Join RMACC leadership to discuss upcoming grant possibilities. Hear about what others are working on, pitch your ideas to other members of RMACC and network to form connections/collaborations.
Modern single-cell and single-nucleus RNA sequencing allow us to profile every cell within a brain tumor, uncovering the diverse lineages and cell signals that drive growth and therapy resistance. However, each experiment can produce terabytes of raw reads and millions of barcodes, demanding significant CPU, GPU, and memory resources – far beyond the limits of a laptop. This talk will show how high-performance computing (HPC) systems transform that data deluge into biological insight.
I will walk through an end-to-end analysis pipeline that pairs the Cell Ranger aligner with an nf-core workflow for efficient, reproducible processing on CPU and GPU nodes. Interactive exploration then transitions to Seurat, where large-memory nodes accelerate dimensionality reduction, clustering, and differential expression analysis and integration of hundreds of thousands of cells. HPC infrastructure also enables RNA velocity calculations, and trajectory analysis that would be impractical on local workstations.
Benchmarks will illustrate how parallel job arrays and optimized space management can cut runtimes from days to hours while lowering costs. My goal is to provide researchers and students with a clear roadmap for harnessing supercomputers to advance neuro-oncology and other data-intensive areas of life science.
Fuzzball transforms your on-premises or cloud environment into a user-friendly, manageable HPC cluster. In this talk, I’ll share exciting updates from the past year. Fuzzball is now generally available (GA) for on-prem installations and is also offered as a tech preview in the AWS Marketplace. This means you can try Fuzzball in your AWS account and deploy it directly on your on-prem cluster. We’ve also introduced Fuzzball Federate as a tech preview, enabling you to manage multiple Fuzzball clusters through a single console and unified user interface. Additionally, Fuzzball now includes a built-in workflow catalog, featuring CIQ-certified examples of popular applications in an easy-to-use format.
NVIDIA's GH200 Grace Hopper Superchip offers strong potential for accelerating large-scale AI and HPC workflows through its tightly integrated CPU-GPU architecture. In this talk, we share CU Boulder Research Computing’s first-hand experience providing GH200 nodes to users. We'll cover the GH200 architecture, our approach to the software stack, an overview of our beta testing phase, and successful use cases run on the GH200s. We'll conclude with potential future directions for CURC’s GH200 resources and describe how RMACC members can access the nodes for their research or educational purposes.
Fostering a positive work environment is crucial for both employee satisfaction and organizational success. This presentation will focus on two key pillars: building trust and fostering belonging. We will explore how trust serves as the foundation for effective communication, collaboration, and performance, while a sense of belonging ensures that employees feel valued and connected to the organization. Attendees will gain practical insights and actionable strategies for cultivating trust through transparency, consistency, and empathy, as well as creating an inclusive culture where every individual feels respected and integral to the team’s success.
Please join Mark III and NVIDIA at RMACC for an overview, walk-through, and live demo of building an AI agent app quickly and easily using NVIDIA NIMs and Blueprints. The first segment of this session will focus on a build of a simple AI agent bot built with a Llama 3 NIM. The second segment will focus on a live build and demo of a more advanced AI agent build with an NVIDIA Blueprint (powered by multiple NIMs). Lastly, the session will show how to quickly and easily finetune a model before being served up for API consumption by apps via NIMs. This session will be of interest to participants of all levels and skillsets looking to deploy AI services into their existing and new cloud-native and modern apps and research as quickly and effectively as possible.
Are you still debugging your code with print statements and commenting out chunks of code? Would you like a more efficient way to observe and test your programs? Then you should consider incorporating Visual Studio Code’s (VS Code) built-in debugger! In this workshop we will cover how to use the VS Code debugger for testing a software program at run time. You will learn how to pause a running program at key locations in your code (breakpoints) and observe and/or update your project’s variables. To follow along with the hands-on component of this workshop, you will need to bring a personal laptop computer. You will also need to either have VS Code installed on your laptop computer or have a Research Computing account so that you can use VS Code through OnDemand. You can register for a Research Computing account beforehand (this will take about 10 minutes).
Warewulf is an open-source cluster provisioning and management system primarily targeting HPC deployments. Its latest feature release, v4.6, includes a ton of new features, including an automated upgrade process, a completely new kernel management system, a new REST API, better management of site-local overlays, more dynamic and expressive overlay templates, a brand new abstract resources system, and significant performance improvements for large clusters.
In this session we will go over these changes and more in detail and showcase a new web interface from CIQ built specifically for Warewulf utilizing its new REST API. We'll also take audience questions, with a look into the future of the v4.6 release series.
