Rockfish Citizen

The Rockfish cluster is used by hundreds of users on a daily basis. The Rockfish cluster has a few login nodes where all users connect to conduct their computing and research. The login nodes are an interface between all compute nodes and the users. We all need to follow good practices to create a friendly environment and ensure that one’s activity does not negatively impact the system and many or all other users.

DO NOT run jobs on the login nodes: Login nodes are used by hundreds of users to monitor their jobs, submit jobs, edit and manipulate files and sometimes to compile codes. We strongly request that users abstain from running jobs on login nodes. Sometimes users may want to run quick jobs to check that input files are correct or scientific applications are working properly. If this is the case, make sure this activity does not take more than a couple of minutes or even better request an interactive session (interact) to fully test your codes. Do not perform activities that may impact the file systems, for example rsync or copying large or many files from one file system to the other. Please use Globus (FAQs) to transfer large amounts of data. In short, do not run jobs or perform intensive computing on login nodes. Your account may be suspended and you will lose access to the queues if your activities are impacting other users.
What is allowed on login nodes:
    1. Request an interactive session
    2. Compile your codes, for example run “make”. Be careful if you are running commands with multiple processes, for example while “make -j 4” may be fine, “make -j 20” may impact other users.
    3. Check your jobs
      1. sqme
    4. Edit files, scripts, manipulate files
    5. Submit jobs
      1. sbatch script-name
    6. Look at output files
      1. more file-name
What is NOT allowed on the login nodes:
  1. Run jobs (./a.out or any other executables)
  2. rsync or copy large number of files
  3. Run applications like Matlab (use an interactive session)

Terms of Use

Definitions

User

Any individual with access to the Advanced Research Computing @ Hopkins (ARCH) facility resources.

Account

A username (userid) associated with a user, that allows access to ARCH resources. Read more here: Request an Account.

Allocation

A set number of service units (SUs) granted to a specific project. A single account (userid) may have access to one or more allocations.

Principal Investigator (PI)

A user that manages one or more allocations, including the ability to grant access to the allocation(s). PIs are usually faculty members.

Accounts and Allocations

PIs can create projects using the Coldfront Portal and request allocations according to the approved proposal (see JHU-Allocations). PIs can also give proxy status to any user if they require help to manage their projects, allocations and users. 

All users request an account through the Coldfront Portal. To be added to projects and allocations, a user must notify their PI after requesting the account.

One of ARCH’s goals is to promote collaborations. PIs can sponsor external accounts (ext-userid, this tag is used to identify users and more easily manage accounts).

Acceptance of Terms

Users, including external users, must agree to the Terms of Use herein to use any of ARCH resources. These Terms may be modified at any time and users must agree with the new Terms to continue using the resources. A breach of these Terms by a user may be grounds for termination of the account associated with the user.

Security

Each account is associated with a single user and must not be shared with other individuals, whether or not those individuals are affiliated with the user’s research group. Users must refrain from sharing passwords and have a responsibility to keep their passwords safe. The technical support team will never ask a user for their account password. However, there may be some cases when resetting a user’s password is necessary to troubleshoot issues. Users agree to immediately report the ARCH administration any security flaw found while refraining from exploiting the flaw.

Usage

Users agree to utilize the resources solely for the purposes outlined in the allocations associated with the account. All data and software stored and used must be legally obtained and follow the associated licenses and terms of use as required.

Users may only run computationally intensive applications using the compute nodes and by submitting jobs to the resource manager. 

Users must refrain from running resource intensive programs on the login nodes. Any procedure to circumvent the resource manager and directly use a compute node must be reported to the ARCH administration and not be exploited.

Users are responsible for abiding by the U.S. export controls regulation and acquiring licenses when needed.

Publications

All publications relying directly on data generated or software developed with ARCH resources shall acknowledge the allocation granted under which resources were used. PI’s should abide with the proposal guidelines and upload publications to the Rockfish portal as soon as these are available.