Skip to main content

30 posts tagged with "klone"

View All Tags

September 2025 Maintenance Update

· 6 min read
Kristen Finch
Kristen Finch
Director of Research Computing Solutions

During September’s maintenance, we refreshed the operating system images for both login and compute nodes, upgraded our NVIDIA driver, and performed network configuration steps to support Tillicum GPU cluster.

Stay informed by subscribing to our mailing list and the UW-IT Research Computing Events Calendar. The next maintenance is scheduled for Tuesday, October 14, 2025 (the second Tuesday of the month).

Notable Updates​

  • Node image updates – Routine updates and security patches.
  • NVIDIA Driver upgrade to 580.82.07 – this upgrade enables CUDA 13.
  • Network configuration changes to support Tillicum.

A New Look!​

We continue to make updates to our documentation website to make sure you have access to everything you need from UW-IT Research Computing. Check out these pages and features and let us know what you think by emailing help@uw.edu with "hyak.uw.edu" in the subject line:

  • Home page content cards - find content cards that serve as quick links to documentation, learning resources, and support.
  • Learn page - the new home for our training and video library either linking directly to videos or tutorial pages.
  • Pricing menu - our pricing menu now leads to our knowledge base articles for each of our services. Learn more about your compute and storage options from these resources.
  • UW-IT Research Computing Landing Page - this documentation website (hyak.uw.edu) is shifting to a focus on documentation alone, to learn more about our services, use the Pricing menu mentioned above or visit our Research Computing Landing Page on the UW-IT website.

Office Hours​

If you would like to request 1 on 1 help, please send an email to help@uw.edu with "Hyak Office Hour" in the subject line to coordinate a meeting.

AWS GameDay​

Save the date! Research Computing Club along with UW-IT Research Computing and eScience will be hosting a AWS GameDay on Thursday November 20, 2025. Registration link coming soon!

What is it? An epic hands-on competition where you'll build and deploy real AI solutions using AWS's latest generative AI services. Work in teams to solve fun challenges and create magical applications!

Why Should You Join?

  • Get practical experience with cutting-edge AI technology
  • Build your AWS cloud resume
  • Network with tech industry professionals
  • Win awesome prizes and swag
  • Free food and energy drinks

UW Community Opportunities​

  • The Data Science and AI Accelerator pairs eScience Institute data scientists with researchers from any field of study to work on focused, collaborative projects. Collaborations may center on analysis of an existing dataset to answer a specific research question, an implementation of software for processing or analyzing data, data visualization tools, or tools for data interpretation. Accelerator Projects may be submitted at any time. Projects for Winter 2026 must be received by October 26th, 2025. LEARN MORE HERE.
  • The eScience Institute invites applications for the title of “UW Data Science Postdoctoral Fellow”. As a former Data Science Fellow, I can say this is a great program with opportunities for funding for travel and more! Applications will be accepted and evaluated twice per year with due dates of 12pm midnight PT on September 15th and March 15th. LEARN MORE HERE..
  • DubHacks - October 18 - October 19, 2025 - DubHacks 2025 takes you back to where it all began—the childhood bedroom. A space for imagination, curiosity, and bold ideas. Now, with code instead of crayons, you get to build what makes your younger self proud. No limits, just pure creativity. What will you create when you let yourself play?

External Training Opportunities​

  • Intro to R: Tidyverse, ggplot2, dplyr 09/12/25 - 1:00 PM - 2:00 PM EDT Learn more and register.
  • CU-RMACC Webinars: Introduction to AI for All 09/15/25 - 12:00 PM - 1:30 PM EDT Learn more and register.
  • ACES: Python for New Programmers 09/16/25 - 10:00AM-4:00PM CT Learn more and register.
  • COMPLECS: Data Transfer 09/18/25 - 2:00 PM - 3:30 PM EDT Learn more and register.
  • ACES: AI TechLab in Jupyter Notebooks 09/23/25 - 1:30PM-4:00PM CT Learn more and register.
  • ALCF Hands-on HPC Workshop – Sep 23 – Sep 25, 2025 The workshop is designed to advance users’ skills in high-performance computing and artificial intelligence on ALCF systems (Polaris and Aurora). The virtual week (Sept 23-25) covers the fundamentals of building, compiling, and running applications. Learn more and register.
  • Globus Office Hours 09/24/25 - 4:00 PM - 5:00 PM EDT Learn more and register.
  • ACES: Python for Programmers New to Python 09/30/25 - 10:00AM-4:00PM CT Learn more and register.
  • Applied Parallel Programming with TACC 10/01/25 - 10:00 AM to 10/08/25 - 5:00 PM EDT Learn more and register.
  • ParaView - Scientific Data Analysis and Visualization Training – Oct 2, 2025, 10 am – 2 pm PDT ParaView is an open-source, multi-platform scientific data analysis and visualization tool that enables the analysis and visualization of extremely large datasets. This course provides an overview of ParaView, including how to visualize and process data. Learn more and register.
  • COMPLECS: Interactive Computing 10/09/25 - 2:00 PM - 3:30 PM EDT Learn more and register.
  • COMPLECS: Linux Shell Scripting 10/23/25 - 2:00 PM - 3:30 PM EDT Learn more and register.
  • COMPLECS: Using Regular Expressions with Linux Tools 11/06/25 - 2:00 PM - 3:30 PM EST Learn more and register.
  • COMPLECS: Batch Computing (Part III) High-Throughput and Many-Task Computing - Slurm Edition 12/04/25 - 2:00 PM - 3:30 PM EST Learn more and register.
  • R for HPC 12/04/25 - 2:00 PM - 3:30 PM EST Learn more and register.

