You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: docs/clusters/eiger.md
+31-35Lines changed: 31 additions & 35 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -4,13 +4,11 @@
4
4
Eiger is an Alps cluster that provides compute nodes and file systems designed to meet the needs of CPU-only workloads for the [HPC Platform][ref-platform-hpcp].
5
5
6
6
!!! under-construction
7
-
This documentation is for `eiger.alps.cscs.ch` - an updated version of Eiger that will replace the existing `eiger.cscs.ch` cluster.
8
-
For help using the existing Eiger, see the [Eiger User Guide](https://confluence.cscs.ch/spaces/KB/pages/284426490/Alps+Eiger+User+Guide) on the legacy KB documentation site.
9
-
10
-
The target date for full deployment of the new Eiger is **July 1, 2025**.
7
+
This documentation is for the updated cluster `Eiger.Alps` reachable at `eiger.alps.cscs.ch`, that has replaced the former cluster as of June 30 2025.
8
+
The previous [Eiger User Guide](https://confluence.cscs.ch/spaces/KB/pages/284426490/Alps+Eiger+User+Guide) is still available on the legacy Knowledge Base.
11
9
12
10
!!! change "Important changes"
13
-
The redeployment of `eiger.cscs.ch` as `eiger.alps.cscs.ch`introduces changes that may affect some users.
11
+
The redeployment of `eiger.cscs.ch` as `eiger.alps.cscs.ch`has introduced changes that may affect some users.
14
12
15
13
### Breaking changes
16
14
@@ -31,10 +29,10 @@ Eiger is an Alps cluster that provides compute nodes and file systems designed t
31
29
32
30
### Unimplemented features
33
31
34
-
!!! under-construction "FirecREST is not yet available"
35
-
[FirecREST][ref-firecrest] has not been configured on `eiger.alps` - it is still running on the old Eiger.
32
+
!!! under-construction "Jupyter and FirecREST is not yet available"
33
+
[Jupyter and FirecREST][ref-firecrest] have not been configured on `Eiger.Alps`.
36
34
37
-
**It will be deployed, and this documentation updated when it is.**
35
+
**They will be deployed as soon as possible and this documentation will be updated accordingly**
38
36
39
37
### Minor changes
40
38
@@ -44,18 +42,16 @@ Eiger is an Alps cluster that provides compute nodes and file systems designed t
44
42
45
43
### Compute nodes
46
44
47
-
!!! under-construction
48
-
During this Early Access phase, there are 19 compute nodes for you to test and port your workflows to the new Eiger deployment. There is one compute node in the `debug` partition and one in the `xfer` partition for internal data transfer. The remaining compute nodes will be moved from `eiger.cscs.ch` to `eiger.alps.cscs.ch` at a later date (provisionally, 1 July 2025).
49
-
50
-
Eiger consists of 19 [AMD Epyc Rome][ref-alps-zen2-node] compute nodes.
51
-
52
-
There is one login node, `eiger-ln010`.
53
-
54
-
[//]: #(TODO: You will be assigned to one of the four login nodes when you ssh onto the system, from where you can edit files, compile applications and start simulation jobs.)
55
-
56
-
| node type | number of nodes | total CPU sockets | total GPUs |
Eiger consists of multicore [AMD Epyc Rome][ref-alps-zen2-node] compute nodes: please note that the total number of available compute nodes on the system might vary over time, therefore you might want to check them with the Slurm command `sinfo -s`.
46
+
```
47
+
PARTITION AVAIL TIMELIMIT NODES(A/I/O/T) NODELIST
48
+
debug up 30:00 0/12/0/12 nid[002236-002247]
49
+
xfer up 1-00:00:00 0/4/0/4 nid[002232-002235]
50
+
prepost up 30:00 0/560/0/560 nid[001000-001023,001028-001031,001064-001127,001160-001191,001256-001267,001272-001287,001320-001447,001504-001539,001541-001543,001573-001599,001640-001767,001797-001799,001829-001831,002152-002231]
51
+
normal* up 1-00:00:00 0/560/0/560 nid[001000-001023,001028-001031,001064-001127,001160-001191,001256-001267,001272-001287,001320-001447,001504-001539,001541-001543,001573-001599,001640-001767,001797-001799,001829-001831,002152-002231]
52
+
low up 1-00:00:00 0/560/0/560 nid[001000-001023,001028-001031,001064-001127,001160-001191,001256-001267,001272-001287,001320-001447,001504-001539,001541-001543,001573-001599,001640-001767,001797-001799,001829-001831,002152-002231]
53
+
```
54
+
Additionally, there are four login nodes with hostnames `eiger-ln00[1-4]`: .
59
55
60
56
### Storage and file systems
61
57
@@ -148,31 +144,33 @@ To build images, see the [guide to building container images on Alps][ref-build-
148
144
149
145
Eiger uses [Slurm][ref-slurm] as the workload manager, which is used to launch and monitor workloads on compute nodes.
150
146
151
-
There are four[Slurm partitions][ref-slurm-partitions] on the system:
147
+
There are multiple[Slurm partitions][ref-slurm-partitions] on the system:
152
148
149
+
* the `debug` partition can be used to access a small allocation for up to 30 minutes for debugging and testing purposes
150
+
* the `prepost` partition is meant for small high priority allocations up to 30 minutes, for pre- and post-processing jobs.
153
151
* the `normal` partition is for all production workloads.
154
-
* the `debug` partition can be used to access a small allocation for up to 30 minutes for debugging and testing purposes.
155
152
* the `xfer` partition is for [internal data transfer][ref-data-xfer-internal].
156
153
* the `low` partition is a low-priority partition, which may be enabled for specific projects at specific times.
157
154
158
-
| name | nodes | max nodes per job | time limit |
159
-
| -- | -- | -- | -- |
160
-
|`normal`| unlim | - | 24 hours |
161
-
|`debug`| 32 | 1 | 30 minutes |
162
-
|`xfer`| 2 | 1 | 24 hours |
163
-
|`low`| unlim | - | 24 hours |
155
+
| name | max nodes per job | time limit |
156
+
| -- || -- | -- |
157
+
|`debug`| 1 | 30 minutes |
158
+
|`prepost`| 1 | 30 minutes |
159
+
|`normal`| - | 24 hours |
160
+
|`xfer`| 1 | 24 hours |
161
+
|`low`| - | 24 hours |
164
162
165
163
* nodes in the `normal` and `debug` partitions are not shared
166
164
* nodes in the `xfer` partition can be shared
167
165
168
166
See the Slurm documentation for instructions on how to run jobs on the [AMD CPU nodes][ref-slurm-amdcpu].
169
167
170
-
### FirecREST
168
+
### Jupyter and FirecREST
171
169
172
170
!!! under-construction "FirecREST is not yet available"
173
-
[FirecREST][ref-firecrest]has not been configured on `eiger.alps` - it is still running on the old Eiger.
171
+
[Jupyter and FirecREST][ref-firecrest]have not been configured on `Eiger.Alps`.
174
172
175
-
**It will be deployed, and this documentation updated when it is.**
173
+
**They will be deployed as soon as possible and this documentation will be updated accordingly**
176
174
177
175
## Maintenance and status
178
176
@@ -184,12 +182,10 @@ Exceptional and non-disruptive updates may happen outside this time frame and wi
0 commit comments