Page History

Policies

LCLS users are responsible for complying with the data management and curation policies of their home institutions and funding agents and authorities. To enhance the scientific productivity of the LCLS user community, LCLS supplies on-site disk, tape and compute resources for prompt analysis of LCLS data, and software to access those resources consistent with the published data retention policy. Compute resources are preferentially allocated to recent and running experiments.

Data Management

Please be aware that the system administration team reserves the right for computer system maintenance on the 1st Wednesday of each month. Computers and storage systems might experience short and announced outages during these days.

Getting Access to the System

You can get into the LCLS photon computing system in the NEH by ssh'ing to one of these nodes:

psexport.slac.stanford.edu
psimport.slac.stanford.edu

From these nodes you can move data files in and out of the system and you can connect to the bastion hosts:

pslogin
psdev

Note that, from within SLAC, you can directly connect to the bastion hosts without going through psimport/psexport.

The SLAC wireless visitor network is not considered part of SLAC so you'll need to go through psexport/psimport when using your laptop on-site.

From the bastion hosts you can then reach the analysis nodes (see below).

Each control room has a number of nodes for local login. These nodes have access to the Internet and are reachable from pslogin and psdev.

The controls and DAQ nodes used for operating an instrument work in kiosk mode so you don't need a personal account to run an experiment from the control room. Remore access to these nodes is not allowed for normal users.

You will need a valid SLAC UNIX account in order to run your analysis in the NEH system. This UNIX account must be enabled in the NEH system in order to grant access to data and elog. The instructions for getting a SLAC UNIX account are here:

http://www-ssrl.slac.stanford.edu/lcls/users/logistics.html#compaccts

Getting Access to the Electronic Logbook

The elog is accessible at the following location:

https://pswww.slac.stanford.edu/apps/logbook

Each user is allowed to view and edit only the experiments she/he belongs to. The elog can also be accessed through the experiment shared account. The PI for the experiment is the custodian of the password of the experiment shared account and he/she can share it with the members of the experiment group. The name of the shared account is the same as the experiment's name.

Running the Analysis

Instrument Dedicated Nodes

These nodes are reserved for the users who are currently running an experiment. Each instrument has three dedicated interactive compute systems:

• AMO:

  psanaamo01
psanaamo02
psanaamo03

• SXR:

  psanasxr01
psanasxr02
psanasxr03

• XPP:

  psanaxpp01
psanaxpp02
psanaxpp03

• XCS:

  psanaxcs01
psanaxcs02
psanaxcs03

• CXI:

  psanacxi01
psanacxi02
psanacxi03

• MEC:

  psanamec01
psanamec02
psanamec03

The general specifications for these nodes are:

• psana<instr>01: 8 cores, Xeon E5520, 24GB, 500GB disk, 1Gb/s, dedicated Matlab license
• psana<instr>02: 8 cores, Xeon E5520, 24GB, 500GB disk, 1Gb/s
• psana<instr>03: 8 cores, Opteron 2384, 8GB, diskless, 10Gb/s

Interactive Farm

In order to get access to the interactive farm, ssh to the address psana. A load-balancing mechanism will connect you to the least loaded of the nodes in the farm. This farm is currently made of six 8-cores Opteron 2384 nodes with one 10Gb/s connection to the data.

Batch Farm

Login first to psdev or pslogin (from SLAC) or psimport or psexport (from anywhere). From there you can submit a job with the following command:

bsub -q lclsq -o <output file name> <job_script_command>

For example:

bsub -q lclsq -o ~/output/job.out my_program

This will submit a job (my_program) to the queue lclsq and write it's output to a file named ~/output/job.out.

You may check on the status of your jobs using the bjobs command.

The batch farm is made of sixty 8-cores Xeon E5520 with 1Gb/s connection to the data.

For a more detailed description and more more LSF commands, please see
http://www.slac.stanford.edu/comp/unix/unix-hpc.html

...

LCLS provides space for all your experiment's data at no cost for you. This includes the measurements raw data from the detectors as well as the data derived data from your analysis software. Your raw data are available as XTC files or, on demand, as HDF5 files. The tools for managing files are described here.

Short-term Storage

All your data is available on disk for one year after data taking. The path name is /reg/d/psdm. The data files are currently stored in a Lustre file system. Each experiment is allocated three directories: xtc, scratch and hdf5. The xtc directory contains the raw data from the DAQ system. Its contents are archived to tape. The scratch and hdf5 directories are not backed up. Please write the output of your analysis to the scratch area and not in your NFS space. Keep your analysis code under your NFS home or under your NFS group space (if you have one). Your NFS space is backep up.

Long-term Storage

After one year, your data files are removed from disk. The XTC files remain stored on tape for up to 10 years. LCLS may restore your data from tape back to disk for you to access. Restoring the data to disk more than once will require the approval of the LCLS management.

Data Export

There is a web interface to the experimental data accessible via https://pswww.slac.stanford.edu/apps/explorer/

The web interface also allows you to generate file lists that can be fed into bbcp to export your data from SLAC to your home institution. You can use psexport or psimport for copying your data.

See the DataExportation page for more information.

Printing

The following printers are available in the NEH building from all the UNIX nodes:

Getting Help

The first line of help is available by sending email to pcdshelp@slac.stanford.edu.

If you have problems with a specific piece of the analysis software, please have a look at our bug-tracking system https://pswww.slac.stanford.edu/trac/psdm.

Running the Analysis

The analysis framework is documented in the Data Analysis page for the LCLS-I/HXR systems and psana for the LCLS-II (SXR&UED) systems. This section describes the resources available for running the analysis. The following figure shows a logic diagram of the LCLS data flow and indicates the different stages where data analysis can be performed in LCLS: 

1. Data acquisition - The online monitoring nodes get detector data over the network and place it in shared memory on each node. There is a set of monitoring nodes allocated for each instrument. The detector data are received over the network by snooping the multicast traffic between the DAQ readout nodes and the data cache layer. Analysis performed at this stage provides < 1 s feedback capabilities. The methods for doing (quasi) real time analysis are described in the Prompt Analysis page. Users should be aware of the different possibilities and choose the approach that works best for their experiment. 
2. Fast feedback - The processing nodes in the FFB system read the detector data from a dedicated, flash-based file system. It is possible to read the data as they are written to the FFB storage layer by the DAQ without waiting for the run to end.  Analysis performed at this stage provides < 5 min feedback capabilities. The resources reserved for this stage are described in the Fast Feedback System page.
3. Offline - The offline nodes read the detector data from disk. These systems include both interactive nodes and batch queues and are the main resource for doing data analysis. We currently support sending the data to three offline systems: psana, S3DF and NERSC. The psana system is the default offline system and your data will end up in psana unless you arrange a different destination with your experiment POC. The psana system is also relatively old and it will be retired when more storage becomes available in the S3DF system. Please consider running at NERSC if you expect to have intensive computing requirements (> 1 PFLOPS).
    

Image AddedIn case of an emergency that affect your data taking ability, please ask the instrument scientist or the floor coordinator. They have a contact list with all the people in the PCDS group.