Questions about Hyak Klone, Tillicum, or any other UW-IT Research Computing Service? Fill out our Research Computing Consulting intake form. We are here to help!

Happy Computing,

Hyak Team

August 2025 Maintenance Update

· 6 min read
Kristen Finch
Kristen Finch
Director of Research Computing Solutions

During August’s maintenance, we refreshed the operating system images for both login and compute nodes, upgraded Slurm to version 25.5.2, and upgraded Klone's filesystem (GPFS) for increased stability. We also introduced a new Globus OneDrive connector, making it easier than ever to transfer files between OneDrive and Hyak Klone or Kopah Storage.

Stay informed by subscribing to our mailing list and the UW-IT Research Computing Events Calendar. The next maintenance is scheduled for Tuesday, September 9, 2025 (the second Tuesday of the month).

Notable Updates​

  • Node image updates – Routine updates plus installation of new Slurm utilities that we will test for job efficiency monitoring.
  • Slurm upgrade to 25.5.2 – Resolves a bugs the --gres flag allowed resources to be allocated to more than one job in version 25.05.0 and fixes X11 forwarding. Read more about this version.
  • GPFS upgrade to 5.1.9.11 – Improves stability and includes several bug fixes. Read more about this version.

New Features​

Globus OneDrive Connector – UW-IT Research Computing has added OneDrive as a connector to Globus, making transfers between OneDrive and Hyak Klone or OneDrive and Kopah Storage easier than ever before!

Search Uw OneDrive to utilize the connector.

Things to note​

  • Did you know that the UW Community are eligible for 5TB of storage on OneDrive as part of the Office 365 Suite? Click here to learn more.
  • While OneDrive is HIPAA and FERPA compatible, encryption is not enforced for Globus transfers on any of our connectors (OneDrive, Kopah, Klone). As a reminder, Klone and Kopah are NOT aligned with HIPPA, please keep this in mind now that OneDrive can transfer to either cluster.
  • Sharing with external partners is not enabled for our OneDrive or Klone connectors via Globus. Sharing is permitted for Kopah.
  • Read more

Office Hours​

  • Wednesdays at 2pm on Zoom. Attendees need only register once and can attend any of the occurrences with the Zoom link that will arrive via email. Click here to Register for Zoom Office Hours.
  • Thursdays at 2pm in person in eScience. (address: WRF Data Science Studio, UW Physics/Astronomy Tower, 6th Floor, 3910 15th Ave NE, Seattle, WA 98195).
  • See our office hours schedule, subscribe to event updates, and bookmark our UW-IT Research Computing Events Calendar.

If you would like to request 1 on 1 help, please send an email to help@uw.edu with "Hyak Office Hour" in the subject line to coordinate a meeting.

UW Community Opportunities​

  • The Data Science and AI Accelerator pairs eScience Institute data scientists with researchers from any field of study to work on focused, collaborative projects. Collaborations may center on analysis of an existing dataset to answer a specific research question, an implementation of software for processing or analyzing data, data visualization tools, or tools for data interpretation. Accelerator Projects may be submitted at any time. Projects for Fall 2025 must be received by August 14th, 2025. LEARN MORE HERE.
  • Applications for the CSDE Data Science and Demography Training program are due Friday, August 22nd by 5pm. An information session will take place Wednesday, August 13th at 10:00 a.m. DETAILS HERE.
  • Cloud Clinic August 14 10-11am - guest presenter Niris Okram from AWS presenting on “The Utility of Capacity Blocks: Optimizing computing horsepower per budget dollar.” This will be followed by a short presentation on building small-scale (“Littlest”) JupyterHubs. LEARN MORE HERE.
  • DubHacks - October 18 - October 19, 2025 - DubHacks 2025 takes you back to where it all began—the childhood bedroom. A space for imagination, curiosity, and bold ideas. Now, with code instead of crayons, you get to build what makes your younger self proud. No limits, just pure creativity. What will you create when you let yourself play?