AI4WY is an NSF MRI-supported project seeking to build regional partnerships to transform the research landscape by acquiring a state-of-the-art high-performance computing (HPC) system. Led by the University of Wyoming in collaboration with Colorado State University and the Rocky Mountain Advanced Computing Consortium (RMACC), the AI4WY cluster will feature NVIDIA Grace Hopper Superchips for empowering AI-driven research and big data modeling across key domains such as environment, agriculture, society, and energy. In this presentation, we will provide an update on the system acquisition, expanding HPC access to RMACC through the NSF ACCESS program, and plans for fostering regional collaboration.
High-performance computing (HPC) research thrives on collaboration, yet institutional data silos often hinder progress. This extends beyond the storage hardware itself to where data is being siloed in departments and institutions alike. Accelerating scientific progress and innovation requires enabling secure data discovery and sharing across not only institutions, but scientific disciplines as well. Federated access models, controlled permissions, and distributed compute environments enable seamless yet secure collaboration. By facilitating data discovery across disciplines and optimizing shared infrastructure, organizations can break down barriers, enhance research efficiency, and drive cross-disciplinary insights that push the boundaries of scientific advancement.
The National Artificial Intelligence Research Resource (NAIRR) is a federal initiative aimed at providing researchers with greater access to advanced AI tools, datasets, and computing infrastructure. By connecting academic institutions, national labs, and government agencies, NAIRR is building a shared ecosystem to support AI-driven discovery across scientific disciplines. This talk will highlight the goals of the NAIRR pilot, outline its current offerings, and explore how institutions in the RMACC community can engage with and benefit from this growing national effort.
With the recent changes to the NIH Genomic Data Sharing Policy, more focus is being placed on compliance today. Organizations are looking for solutions to meet these new requirements. In this session we will discuss how Globus data management services allow users to confidently work with controlled-access data. We will cover all the product features that address managing controlled-access data to enable compliance with NIST SP 800-171. We will demonstrate how with High Assurance collections the service meets these new requirements.
This presentation explores key milestones in the Colorado quantum ecosystem’s growth, including the establishment of anchor institutions like JILA and NIST, the role of state and federal policy, and the rise of quantum-focused companies across sensing, computing, and communication. We also examine how Colorado’s collaborative culture, talent pipeline, and public-private partnerships have created a fertile ground for quantum innovation. Looking ahead, the presentation highlights the challenges and opportunities that will shape the next phase of Colorado’s quantum leadership.
CCMNet unites cyberinfrastructure professionals to enhance community support for research and academic institutions, maximizing cyberinfrastructure utilization for research advancement through peer mentoring engagements. This presentation will introduce the program, and feature a real time walkthrough of the web portal to help interested individuals dive straight in.
Many research computing providers are currently working to comply with enhanced data security requirements mandated by U.S. funding agencies to protect Controlled Unclassified Information (CUI) and other sensitive data when used on their cyberinfrastructure. The most common data security frameworks include the National Institute of Standards and Technology (NIST) 800-171 for CUI, the Cybersecurity Maturity Model Certification (CMMC) for defense-related CUI, and the Health Insurance Portability and Accountability Act (HIPAA) for health information. Progress toward compliance with these frameworks is a complex and iterative process, and therefore numerous approaches have been undertaken. This panel discussion provides a forum to share information, experiences, and ideas regarding secure HPC.
This presentation provides a comprehensive overview of AMD's current and future technology roadmap, highlighting key innovations, product developments, and strategic direction across its CPU, GPU, and APU portfolios.
Modern scientific computing demands flexible and scalable solutions that bring computing power closer to data while maintaining security and ease of use. We propose to present our solution which leverages Kubernetes to provide a platform to our employees, university members, and partner organizations that meets those demands and complements our existing HPC system. This presentation will cover how we utilize Continuous Integration and Continuous Delivery (CI/CD), coupled with GitOps and DevOps practices, to provide a robust and secure platform for hosting container-based workloads. These workloads include interactive web visualizations, JupyterHub instances, science gateways, data assimilation tools, data analysis tools, and those that require access to GPUs, including, but not limited to, AI/ML. The presentation will also cover how Cilium network and Kyverno access policies are implemented to secure the platform. It will also discuss how GitHub Actions are utilized to test and build codebases into containers that can run on the platform. Attendees will learn why we chose Kubernetes as our platform as well as practical strategies to implement similar solutions and common pitfalls to avoid.
Tired of wrestling with Slurm? This fast-paced, hands-on workshop will equip researchers with the essential tools and strategies to handle common challenges such as job failures, long queue times, and resource allocation issues. Whether you're a beginner or have some experience with SLURM, this workshop will enable you to design, build, and troubleshoot efficient HPC research workflows. If you would like to practice some of the hands-on exercises at the end of the session, register for a Research Computing account beforehand (this will take about 10 minutes).
