Research Computing SeminarsResearch Computing Seminars
Class Notes
Any member of the Lehigh community who would like to learn more about Research Computing and High-Performance Computing (HPC), but cannot easily attend our seminar series, is welcome to review our Research Computing Notes at the following location. Note that you must be on the campus network or VPN to access these notes:
Research Computing Seminar Notes
These note will be continuously updated during each Seminar Series listed below. Research Computing is characterized by a wide diversity of different projects and goals. If you would like to discuss your specific goals with our staff, schedule a "Research" consultation here.
Spring 2024 Seminars
In the Spring, 2024 semester we will continue our high-performance computing (HPC) seminar series. This series contains five parts which cover a wide breadth of high-performance computing fundamentals with an emphasis on surveying many different techniques that may help with your research. Experienced HPC users are welcome to attend any individual sessions that interest them, while we recommend that novice users attend the first three sessions, namely the quickstart, parallelism, and interactive computing sessions.
While there are no formal prerequisites, this series does not cover Linux fundamentals and the use of version control (typically git). To acquire these skills before the HPC seminar series, researchers of all skill levels can benefit from joining LinkedIn Learning, which is free for the Lehigh community. This provides access to two critical courses. First, the "Learning Linux Command Line" course will introduce you the Linux command line, which is essential for interacting with almost all high-performance computing platforms. Second, the "Git Essentials" course will teach you the skills required to collaboratively develop and share your code and electronic lab notebooks.
Notes from the Quickstart are available online (you must be on the Lehigh network to access this).
This series is held Tuesdays from 3/19-4/16 at 2PM in the EWFM Digital Media Studio.
| Date | Title |
|---|
| March 19 (register here) | | Expand |
|---|
| title | High-Performance Computing: Quickstart |
|---|
| Description: This seminar provides an introduction to high-performance computing (HPC) resources at Lehigh along with a walkthrough of the essential ingredients required to build research computing projects. Participants will learn the best practices for designing small, exploratory projects so that they can be scaled up to answer questions that require big data and parallel computations. This course will teach newcomers how to use Linux and Python to complete a simple but powerful research computing project. Instructor: Ryan Bradley |
|
| March 26 (register here) | | Expand |
|---|
| title | High-performance computing: Easy Parallel Computations |
|---|
| Description: Many exploratory research computing projects will eventually grow to require larger amounts of computational resources, either because they need to demonstrate that they are correct and reproducible, or to answer larger, more complicated research questions. This seminar will teach you to use the Lehigh high-performance computing (HPC) cluster to build parallel scientific software for answering larger questions using simple Linux tools along with the HPC scheduler (named SLURM). Instructor: Ryan Bradley |
|
| April 2 (register here) | | Expand |
|---|
| title | High-Performance Computing: Interactive Computing and Visualization |
|---|
| Description: Research computing projects of all sizes can benefit from interactive tools which make it easy to develop, debug, and visualize code and data on our local high-performance computing (HPC) systems. This seminar will demonstrate the use of a web portal to interact with your data and code, with a particular emphasis on connecting Jupyter notebooks to very large datasets and parallel computations. Instructor: Ryan Bradley |
|
| April 9 (register here) | | Expand |
|---|
| title | High-Performance Computing: Containers and Reproducibility |
|---|
| Description: Research computing projects can have a large impact if they are portable and easy to share with new collaborators. In this seminar, you will learn how to package your scientific computations in a Linux-based container system (Docker and Apptainer). Containers can make it easy to take your work to larger public supercomputing centers, and improve the reproducibility and openness of your scientific software. Instructor: Ryan Bradley |
|
| April 16 (register here) | | Expand |
|---|
| title | High-Performance Computing: Data Structures |
|---|
| Description: When a research computing project evolves to answer more nuanced questions, it inevitably requires more computational power and generates more data. This seminar will teach you the best practices for storing large, complex datasets so they can easily be analyzed and mined for new insights. These methods will apply to any research questions that depend on big data. Instructor: Ryan Bradley |
|
Fall 2023 Seminars
In the fall we will resume our high-performance computing (HPC) seminar series. We begin with two identical sessions of the HPC Quickstart session, and then learn about parallelism, interactive computing, containers, and data structures. Notes from the Quickstart are available online (you must be on the Lehigh network to access this).