External Training Opportunities​

  • Automating Research with Globus: The Modern Research IT Platform - Aug. 18, 2025, 9 a.m. – 12 p.m. (Pacific Time) This workshop introduces Globus Flows and its role in automating research workflows. Participants will explore data portals, science gateways, and commons, enabling seamless data discovery and access. Enroll here.
  • CU-RMACC Webinars: Should I be Scared of AI? Aug. 18, 2025 - 3:00 PM - 4:00 PM EDT Throughout history, new technologies have sparked both excitement and fear—AI is no different. In this talk, Dr. Shelley Knuth, Assistant Vice Chancellor for Research Computing at the University of Colorado explores the common fears surrounding artificial intelligence, why we feel them, and how we can shift our perspective to focus on positive outcomes. We’ll look at practical ways to address risks, embrace innovation, and move forward with AI as a powerful tool rather than something to fear. Learn more and register.
  • COMPLECS: Batch Computing (Part II): Getting Started with Batch Job Scheduling 08/21/25 - 2:00 PM - 3:30 PM EDT Learn more and register.
  • NUG Community Call: A Birds-Eye View of Using Cuda with C/C++ on Perlmutter (Part 2) August 27, 2025, 11 a.m. - 12:30 p.m. PDT - In this two-part training series, users will be introduced to the basics of using CUDA on Perlmutter at NERSC. The training will focus on the basics of the Perlmutter architecture and NVIDIA GPUs, programming concepts with CUDA using C/C++. Event 2 focuses on advance kernel and custom cuda kernels in C/C++. Learn more and register.
  • COMPLECS: Linux Tools for Text Processing 09/04/25 - 2:00 PM - 3:30 PM EDT Learn more and register.
  • Python for HPC 09/09/25 - 2:00 PM - 3:30 PM EDT Learn more and register.
  • COMPLECS: Data Transfer 09/18/25 - 2:00 PM - 3:30 PM EDT Learn more and register.
  • COMPLECS: Interactive Computing 10/09/25 - 2:00 PM - 3:30 PM EDT Learn more and register.
  • COMPLECS: Linux Shell Scripting 10/23/25 - 2:00 PM - 3:30 PM EDT Learn more and register.
  • COMPLECS: Using Regular Expressions with Linux Tools 11/06/25 - 2:00 PM - 3:30 PM EST Learn more and register.
  • COMPLECS: Batch Computing (Part III) High-Throughput and Many-Task Computing - Slurm Edition 12/04/25 - 2:00 PM - 3:30 PM EST Learn more and register.
  • R for HPC 12/04/25 - 2:00 PM - 3:30 PM EST Learn more and register.

Positions​

  • Two PhD positions in Artificial Intelligence - in collaboration with German Aerospace Center and TU Dresden, Germany. Deadline to apply: 27 August 2025. Apply Now!

Questions about Hyak Klone, Tillicum, or any other UW-IT Research Computing Service? Fill out our Research Computing Consulting intake form. We are here to help!

Happy Computing,

Hyak Team

July 2025 Maintenance Update

· 6 min read
Kristen Finch
Kristen Finch
Director of Research Computing Solutions

During June's maintenance, we've refreshed the operating system images for both login and compute nodes including the newest version of Slurm, and we have implemented some changes critical to provisioning our new GPU system, Tillicum (launching in Fall 2025). Stay informed by subscribing to our mailing list and the UW-IT Research Computing Events Calendar. The next maintenance is scheduled for Tuesday August 12, 2025 (AKA the 2nd Tuesday of the month).

Notable Updates​

  • Routine package updates – images for both the login and compute nodes have been refreshed to incorporate the latest Linux OS security updates and system patches.
  • Slurm Upgrade to version 25.05 – Slurm 25.05 introduces encrypted job communication, improved support for complex network topologies, and new features like optional TLS, job start events in Kafka, and better license request handling. While you won’t notice major changes in your day-to-day workflow, this upgrade improves security, enables more flexible job scheduling, and lays the groundwork for new features in the future. Learn more from Slurm's release notes.
  • SSHD changes – We’ve updated some behind-the-scenes SSH settings to improve login handling. These changes help ensure account access stays consistent across Klone and Tillicum, but you won’t need to do anything differently when connecting.

New Training Videos​

This month we uploaded several training videos to our YouTube Playlist that may be of interest:

Summer Office Hours​

  • Wednesdays at 2pm on Zoom. Attendees need only register once and can attend any of the occurrences with the Zoom link that will arrive via email. Click here to Register for Zoom Office Hours.
  • Thursdays at 2pm in person in eScience. (address: WRF Data Science Studio, UW Physics/Astronomy Tower, 6th Floor, 3910 15th Ave NE, Seattle, WA 98195).
  • See our office hours schedule, subscribe to event updates, and bookmark our UW-IT Research Computing Events Calendar.

If you would like to request 1 on 1 help, please send an email to help@uw.edu with "Hyak Office Hour" in the subject line to coordinate a meeting.

External Training Opportunities​

  • NVIDIA Workshop: Building Transformer-Based Natural Language Processing Applications - 07/09/25 - 10:00 AM - 6:00 PM EDT Learn how to apply and fine-tune a Transformer-based Deep Learning model to Natural Language Processing (NLP) tasks. In this course, you'll: · Construct a Transformer neural network in PyTorch · Build a named-entity recognition (NER) application with BERT · Deploy the NER application with ONNX and TensorRT to a Triton inference server Upon completion, you’ll be proficient in task-agnostic applications of Transformer-based models. Learn More and Register.

  • COMPLECS: Intermediate Linux - 07/10/25 - 2:00 PM - 3:30 PM EDT Knowledge of Linux is indispensable for using advanced CI. While GUIs are becoming more prevalent, being able to work at the command line interface (CLI) provides the greatest power and flexibility. In this session, we assume that participants are already comfortable with basic Linux operations such as creating, deleting and renaming files, and navigating between directories. Topics covered include the filesystem hierarchy, file permissions, symbolic and hard links, wildcards and file globbing, finding commands and files, environment variables and modules, configuration files, aliases, and history. Learn More and Register.

