Merge branch 'main' into jupyter-update

lukasgd · lukasgd · commit ccea2c4ee6c4 · 2025-07-07T10:25:08.000+02:00
diff --git a/docs/access/jupyterlab.md b/docs/access/jupyterlab.md
@@ -3,9 +3,9 @@
 
 ## Access and setup
 
-The JupyterHub service enables the interactive execution of JupyterLab on the compute nodes of [Daint][ref-cluster-daint], [Clariden][ref-cluster-clariden] and [Santis][ref-cluster-santis]. 
+The JupyterHub service enables the interactive execution of JupyterLab on the compute nodes of [Daint][ref-cluster-daint], [Clariden][ref-cluster-clariden], [Santis][ref-cluster-santis] and [Eiger][ref-cluster-eiger].
 
-The service is accessed at [jupyter-daint.cscs.ch](https://jupyter-daint.cscs.ch/), [jupyter-clariden.cscs.ch](https://jupyter-clariden.cscs.ch/) and [jupyter-santis.cscs.ch](https://jupyter-clariden.cscs.ch/), respectively. As the notebook servers are executed on compute nodes, you must have a project with compute resources available on the respective cluster. 
+The service is accessed at [jupyter-daint.cscs.ch](https://jupyter-daint.cscs.ch/), [jupyter-clariden.cscs.ch](https://jupyter-clariden.cscs.ch/), [jupyter-santis.cscs.ch](https://jupyter-clariden.cscs.ch/) and [jupyter-eiger.cscs.ch](https://jupyter-eiger.cscs.ch), respectively. As the notebook servers are executed on compute nodes, you must have a project with compute resources available on the respective cluster.
 
 Once logged in, you will be redirected to the JupyterHub Spawner Options form, where typical job configuration options can be selected. These options might include the type and number of compute nodes, the wall time limit, and your project account.
 
diff --git a/docs/alps/hardware.md b/docs/alps/hardware.md
@@ -78,6 +78,17 @@ Each node contains four Grace-Hopper modules and four corresponding network inte
     x1100c0s6b1n0
     ```
 
+??? info "Core-to-core latency"
+    The core-to-core latency on a Grace CPU (collected using the [`core-to-core-latency`](https://github.com/nviennot/core-to-core-latency) program):
+
+    ![](../images/alps/grace-core-to-core-latency.svg)
+
+    The latencies between the first cores on each of the four Grace CPUs within a node:
+
+    ![](../images/alps/grace-core-to-core-latency-cross-socket.svg)
+
+    Note the significantly higher latencies compared to within a single Grace CPU.
+
 [](){#ref-alps-zen2-node}
 ### AMD Rome CPU Nodes
 
diff --git a/docs/clusters/daint.md b/docs/clusters/daint.md
@@ -127,7 +127,7 @@ See the Slurm documentation for instructions on how to run jobs on the [Grace-Ho
 
 ### FirecREST
 
-Daint can also be accessed using [FirecREST][ref-firecrest] at the `https://api.cscs.ch/ml/firecrest/v2` API endpoint.
+Daint can also be accessed using [FirecREST][ref-firecrest] at the `https://api.cscs.ch/hpc/firecrest/v2` API endpoint.
 
 !!! warning "The FirecREST v1 API is still available, but deprecated"
 
diff --git a/docs/clusters/eiger.md b/docs/clusters/eiger.md
@@ -158,12 +158,15 @@ There are multiple [Slurm partitions][ref-slurm-partitions] on the system:
 
 See the Slurm documentation for instructions on how to run jobs on the [AMD CPU nodes][ref-slurm-amdcpu].
 
-### Jupyter and FirecREST
+### JupyterHub 
 
-!!! under-construction "Jupyter is not yet available"
-    [Jupyter][ref-jupyter] has not yet been configured on `Eiger.Alps`.
+A [JupyterHub][ref-jupyter] service for Eiger is available at [https://jupyter-eiger.cscs.ch](https://jupyter-eiger.cscs.ch).
 
-    **It will be deployed as soon as possible and this documentation will be updated accordingly**
+### FirecREST
+
+Eiger can also be accessed using [FirecREST][ref-firecrest] at the `https://api.cscs.ch/hpc/firecrest/v2` API endpoint.
+
+!!! warning "The FirecREST v1 API is still available, but deprecated"
 
 ## Maintenance and status
 
diff --git a/docs/images/alps/grace-core-to-core-latency-cross-socket.svg b/docs/images/alps/grace-core-to-core-latency-cross-socket.svg
diff --git a/docs/images/alps/grace-core-to-core-latency.svg b/docs/images/alps/grace-core-to-core-latency.svg
diff --git a/docs/running/slurm.md b/docs/running/slurm.md
@@ -300,8 +300,10 @@ if [[ $SLURM_LOCALID -eq 0 ]]; then
     CUDA_VISIBLE_DEVICES=0,1,2,3 nvidia-cuda-mps-control -d
 fi
 
-# Set CUDA device
-numa_nodes=$(hwloc-calc --physical --intersect NUMAnode $(hwloc-bind --get --taskset))
+# Set CUDA device. Disable HWLOC_KEEP_NVIDIA_GPU_NUMA_NODES to avoid GPU NUMA
+# nodes appearing in the list of CUDA devices. They start appearing in hwloc
+# version 2.11.
+numa_nodes=$(HWLOC_KEEP_NVIDIA_GPU_NUMA_NODES=0 hwloc-calc --physical --intersect NUMAnode $(hwloc-bind --get --taskset))
 export CUDA_VISIBLE_DEVICES=$numa_nodes
 
 # Wait for MPS to start
diff --git a/docs/software/uenv/index.md b/docs/software/uenv/index.md
@@ -52,16 +52,11 @@ $ uenv --version
 
 On Alps clusters the most recent version 8.1.0 is installed.
 
-??? warning "Out of date uenv version on Eiger and Balfrin"
+??? warning "Out of date uenv version on Balfrin"
 
-    The uenv tool available on Eiger and Balfrin is a different version than the one described below, and some commands will be different to those documented here.
+    The uenv tool available on Balfrin is a different version than the one described below, and some commands will be different to those documented here.
 
-    !!! note
-        This note only applies to the current `eigen.cscs.ch` deployment.
-
-        The new [`eiger.alps.cscs.ch`][ref-cluster-eiger] deployment has version 8.1.0 of uenv installed.
-
-    Please refer to `uenv --help` for the correct usage on these systems.
+    Please refer to `uenv --help` for the correct usage on Balfrin.
 
 ## Finding uenv
 
@@ -121,8 +116,8 @@ The search can be refined by providing a [label][ref-uenv-labels].
 
 ## Downloading uenv
 
-??? note "Using uenv for the first time on Balfrin and Eiger"
-    With the old version of uenv installed on Balfrin and Eiger, before downloading your first image, a local directory for storing the images must first be created, otherwise you will receive an error message that the repository does not exist.
+??? warning "Using uenv for the first time on Balfrin"
+    With the old version of uenv installed on Balfrin, before downloading your first image, a local directory for storing the images must first be created, otherwise you will receive an error message that the repository does not exist.
 
     To create a repo in the default location, use the following command:
 