Click here to visit the LTS website for dates, times, and registration information.
| Date | Title |
|---|
| Sep 19 or Sep 26 | | Expand |
|---|
| title | High-Performance Computing: Quickstart |
|---|
| Description: This seminar provides an introduction to high-performance computing (HPC) resources at Lehigh along with a walkthrough of the essential ingredients required to build research computing projects. Participants will learn the best practices for designing small, exploratory projects so that they can be scaled up to answer questions that require big data and parallel computations. This course will teach newcomers how to use Linux and Python to complete a simple but powerful research computing project. Instructor: Ryan Bradley |
|
| Oct 3 | | Expand |
|---|
| title | High-performance computing: Easy Parallel Computations |
|---|
| Description: Many exploratory research computing projects will eventually grow to require larger amounts of computational resources, either because they need to demonstrate that they are correct and reproducible, or to answer larger, more complicated research questions. This seminar will teach you to use the Lehigh high-performance computing (HPC) cluster to build parallel scientific software for answering larger questions using simple Linux tools along with the HPC scheduler (named SLURM). Instructor: Ryan Bradley |
|
| Oct 10 | | Expand |
|---|
| title | High-Performance Computing: Interactive Computing and Visualization |
|---|
| Description: Research computing projects of all sizes can benefit from interactive tools which make it easy to develop, debug, and visualize code and data on our local high-performance computing (HPC) systems. This seminar will demonstrate the use of a web portal to interact with your data and code, with a particular emphasis on connecting Jupyter notebooks to very large datasets and parallel computations. Instructor: Ryan Bradley |
|
| Oct 17 | | Expand |
|---|
| title | High-Performance Computing: Containers and Reproducibility |
|---|
| Description: Research computing projects can have a large impact if they are portable and easy to share with new collaborators. In this seminar, you will learn how to package your scientific computations in a Linux-based container system (Docker and Apptainer). Containers can make it easy to take your work to larger public supercomputing centers, and improve the reproducibility and openness of your scientific software. Instructor: Ryan Bradley |
|
| Oct 24 | | Expand |
|---|
| title | High-Performance Computing: Data Structures |
|---|
| Description: When a research computing project evolves to answer more nuanced questions, it inevitably requires more computational power and generates more data. This seminar will teach you the best practices for storing large, complex datasets so they can easily be analyzed and mined for new insights. These methods will apply to any research questions that depend on big data. Instructor: Ryan Bradley |
|
Summer 2023 Seminars
For the summer session, the research computing group will offer a single High-Performance Computing Quickstart training session, designed for small groups to learn more about completing their research computing projects on Lehigh University high-performance computing infrastructure.
This seminar reviews the minimum essential ingredients required to build research computing projects on Linux platforms. Students will learn the basics of Linux, use shell scripts, Python, and the scheduler on our high-performance computing (HPC) cluster to perform simple calculations. This course will outline the best practices for scaling and repeating many kinds of calculations. All students, faculty, and staff with an interest in learning more about HPC and Linux are welcome to attend.
Registration is available here for any one of the following dates.