  • Codee for Beginners: Automatic Code Optimization with Codee - July 29, 2025, 9 - 10:30 a.m. PDT This is an introductory webinar showing how Codee’s AutoFix feature can automatically accelerate computational kernels, representing performance hotspots, on both CPUs and GPUs. With AutoFix, developers can simply instruct Codee to insert OpenMP, OpenACC, and compiler-specific directives, as well as language-specific constructs (e.g., Fortran’s “do concurrent”) to vectorize, parallelize, and offload compute-intensive loops. AutoFix can even combine optimization techniques, such as multithreading and vectorization for nested loops, or OpenACC alongside OpenMP to maximize compatibility, allowing even novice programmers to write expert-level parallel code. Learn More and Register.

  • NUG Community Call: A Birds-Eye View of Using Cuda with C/C++ on Perlmutter (Part 1) - July 30, 2025, 11 a.m. - 12:30 p.m. PDT NERSC will be hosting a 2-part event series that focuses on using learning introductory GPU Programming concepts with CUDA on Perlmutter at NERSC. In this two-part training series, users will be introduced to the basics of using CUDA on Perlmutter at NERSC. The training will focus on the basics of the Perlmutter architecture and NVIDIA GPUs, programming concepts with CUDA using C/C++. This training is also open to non-NERSC users. Learn More and Register.

  • Accelerating and Scaling Python for HPC - August 8, 2025, 9 a.m. - 5 p.m. PDT In this interactive tutorial you’ll learn how to write, debug, profile, and optimize high-performance, multi-node GPU applications in Python. You'll learn and master: CuPy for drop-in GPU acceleration of NumPy workflows; Nvmath-python for high level api for integrating Python with NVIDIA math libraries; Numba for writing custom kernels that match the performance of C++ and Fortran; and mpi4py for scaling across thousands of nodes. Along the way we’ll learn how to profile our code, debug tricky kernels, and leverage foundational and domain-specific accelerated libraries. Learn More and Register.

  • Automating Research with Globus: The Modern Research IT Platform - Aug. 18, 2025, 9 a.m. – 12 p.m. (Pacific Time) This workshop introduces Globus Flows and its role in automating research workflows. Participants will explore data portals, science gateways, and commons, enabling seamless data discovery and access. Enroll here.

Questions about Hyak Klone, Tillicum, or any other UW-IT Research Computing Service? Fill out our Research Computing Consulting intake form. We are here to help!

Happy Computing,

Hyak Team

June 2025 Maintenance Update

· 6 min read
Kristen Finch
Kristen Finch
Director of Research Computing Solutions

During June's maintenance, we've refreshed the operating system images for both login and compute nodes, and we've responded to user feedback with a solution to make Cron jobs persistent. Good news: we are holding office hours all summer to support your research grind. Stay informed by subscribing to our mailing list and the UW-IT Research Computing Events Calendar. The next maintenance is scheduled for Tuesday July 8, 2025 (AKA the 2nd Tuesday of the month).

Notable Updates​

  • Routine package updates - images for both the login and compute nodes have been refreshed to incorporate the latest Linux OS security updates and system patches.
  • Slurm database lock timeout settings adjusted to match documentation best practices.
  • Cron job system improvements – our users provided feedback that their Cron jobs were being lost after monthly maintenance. We resolve this:
    • User crontabs moved to GPFS (/gscratch) for persistence across maintenance
    • Only one login node will now run user cron jobs (preventing duplication)
    • Users will need to re-create their crontabs one more time after this maintenance
    • This is intended to be a permanent fix—no more resets in future maintenance
    • FYI - Cron jobs are recurring scheduled tasks run by the system using each user's crontab.
    • We recommend scrontab for routine operations. Learn more from Slurm. Learn more from NERSC.

Action Required: Research Computing Club (stf account) Members​

To keep your access to RCC-supported Hyak accounts, please fill out the following form by Friday, June 13, 2025:

2025 RCC Usage Check-In Form

This short form is required for Student Technology Fee reporting and ensures the RCC can continue offering free computing resources to UW students. It only takes a few minutes, just tell us how you’ve used RCC resources this past year. Thanks for helping us keep RCC resources funded and accessible!

Spotlight: Kopah Object Storage​

Our Kopah S3-compatible storage service is available to all campus researchers and staff. It’s a flexible, scalable storage solution to complement your research computing portfolio.

If you missed our recent Data Storage Day on May 5, we’ve published the full set of demonstration videos on our YouTube Playlist. Topics covered and relevant links:

Whether you're just getting started or looking to expand your use of campus storage resources, this is a great place to learn more.

Summer Office Hours​

  • Wednesdays at 2pm on Zoom. Attendees need only register once and can attend any of the occurrences with the Zoom link that will arrive via email. Click here to Register for Zoom Office Hours.
  • Thursdays at 2pm in person in eScience. (address: WRF Data Science Studio, UW Physics/Astronomy Tower, 6th Floor, 3910 15th Ave NE, Seattle, WA 98195).
  • See our office hours schedule, subscribe to event updates, and bookmark our UW-IT Research Computing Events Calendar.

