Skip to Content

Wiki source code of 00 - How to login to Maxwell

Last modified by flenners on 2025-06-24 16:56
Show last authors
1 The DESY has a quite powerful compute cluster called the Maxwell cluster. The documentation can be found here [[https:~~/~~/confluence.desy.de/display/MXW/Maxwell+Cluster>>doc:MXW.Maxwell Cluster.WebHome||shape="rect"]], however as this can be confusing sometimes, we will try to condensate this to a step by step manual.
2
3
4
5 {{toc/}}
6
7 = {{id name="00-HowtologintoMaxwell-GettingaDESYAccount"/}}Getting a DESY Account =
8
9 During you beamtime you will encounter multiple systems, where you will need two different types of accounts:
10
11 == {{id name="00-HowtologintoMaxwell-TheDOORAccount"/}}The DOOR Account ==
12
13 Before you arrive you have to create a DOOR account and do all the safety trainings. This account is also being used for the gamma-portal, where you can manage you beamtime data, grant access to other users and manage FTP access. However this account does not work with other resources. For this you will have to request a second account:
14
15 == {{id name="00-HowtologintoMaxwell-ThePSXAccount"/}}The PSX Account ==
16
17 If you decide during a beamtime, you want to have access to the cluster, tell your local contact so, and they will request a PSX account for you. With this you will get access to the Kerberos, Windows and afs resources at DESY, which includes the cluster.
18
19
20 After you got the account, you have to change the initial password within 6 days. For this, go to [[https:~~/~~/passwd.desy.de/>>url:https://passwd.desy.de/||shape="rect"]] and log in with your user name and initial password (you do not need any OTP when you sign in for the first time). Then agree to the terms and change your password.
21
22 = {{id name="00-HowtologintoMaxwell-UsingtheCluster"/}}Using the Cluster =
23
24 == {{id name="00-HowtologintoMaxwell-StructureoftheCluster"/}}Structure of the Cluster ==
25
26 === {{id name="00-HowtologintoMaxwell-Overview"/}}Overview ===
27
28 The Maxwell Cluster has (status 2021) more than 750 nodes in it. To organize this, you cannot access any node directly, but you have to request compute resources at first. You then can connect form an entrance node to you compute node
29
30 === {{id name="00-HowtologintoMaxwell-EntranceNodes"/}}Entrance Nodes ===
31
32 If you have successfully obtained an PSX account you can get started. The entrance node are:
33
34
35 [[https:~~/~~/max-display.desy.de:3389/auth/ssh>>url:https://max-display.desy.de:3443/auth/ssh||shape="rect"]]  (in any case)
36
37 These nodes are **not **for processing, as you will share them with many other users. So please do not do anything computational intensive on them, like reconstruction or visualization. Viewing images is ok.
38
39 === {{id name="00-HowtologintoMaxwell-FastX2"/}}Fast X3 ===
40
41 The cluster uses the software FastX3 for connection and virtual desktop. To get the right version of this, use the web interface, log in, and in the bottom right corner is a download link for the desktop client. The version has to match exactly to work properly.
42
43 If you want to add a connection in the desktop client, click the plus, select web, use the address above (including the port), and your username and force ssh authentication. Then you can choose if you want a virtual desktop (XFCE) or a terminal.
44
45 === {{id name="00-HowtologintoMaxwell-Partitions"/}}Partitions ===
46
47 Starting from an entrance node, you can connect to a compute node. As there are multiple levels of priorities etc. the nodes are organizes in partitions. You can only access some of these. To view which one, open a terminal and use the commad:
48
49 {{code}}
50 my-partitions
51 {{/code}}
52
53 Your result will look something like this:
54
55 [[image:attach:P5I.User Guide\: NanoCT.4\. Reconstruction Guide.00 - How to login to Maxwell.WebHome@image2021-5-4_10-28-14.png||queryString="version=1&modificationDate=1620116894626&api=v2"]]
56
57 == {{id name="00-HowtologintoMaxwell-SLURM"/}}SLURM ==
58
59 The access to the resources of the cluster is managed via a scheduler, SLURM.
60
61 SLURM schedules the access to nodes and can revokes access if higher priority jobs come.
62
63 === {{id name="00-HowtologintoMaxwell-PSXPartition"/}}PSX Partition ===
64
65 Here you cannot be kicked out of your allocation. However, only few nodes are in this partition and you can also only allocate few in parallel (2021: 5). Some of them have GPUs available.
66
67 === {{id name="00-HowtologintoMaxwell-AllPartition"/}}All Partition ===
68
69 Very large number of nodes available and you can allocate many in parallel (2021: 100). However each allocation can be revoked without a warning if s.o. with higher priority comes. This is very common to happen. If you want to use this partition, be sure to design your job accordingly. Only CPU nodes.
