "Ultra2": - - "Overview": services/ultra2/access.md - - "Connect": services/ultra2/connect.md - - "Running codes": services/ultra2/run.md - "Cerebras": - "Overview": services/cerebras/access.md - "Connect": services/cerebras/connect.md