If you would like to request 1 on 1 help, please send an email to help@uw.edu with "Hyak Office Hour" in the subject line to coordinate a meeting.

Opportunities​

Computing Training from eScience and more​

eScience logo.

Introduction to Text Mining - Friday, June 27, 2025, 2 – 3 p.m. Open Scholarship Commons Presentation Space, Discover the power of text mining in this interactive workshop, where you will learn techniques for collecting, cleaning, and analyzing textual data through hands-on exercises using Python and Jupyter Notebooks. In this session, you will: Learn methods for scraping and extracting text from web sources.

  • Gain skills in preprocessing textual data, such as removing HTML tags, tokenization, and handling stop words.
  • Explore techniques for visualizing and analyzing word frequencies to uncover hidden themes and trends.
  • Use deep learning models to identify relevant text by analyzing semantic similarity.
  • Register here

External Training Opportunities​

  • COMPLECS: Code Migration - Thursday, June 12, 2025 - 11:00 a.m. – 12:30 p.m. (Pacific Time) We will cover typical approaches to moving your computations to HPC resources: using applications/software packages already available on the system through Linux environment modules; compiling code from source with information on compilers, libraries, and optimization flags to use; setting up Python and R environments; using conda-based environments; managing workflows; and using containerized solutions via Singularity. Register here!
  • Automating Research with Globus: The Modern Research IT Platform - Aug. 18, 2025, 9 a.m. – 12 p.m. (Pacific Time) This workshop introduces Globus Flows and its role in automating research workflows. Participants will explore data portals, science gateways, and commons, enabling seamless data discovery and access. Enroll here.
  • HPC Fundamentals: June 11, 9 am – 4 pm PDT & June 12, 9 am – 12 pm PDT - This 1.5-day hybrid training, provided in collaboration with HPC Carpentries, is for novice HPC users to learn the basic skills they will need to start using an HPC resource. Capacity is limited to 40 learners; application and registration are required. Register here.
  • OLCF Julia for Science: June 19, 10 am – 1 pm PDT; also June 26, 10 am – 1 pm PDT - The Oak Ridge Leadership Computing Facility (OLCF), in conjunction with the Oak Ridge National Laboratory Computer Science and Mathematics Division (CSMD), will host Julia for Science, a 3-hour tutorial focused on introductory aspects of the Julia programming language, and ecosystem for computation and data analysis. This training provides a hands-on way to learn more about using Julia and parallel code in scientific computing. Register here.
  • Crash Course in Supercomputing: June 23, 9 am – 4 pm PDT - In this course, students will learn to write parallel programs that can be run on a supercomputer. We begin by discussing the concepts of parallelization before introducing MPI and OpenMP, the two leading parallel programming libraries. Finally, the students will put together all the concepts from the class by programming, compiling, and running a parallel code on one of the NERSC supercomputers. Training accounts will be provided for students who have not yet set up a NERSC account. This hybrid training, as part of the 2025 Berkeley Lab Computational Sciences Summer Student Program, is also open to NERSC, ALCF, LANL, OLCF, and TACC users. This training is geared towards novice parallel programmers. Register here.

If you have any questions about using Hyak, please start a help request by emailing help@uw.edu with "Hyak" in the subject line.

Happy Computing,

Hyak Team

May 2025 Maintenance Update

· 7 min read
Kristen Finch
Kristen Finch
Director of Research Computing Solutions

During May's maintenance, we've refreshed the operating system images for both login and compute nodes with the latest Linux security updates and patches, and enhanced the node image to include NVIDIA Fabric Manager and software upgrades for next-generation GPU switching fabrics. There’s still time to register for GPU in EDU on Thursday, May 15th, 10:00 a.m.–4:00 p.m., where we team up with NVIDIA and Cambridge Computer for a day of demos, teaching guidance, and research highlights. Stay informed by subscribing to our mailing list and the UW-IT Research Computing Events Calendar. The next maintenance is scheduled for Tuesday June 10, 2025 (AKA the 2nd Tuesday of the month).

Notable Updates​

  • Operating system - The images for both the login and compute nodes have been refreshed to incorporate the latest Linux OS security updates and system patches.
  • Node image enhancements - This version of the node image includes NVIDIA Fabric Manager and software upgrades necessary to support next-generation GPU switching fabrics.

Upcoming Events​

Subscribe to event updates and bookmark our UW-IT Research Computing Events Calendar.

  • There's still time to sign up for GPU in EDU Thursday May 15th 10:00 a.m. - 4:00 p.m.
    • Join us for a full day of learning about GPUs with experts from NVIDIA and Cambridge Computer. The event will feature recommendations for building GPU workflows, guidance for using GPUs for teaching, highlights of GPU-powered research at UW, and more. Don’t miss it — Lunch is provided! Registration is still open!

Spring Office Hours​

If you would like to request 1 on 1 help, please send an email to help@uw.edu with "Hyak Office Hour" in the subject line to coordinate a meeting.

