The following words and phrases are used throughout the MSI website and unless otherwise stated, the meaning of these terms is defined as follows. Please be aware that the definitions given here are not intended to be universal; rather, the definitions are specifically tailored to fit the environment at MSI.
The resources that are provided to each PI group free of charge each year. Groups will receive allocations of both storage (GBs/TBs) and CPU time (in the form of Service Units (SUs). With additional justification, additional resources can be allocated or purchased for the PI group (see definition "Purchased Resources").
The process by which research groups renew their access to MSI's HPC services for the next calendar year. Each Principal Investigator or Group Administrator is responsible for renewing their group's access, and annual renewal is required to maintain access.
Group Administrator (Group Admin)
A member of a Principal Investigator's group who has privileges to carry out administrative functions for the group, such as adding new users and giving group members access to HPC resources. The PI may assign this role to any MSI user in their group.
The scheduling queues used to efficiently allocate MSI hardware and resources to different types of jobs. Queues act as a sort of "waiting list" for jobs with similar hardware and resource needs, ensuring fairness and efficiency in job completion. Each system (Mesabi and Mangi) have their own queues to accomodate different types of jobs.
A software used to provide users with access to software-on-demand through Linux systems. Modules allow users to load and unload different software versions to meet their computational needs. Users may also create private software modules of software installed within their home or group directories.
Script used in a small text file when submitting a job through the MSI queuing system, known as the Portable Batch System (PBS). This script includes information about requested resources and commands for executing the calculation. PBS Scripts help ensure that jobs are submitted to a relevant queue, allocated the correct amount of resources, and perform the desired functions of the user.
A global high performance network file system that contains user home directories and shared group directories. The file system is available (exported) on all of MSI's computational resources, so that users will see the same files and directories no matter what system they are using.
Principal Investigator (PI)
The primary individual responsible for access and research conducted using MSI systems. Pricipal investigators can apply for MSI accounts for themselves and members of their research groups.
Individuals that may be eligible to become a PI include faculty, instructor faculty, emeritus faculty, and academic professionals at the University of Minnesota, as well as faculty researchers from other accredited institutions of post-secondary education in the state of Minnesota.
Resources that may be purchased beyond the allocations given to each PI group. PIs may request to purchase dedicated resources, such as additional storage and hardware, to suit the needs of their research that normal allocations may not be able to meet. Additional justification may be needed for some purchases.
Shared temporary storage used while applications are running on the supercomputer. Analogous to "scratch paper," scratch storage is not intended for valuable, long-term, or backup data storage, but instead is available for temporary file storage while executing a job. MSI offers both large capacity and high performance temporary storage.
Second Tier Storage
An object storage system in place for storing inactive data, supporting data lifecycles and workflows, and facilitating data management and sharing. Second tier storage is ideal for those looking to share data with colleagues around the world, including some storage through Amazons S3 (Simple Storage Service) interface.
Service Units (SUs)
A unit of time on MSI HPC resources. The SU to CPU-hour conversion rates for various systems are listed on the Queues page. Every group at MSI is allotted a default allocation of 70,000 SUs each calendar year, which is usually enough for many jobs done using MSI resources. However, groups may request more SUs with proper justification. The default allocation of SUs is renewed each calendar year upon renewal of the PI group.
A read-only copy of data stored at a particular point in time, maintaining the same permissions for each file at that point in time. These snapshots static backups of a volume can be used to restore a file to its location in the case that it was inadvertently deleted. You can also snapshot while running Virtual Machines (VM).
Another given name for a machine capable of High Performance Computing (HPC). Supercomputers, including the ones at MSI, have access to the high speed networks, high performance storage, powerful GPUs, and large amounts of memory needed to execute large-scale computations and programs. MSI has two main HPC units, Mesabi and Mangi, designed to support complex computation and memory-intensive programs used in research across the University as well as the state of Minnesota.
Off-line storage used to remove and archive large amounts of data from MSI systems. Data is compressed and stored.
MSI hosts various tutorial seminars to assist users with a variety of topics, including how to utilize MSI's services and various software packages. These tutorials are free, and are hosted both in person and online via YouTube.