- Tuesday, 6/13/23 from 1-2:30PM (in person, Digital Media Studio Lab)
- Wednesday, 6/21/23 1-2:30PM (Zoom)
- Tuesday, 6/27/23 from 1-2:30PM (in person, Digital Media Studio Lab)
- Wednesday, 7/5/23 from 1-2:30PM (Zoom)
Spring 2023 Seminars
The research computing seminars offered in the Spring semester are designed to lead participants through a survey of the wide range of available research computing techniques that they can apply to their training and research. Over the course of this seminar series, students will be expected to imagine how their projects can be formulated to take advantage of both small and large scale computation. We will begin with the end in mind, by first creating a map of the high-performance computing (HPC) resources available to the research community at Lehigh. After this introduction we will study the Linux basics required to effectively work with code and data, and then tackle the scheduler (SLURM) that we use to share a large HPC platform among many users. The final three sessions will provide a deeper dive into interactive computing with the Open OnDemand portal, building your own highly-portable software containers, and using Python for advanced computation on the cluster. Attendees are welcome to bring their own research problems and questions to these sessions.
...
| Date | Title |
|
|
|---|
| February 3 |
| Expand |
|---|
| title | Research Computing resources at Lehigh |
|---|
| Description: This seminar provides an overview of Research Computing resources available to the Lehigh research community. Instructor: Alex Pacheco |
| Slides | Recordings |
| February 10 |
| Expand |
|---|
| title | Linux: Basic Commands & Environment |
|---|
| Description: Linux is a free and open source operating system that is the OS of choice of the world's leading supercomputers as well as LTS HPC clusters and some computer labs at Lehigh. This session will provide an introduction to the linux/unix environment, command line basics, logging in to remote system, transferring data, vi/emacs editors etc to get started with using a Linux/Unix based computer. This seminar is geared towards researchers who want to learn or need to learn how to use a linux/unix based resource. Instructor: Sachin Joshi |
| Slides | Recordings |
| February 17 |
| Expand |
|---|
| title | Using SLURM scheduler on Lehigh's HPC cluster |
|---|
| Description: This seminar provides a hands on introduction to using the SLURM scheduler to submit and monitor jobs. SLURM is the scheduler on Lehigh's HPC resources, Sol and Hawk, and national supercomputing resources including XSEDE and NERSC. Prerequisites: An HPC account or an account on national supercomputing resources (XSEDE, DOE, etc) that uses SLURM. Familiarity with Linux/Unix environment, basic command and *nix editors such as vi or emacs is mandatory. Instructor: Alex Pacheco |
| Slides | Recordings |
| February 24 |
| Expand |
|---|
| Description: In this seminar, you will learn the basics of Python, including language fundamentals and basic programming. Prerequisites: Programming background is beneficial but not required. Instructor: Sachin Joshi |
| Slides | Recordings |
| March 3 |
| Expand |
|---|
| Description: In this seminar, you will learn the basics of R, including language fundamentals, data types, functions and basic programming including File I/O. Prerequisites: Programming background is beneficial but not required. Instructor: Jeremy Mack |
| Slides | Recordings |
| March 10 |
| Expand |
|---|
| title | Introduction to Open OnDemand |
|---|
| Description: Open OnDemand (OOD) is an NSF-funded open-source HPC portal developed by the Ohio Supercomputing Center. The goal of OOD is to provide an easy way for system administrators to provide web access to their HPC resources. This seminar introduces OOD to access interactive applications such as MATLAB, Jupyter Lab/Notebooks, and RStudio on Lehigh's HPC cluster. Prerequisites: An HPC account with an active allocation and a web browser. A Lehigh IP or VPN required. Instructor: Alex Pacheco |
| Slides | Recordings |
| March 17 |
| Expand |
|---|
| title | Data Visualization with Python |
|---|
| Description: This seminar provides a hands on introduction to Data Visualization using the Python programming language. Prerequisites: Programming background in Python is required. Instructor: Sachin Joshi |
| Slides | Recordings |
| March 24 |
| Expand |
|---|
| title | Data Visualization with R |
|---|
| Description: This seminar provides a hands on introduction to Data Visualization using the R programming language Prerequisites: Programming background in R is required. Instructor: Jeremy Mack |
| Slides | Recordings |
| March 31 |