Training Resources​

Opportunities​

Computing Training from eScience and more​

eScience logo.
  • Intro Programming Workshop - eScience is holding a Software Carpentry workshop on May 27th–30th (9:00 a.m. – noon each day). The workshop will teach software tools that can make researchers more effective, automate research tasks, and track research over time. Specifically, the Unix Shell, Git, and Python will be taught with a focus on reproducible research. Register here.
  • Teach your own LLM - On Friday, May 23rd from 10:30 - noon in the Open Scholarship Commons, Jose Cols will lead the workshop “Teach your own LLM: Fine-tuning Models on Custom Datasets” covering how LLMs work and how to fine-tune a Llama 3 model for tasks like sentiment analysis and summarization.

External Training Opportunities​

  • COMPLECS: Using Regular Expressions with Linux Tools - Thursday, May 29, 2025 - 11:00 a.m. – 12:30 p.m. (Pacific Time) Regular expressions (regexes) provide a way to identify strings that match a specified pattern. They are extremely useful for preprocessing text and extracting results from high-performance computing and data science workloads. Primarily in the context of the Linux grep utility, we incrementally introduce the main features of regexes: string literals, specifying multiple characters, quantifiers, wildcards, anchors, character classes, grouping, and alternation. We also explore more advanced topics such as word boundaries, lazy and greedy matching, regex flavors (basic, extended, and Perl compatible), regexes with awk and sed, searching compressed files, and using large language models (LLMs) to create regexes. Register here!
  • 8:00 - 11:30 am (Pacific time), Wednesday, May 28, 2025 - LANL is hosting an Accelerated Python Tutorial, presented by Scot Halverson from Nvidia. This will include measuring, understanding, and improving performance of their python applications, including ML workflows using PyTorch and Tensorflow. This event is open to NERSC users. Learn more and register.
  • Building GPU-Accelerated Differentiable Simulations with NVIDIA Warp Python 1- 4 pm (Pacific time), Wednesday, May 28, 2025 - LANL is hosting a Building GPU-Accelerated Differentiable Simulations with NVIDIA Warp Python training, presented by Eric Shi from Nvidia. This approach lets developers harness GPU performance while maintaining the simplicity and flexibility of Python. This event is open to NERSC users. Learn more and register.
  • NERSC GPU Hackathon - NERSC, in conjunction with NVIDIA and the OpenACC organization, will be hosting an Open Hackathon from July 16th-18th with an opening day on July 9th as part of the annual Open Hackathon Series. Please note the deadline to submit a proposal is 11:59 PM Pacific, May 28, 2025. So apply now! Learn more.
  • A Deep Dive into the HPC SDK - 1:00 - 4:00 pm (Pacific time), Thursday, May 29, 2025 - LANL is hosting a Deep Dive into the Nvidia HPC SDK Training, presented by Scot Halverson from Nvidia. This talk will cover the broad set of compilers, tools, and libraries that make up the NVIDIA HPC SDK. This event is open to NERSC users. Learn more and register.
  • High Throughput Workflow Tools and Strategies - 8:30 am - 12:00 pm (Pacific time), Friday, May 30, 2025 - NERSC is hosting an online webinar presented by William Arndt of NERSC and Geoffrey Lentner from Advanced Computing, Purdue University. This training session will discuss and demonstrate multiple software tools for managing high throughput workloads: GNU Parallel, Snakemake, and Hypershell. The seminar is open to the general public. Learn more and register.
  • Solving Data Management Challenges with Globus - June 6, 2025, 9 a.m. – 12 p.m. (Pacific Time) Participants will engage in hands-on exercises to explore how Globus can streamline data movement across cloud and high-performance computing systems. Whether managing large datasets, enabling secure collaboration, or automating workflows, this session will equip participants with the knowledge and skills to maximize the benefits of Globus. Enroll here.
  • COMPLECS: Code Migration - Thursday, June 12, 2025 - 11:00 a.m. – 12:30 p.m. (Pacific Time) We will cover typical approaches to moving your computations to HPC resources: using applications/software packages already available on the system through Linux environment modules; compiling code from source with information on compilers, libraries, and optimization flags to use; setting up Python and R environments; using conda-based environments; managing workflows; and using containerized solutions via Singularity. Register here!
  • Automating Research with Globus: The Modern Research IT Platform - Aug. 18, 2025, 9 a.m. – 12 p.m. (Pacific Time) This workshop introduces Globus Flows and its role in automating research workflows. Participants will explore data portals, science gateways, and commons, enabling seamless data discovery and access. Enroll here.

If you have any questions about using Hyak, please start a help request by emailing help@uw.edu with "Hyak" in the subject line.

Happy Computing,

Hyak Team

April 2025 Maintenance Update

· 10 min read
Kristen Finch
Kristen Finch
Director of Research Computing Solutions

April maintenance included the latest OS patches and security enhancements for login and compute nodes, along with testing endpoint detection and response (EDR) software to strengthen cluster security. We would like to bring your attention to upcoming events featuring an AWS AI in Research Workshop (April 17), our annual Research Computing Club Hackathon (April 19), and the Kopah S3 Storage Launch Event "Data Storage Day" (May 5), showcasing campus-wide S3-compatible storage. Additionally, GPU in EDU (May 15) will provide insights into GPU workflows demonstrated by experts at Cambridge Computer and NVIDIA. Regular office hours are available for research computing support. Stay informed through training resources, event subscriptions, and the UW-IT Research Computing Events Calendar. The next maintenance is scheduled for Tuesday May 13, 2025 (AKA the 2nd Tuesday of the month).