70
71 === {{id name="00-HowtologintoMaxwell-AllgpuPartition"/}}Allgpu Partition ===
72
73 Like all, but with GPUs
74
75 === {{id name="00-HowtologintoMaxwell-JhubPartition"/}}Jhub Partition ===
76
77 For Jupyter Hub
78
79
80 == {{id name="00-HowtologintoMaxwell-ConnectingtotheCluster"/}}Connecting to the Cluster ==
81
82 Connect to an entrance node via FastX. You will automatically be assigned to a node when you start a session via a load balancer (max-display001-003, max-nova001-002)
83
84 [[image:attach:P5I.User Guide\: NanoCT.4\. Reconstruction Guide.00 - How to login to Maxwell.WebHome@image2021-4-27_13-55-52.png||queryString="version=1&modificationDate=1619524552546&api=v2"]]
85
86 Choose a graphic interface and look around.
87
88
89 == {{id name="00-HowtologintoMaxwell-DataStorage"/}}Data Storage ==
90
91 The Maxwell cluster knows many storage systems. The most important are:
92
93 Your User Folder: This has a hard limit of 30 GB. Be sure not to exceed this.
94
95 The GPFS: here all the beamtime data are stored.
96
97 === {{id name="00-HowtologintoMaxwell-GPFS"/}}GPFS ===
98
99 Usually you can find you data at: /asap3/petra3/gpfs/<beamline>/<year>/data/<beamtime_id>
100
101 In there you will find a substructure:
102
103 * raw: raw measurement data. Only applicant and beamtime leader can write/delete there
104 * processed: for all processed data
105 * scratch_cc: scratch folder w/o backup
106 * shared: for everything else
107
108 The GPFS has regular snapshots. The whole capacity of this is huge (several PB)
109
110 == {{id name="00-HowtologintoMaxwell-HowtoGetaComputeNode"/}}How to Get a Compute Node ==
111
112 If you want to do some processing, there are two ways to start a job in SLURM:
113
114 1. Interactive
115 1. Batch
116
117 In both cases you are the only person working on the node, so use it as much as you like.
118
119 === {{id name="00-HowtologintoMaxwell-StartinganInteractiveJob"/}}Starting an Interactive Job ===
120
121 To get a node you have to allocate one via SLURM e.g. use:
122
123 {{code}}
124 salloc -N 1 -p psx -t 1-05:00:00
125 {{/code}}
126
127 Looking at the individual options:
128
129 * salloc: specifies you want a live allocation
130 * -N 1: for one node
131 * -p psx: on the psx partition. You can also add multiple separated with a comma: -p psx,all
132 * -t 1-05:00:00: for the duration of 1 day and 5h
133 * (((
134 Other options could be: ~-~-mem=500GB with at least 500GB of memory,
135
136 (% class="code" %)
137 (((
138 if you need gpu: (% class="bash plain" %){{code language="none"}}--constraint=P100{{/code}}
139 )))
140 )))
141 * ... see the SLURM documentation for more options
142
143 If your job is scheduled you see your assigned node and can connect via ssh to it. (in the rare case where you do not see anything use my-jobs to find out the host name).
144
145 === {{id name="00-HowtologintoMaxwell-Startingabatchjob"/}}Starting a batch job ===
146
147 For a batch job you need a small shell script describing what you want to do. You do not see the job directly, but the output is written to a log file (and results can be stored on disk)
148
149 With a batch job, you can also start an array job, where the same task is executed on multiple servers in parallel.
150
151 An example for such a script:
152
153 {{code}}
154 #!/bin/bash
155 #SBATCH --time 0-01:00:00
156 #SBATCH --nodes 1
157 #SBATCH --partition all,ps
158 #SBATCH --array 1-80
159 #SBATCH --mem 250GB
160 #SBATCH --job-name ExampleScript
161
162
163 source /etc/profile.d/modules.sh
164 echo "SLURM_JOB_ID $SLURM_JOB_ID"
165 echo "SLURM_ARRAY_JOB_ID $SLURM_ARRAY_JOB_ID"
166 echo "SLURM_ARRAY_TASK_ID $SLURM_ARRAY_TASK_ID"
167 echo "SLURM_ARRAY_TASK_COUNT $SLURM_ARRAY_TASK_COUNT"
168 echo "SLURM_ARRAY_TASK_MAX $SLURM_ARRAY_TASK_MAX"
169 echo "SLURM_ARRAY_TASK_MIN $SLURM_ARRAY_TASK_MIN"
170
171 module load maxwell gcc/8.2
172
173 .local/bin/ipython3 --pylab=qt5 PathToYourScript/Script.py $SLURM_ARRAY_TASK_ID
174
175 exit
176
177
178 {{/code}}
179
180
181 To run this use
182
183 {{code}}
184 sbatch ./your_script.sh
185 {{/code}}
186
187
188 === {{id name="00-HowtologintoMaxwell-Viewingyouallocations"/}}Viewing you allocations ===
189
190 To view your pending or running allocations you can use:
191
192 {{code}}
193 squeue -u <username>
194
195 or
196
197 my-jobs
198 {{/code}}
199
200
201 === {{id name="00-HowtologintoMaxwell-Whatisrealisticintermsofresources"/}}What is realistic in terms of resources ===
202
203 To be fair, you will not get 100 nodes every time you want them. Especially during a user run, the machines are often quite busy. But if you design your scripts to be tolerant to sudden cancellation, it is still worth trying if you profit from massive parallelization.
204
205 If you want to do some small processing, use one of the psx nodes. This should work most of the time.
206
207
208 == {{id name="00-HowtologintoMaxwell-GrantingDataAccesstootherBeamtimes"/}}Granting Data Access to other Beamtimes ==
209
210 If you have to add other users to a past beamtime, this can be done via the gamma-portal (by PI, leader or beamline scientist). After adding the accounts, these people have to make sure to log off from **all **FastX sessions, etc. to update the permissions.
211
212
213
214