| Expand |
|---|
| title | Bring Your Own Software: Containers on HPC Resources |
|---|
| Description: Singularity is an open-source container engine designed to bring operating system-level virtualization to scientific and high-performance computing. In this seminar, we’ll provide an overview of Singularity and how you can Build Your Own Software in a singularity container for use in your research on Sol. Prerequisites: A Linux system with singularity installed (check your package manager if your distribution provides singularity or see installation instructions). Instructor: Alex Pacheco |
| Slides | Recordings |
| April 7 |
| Expand |
|---|
| title | Object-Oriented Programming with Python |
|---|
| Description: Object Oriented Programming or OOP is a programming paradigm which provides a means of structuring programs (combining data and functionality) so that properties and behaviors are bundled into individual objects. In this seminar, you’ll learn the basic concepts of OOP in Python: Python Classes, Object Instances, Defining and Working with Methods and OOP Inheritance. Prerequisites: Programming background in Python is required. Knowledge of OOP’s concepts is beneficial but not required. Instructor: Sachin Joshi |
| Slides | Recordings |
| April 14 |
| Expand |
|---|
| Description: Shiny Apps, interactive web applications built using R programming language, have grown in popularity. This session will provide attendees experience using the Shiny package and the idea of reactive programming, which forms the basis of building an interactive web application. Prerequisites: Programming background in R is required. Instructor: Jeremy Mack |
| Slides | Recordings |
...
| Title | Downloads |
|
|---|
| Expand |
|---|
| title | Research Computing resources at Lehigh |
|---|
| Description: This training provides an overview of Research Computing resources available to the Lehigh research community. |
| Slides | Recordings |
Writing an Allocation Proposal | Slides | Recordings |
| Expand |
|---|
| title | Using SLURM scheduler on Sol |
|---|
| Description: This training provides a hands on introduction to using the SLURM scheduler to submit and monitor jobs. SLURM is the scheduler on Lehigh's HPC resource, Sol, and national supercomputing resources including XSEDE and NERSC. Prerequisites: An account on Sol or an account on national supercomputing resources (XSEDE, DOE, etc) that uses SLURM. Familiarity with Linux/Unix environment, basic command and *nix editors such as vi or emacs is mandatory. |
| Slides | Recordings |
| Expand |
|---|
| title | Introduction to Open OnDemand |
|---|
| Description: Open OnDemand (OOD) is an NSF-funded open-source HPC portal developed by the Ohio Supercomputing Center. The goal of OOD is to provide an easy way for system administrators to provide web access to their HPC resources. This tutorial introduces OOD to access Lehigh's Sol cluster. Prerequisites: An account on Sol with an active allocation and a web browser. A Lehigh IP or VPN required. |
| Slides | Recordings |
| Expand |
|---|
| title | Bring Your Own Software |
|---|
| Description: This seminar is geared towards users who wish to bring or build their own software stack to use on Sol and Hawk (or even local linux systems). Topics to be covered include best practices for installing packages using make, cmake and configure, SPACK package manager, and Singularity. Prerequisites: Some familiarity with compilers, and linux environment is required. |
| Slides |
|
| A Brief Introduction to Linux | Slides | Recordings |
| Expand |
|---|
| title | Linux: Basic Commands & Environment |
|---|
| Description: Linux is a free and open source operating system that is the OS of choice of the world's leading supercomputers as well as LTS HPC clusters and some computer labs at Lehigh. This session will provide an introduction to the linux/unix environment, command line basics, logging in to remote system, transferring data, vi/emacs editors etc to get started with using a Linux/Unix based computer. This training is geared towards researchers who want to learn or need to learn how to use a linux/unix based resource. |
| Slides | Recordings |
| Basic Shell Scripting | Slides | Recordings |
| Advanced Shell Scripting | Slides | Recordings |
| Expand |
|---|
| Description: In this tutorial, you will learn the basics of R, including language fundamentals, data types, functions and basic programming including File I/O. Prerequisites: Programming background is beneficial but not required. |
| Slides | Recordings |
| Expand |
|---|
| title | Data Visualization with R |
|---|