Notable Updates, New Features​

  • Operating system - The login and compute node images were updated to address system patches and security updates in the Linux operating system (OS).
  • Security solutions testing - We tested endpoint detection and response (EDR) software as a potential solution for an enhanced cluster security posture. This testing is part of our ongoing efforts to maintain a secure computing environment in line with funding requirements (e.g., NIST-800, CMMC, HIPAA, NIH).
  • Globus - We’re excited to announce that Globus has been added to Hyak Klone and Kopah S3. Globus makes it easy to transfer large datasets reliably and securely between systems, whether across campus or around the world. With features like automated transfers, fault tolerance, and a simple web interface, it’s a powerful tool for streamlining data movement in research workflows.

Upcoming Training and Events​

Subscribe to event updates and bookmark our UW-IT Research Computing Events Calendar.

  • AWS AI in Research Workshop - UW-IT Research Computing and eScience Institute will host a workshop introducing how AWS enables AI in research on Thursday, April 17th from 12:00 - 4:00 p.m. in the eScience Institute. Whether you are developing your own models, building off existing models, or you need to implement generative AI, AWS has you covered. Bring your laptop and join us for a couple of engaging sessions followed by a hands-on lab building AI agents. Learn more.

  • RCC Hackathon - The Research Computing Club (RCC), in collaboration with UW-IT Research Computing and the eScience Institute, is excited to host an all-day HPC Hackathon on Saturday, April 19th from 10:00 a.m. – 6:00 p.m. in the eScience Institute. Learn more.

  • Discover Kopah: Your New S3-Compatible Campus Storage Solution at Data Storage Day! Join us on Monday, May 5 from 1:00 – 5:00 p.m. in the eScience Institute for the official Kopah S3 Storage Service Launch Event — your gateway to fast, scalable storage right on campus. Whether you're a researcher handling big data, an instructor managing course materials, or simply someone in need of reliable cloud-like storage, Kopah is designed for everyone. This event will feature live demos of S3-compatible tools like s3cmd, Globus, JuiceFS, Cyberduck, and boto3, and staff will be available to help you get started with Kopah S3. Learn more.

  • GPU in EDU Save the Date Thursday May 15th 10:00 a.m. - 4:00 p.m. – Join UW-IT Research Computing for a day learning about GPUs from the experts at NVIDIA and Cambridge Computer. This event will feature live demonstrations for building your GPU workflows, recommendations for incorporating GPUs into instruction, computing research at UW, and more! Registration link coming soon!

Spring Office Hours​

If you would like to request 1 on 1 help, please send an email to help@uw.edu with "Hyak Office Hour" in the subject line to coordinate a meeting.

Training Resources​

Opportunities​

External Training Opportunities​

  • Parallel Computing with MATLAB: Hands on workshop Coming Up April 16 1PM PDT - During this hands-on workshop, we will introduce parallel and distributed computing in MATLAB with a focus on speeding up application codes and offloading computers. By working through common scenarios and workflows using hands-on demos, you will gain a detailed understanding of the parallel constructs in MATLAB, their capabilities, and some of the common hurdles that you'll encounter when using them. SIGN UP! Highlights:

    • Multithreading vs multiprocessing
    • When to use parfor vs parfeval constructs
    • Creating data queues for data transfer
    • Leveraging NVIDIA GPUs
    • Parallelizing Simulink models
    • Working with large data

  • AI in Science Postdoc Workshop - Generative AI systems built upon large language models (LLMs) show great promise as tools that enable people to access information through natural conversation. Scientists can benefit from the breakthroughs these systems enable to create advanced tools that will help accelerate their research outcomes. Last week the UW Scientific Software Engineering Center (SSEC) offered a Generative AI / RAG training opportunity to the Schmidt AI in Science Postdoctoral Fellows. The online tutorial focused on how to utilize the underlying methods in Generative AI to advance scientific research, including the basics of LLMs followed by a demo of using LLMs and RAG for creating a question answering tool based on private data. Learn more.

  • Trainings from Texas A&M University - gain skills that are transferrable to Hyak (Linux, Slurm job scheduler, Open OnDemand, NVIDIA GPU).