| Description: This tutorial provides a hands on introduction to Data Visualization using the R programming language Prerequisites: Programming background in R is required. |
| Slides | Recordings |
| Expand |
|---|
| Description: Shiny Apps, interactive web applications built using R programming language, have grown in popularity. This session will provide attendees experience using the Shiny package and the idea of reactive programming, which forms the basis of building an interactive web application. Prerequisites: Programming background in R is required. |
| Slides | Recordings |
| Expand |
|---|
| Description: In this seminar, you will learn the basics of Python, including language fundamentals and basic programming. Prerequisites: Programming background is beneficial but not required. |
| Slides | Recordings |
| Expand |
|---|
| title | Python Data Structures |
|---|
| Description: Data Structures are the fundamental constructs around which you build your programs. Python is a high-level, interpreted, interactive and object-oriented scripting language using which we can study the fundamentals of data structure in a simpler way as compared to other programming languages. In this seminar we are going to study a short overview of some frequently used data structures in general and how they are related to some specific python data types. Prerequisites: Programming background in Python is required. Knowledge of data structure concepts is beneficial but not required. |
| Slides | Recordings |
| Expand |
|---|
| title | Data Visualization with Python |
|---|
| Description: This seminar provides a hands on introduction to Data Visualization using the Python programming language. Prerequisites: Programming background in Python is required. |
| Slides | Recordings |
| Expand |
|---|
| title | Object-Oriented Programming with Python |
|---|
| Description: Object Oriented Programming or OOP is a programming paradigm which provides a means of structuring programs (combining data and functionality) so that properties and behaviors are bundled into individual objects. In this seminar, you’ll learn the basic concepts of OOP in Python: Python Classes, Object Instances, Defining and Working with Methods and OOP Inheritance. Prerequisites: Programming background in Python is required. Knowledge of OOP’s concepts is beneficial but not required. |
| Slides | Recordings |
| Expand |
|---|
| Description: Machine learning is a method of data analysis that automates analytical model building. It is a branch of artificial intelligence (AI) based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention. In this two part seminar you’ll learn the basic machine learning concepts: supervised and unsupervised learning, classification and regression algorithms. This is a two part seminar. Prerequisites: Basic knowledge of general CS concepts such as data structures, linear algebra and design of algorithm is required. |
| Slides (Part I, Part II) | Recordings (Part I, Part II) |
| Expand |
|---|
| Description: Text mining, also known as Text Analytics is the process of deriving high-quality information from text. It is the process of examining large collections of written resources to generate new information and to transform the unstructured text into structured data for use in further analysis. In this seminar you’ll learn the basic introduction to text analysis, different steps of text analysis: gathering textual data, cleaning and preparing textual data , analyzing the data (ETL) and visualizing textual data using tools from the HathiTrust Research Center (HTRC). Prerequisites: Programming background is required. Basic knowledge of general CS concepts such as data structures, linear algebra and design of algorithms is required. |
| Slides | Recordings |
| Expand |
|---|
| Description: MATLAB is a programming platform designed specifically for engineers and scientists. MATLAB is a special-purpose language that is an excellent choice for writing moderate-size programs that solve problems involving the manipulation of numbers. MATLAB is easy to learn, versatile and the design of the language makes it possible to write a powerful program in a few lines. Using MATLAB, you can: Analyze data, Develop algorithms and Create models and applications. In this seminar you’ll learn the basics of MATLAB, including language fundamentals and basic programming that will help you achieve the above functionalities. Prerequisites: Programming background is beneficial but not required. |
| Slides |
|
| Version Control with GIT | Slides |
|
| Document Creation with LaTeX | Slides |
|
| Using Virtualized Software at Lehigh |
|
|
| Storage Options at Lehigh | Slides |
|
| Research Data Management | Slides |
|
| Enhancing Research Impact | Slides |
|
...