    • ACES: Python for Data Science 04/11/25 - 10:00 AM - 04:00 PM CDT - This short course for experienced programmers introduces the Numpy, Pandas, and Matplotlib libraries commonly used to manage and display large datasets in Python. Exercises will be performed in the learner's web browser in Jupyter Notebooks running in the Open OnDemand portal of the ACES cluster. More information about this course at https://hprc.tamu.edu/training/aces_python4data.html.
    • ACES: Metagenomics 04/15/25 - 01:30 PM - 04:00 PM CDT - This short course introduces concepts of metagenomic analyses based on data generated by Next Generation Sequencing (NGS) technology using ACES cluster, a composable accelerator testbed at Texas A&M University. Students will learn to complete a metagenomics pipeline using QIIME2 software. Read more at https://hprc.tamu.edu/training/aces_metagenomics.html
    • ACES: Rust 04/22/25 - 10:00 AM - 12:30 PM CDT - This short course provides an introduction to Rust, a modern systems programming language known for its safety, speed, and concurrency, through hands-on coding exercises and practical examples using the ACES cluster, a composable accelerator testbed at Texas A&M University. Learn more at https://hprc.tamu.edu/training/aces_rust.html
    • ACES: GPU Programming (CUDA) 04/22/25 - 01:30 PM - 04:00 PM CDT - This short course covers basic topics in CUDA programming on NVIDIA GPUs. More information about this Short Course https://hprc.tamu.edu/training/intro_cuda.html Topics include:
      • CUDA architecture
      • basic language usage of CUDA C/C++
      • writing, executing CUDA code.
    • ACES: Python for HPC and Advanced Topics 04/25/25 - 10:00 AM - 04:00 PM CDT - This short course for experienced Python programmers will cover several topics relevant to Python workloads running on HPC systems, including environment and data handling. Exercises will be performed in the learner's web browser in Jupyter Notebooks running in the Open OnDemand portal of the ACES cluster. More information about this course at https://hprc.tamu.edu/training/aces_python4HPC.html.

  • Trainings from the National Energy Research Scientific Computing Center (NERSC) - gain skills that are transferrable to Hyak

    • NERSC N-Ways to GPU Programming Bootcamp - NERSC, in collaboration with the OpenACC Organization and NVIDIA, is hosting an N-Ways to GPU Programming Bootcamp for 3 days from Wednesday, April 30 to Friday, May 2. Beginner users in GPU programming are especially encouraged to attend. This is a virtual event. For detailed information, and how to apply, please refer to Open Hackathons's Bootcamp Events page. The application deadline is April 13, 2025.
    • Building GPU-Accelerated Differentiable Simulations with NVIDIA Warp Python 1- 4 pm (Pacific time), Wednesday, May 28, 2025 - LANL is hosting a Building GPU-Accelerated Differentiable Simulations with NVIDIA Warp Python training, presented by Eric Shi from Nvidia. Warp is a Python framework designed for authoring high-performance, GPU-accelerated code directly in Python. At its core, Warp uses a programming model where Python functions are just-in-time (JIT) compiled into efficient code that can run on both CPUs and NVIDIA GPUs, using C++/CUDA as an intermediate representation. This approach lets developers harness GPU performance while maintaining the simplicity and flexibility of Python. Learn more and register.
    • NERSC GPU Hackathon - NERSC, in conjunction with NVIDIA and the OpenACC organization, will be hosting an Open Hackathon from July 16th-18th with an opening day on July 9th as part of the annual Open Hackathon Series. The NERSC Open Hackathon will be hosted as a hybrid event. Days 0-1 (July 2 and 9) will be held virtually, and attendees will have the option to attend Days 2-4 (July 16 - 18) either in-person at NERSC or online. This hackathon is open to everyone looking to take their projects to the next level; however, priority acceptance will be given to NERSC collaborators. Please note the deadline to submit a proposal is 11:59 PM Pacific, May 28, 2025. So apply now! Please note acceptance is not confirmed until you have received a confirmation email. Learn more.
    • More opportunities from NERSC

Jobs​

  • The UW Echospace research group is seeking a student or an early career professional to help develop open-source scientific software for processing large-scale ocean sonar data to accelerate marine ecosystem research. APPLY!

  • Metascience & AI Postdoctoral Fellowship - The Sloan Foundation is awarding grants of up to $250k to support early career researchers in the social sciences and humanities who are interested in understanding the implications of AI for the science and research ecosystem. Applicants are required to have a faculty mentor based at the research organization where the grant is to be held. Feel free to reach out to eScience for support with identifying a potential UW mentor. Applications due April 10, 2025! Learn more.

  • The Center for Geospatial Intelligence at George Mason University is looking for a Postdoctoral Research Fellow in geospatial data science and machine learning to work on research projects related to large-scale urban simulations and geospatial machine learning. APPLY!

  • Yale Center for Research Computing is hiring an HPC System Administrator to join the center’s Engineering team to provide hardware and software administration for a growing number of high performance computing (HPC) clusters used in faculty research. This position will support the infrastructure behind all of the above, including hardware, system and resource-management software, networking, storage, monitoring and security measures. This is a highly-collaborative effort, so frequent interaction with other system administrators, research support staff, management, vendors and researchers is a regular part of the role. Learn more.

  • Job opening for a High Performance Computing (HPC) Research Computing Associate at Colby College in Maine. Learn more.

If you have any questions about using Hyak, please start a help request by emailing help@uw.edu with "Hyak" in the subject line.

Happy Computing,

Hyak Team

March 2025 Maintenance Details

· 5 min read
Kristen Finch
Kristen Finch
Director of Research Computing Solutions

March maintenance is complete, and Klone is back in operation. Notable updates from this maintenance include new features and documentation (VS Code via Open OnDemand), upcoming events (SAVE THE DATE for Hackathon and GPU Day), newly-released training videos (Advanced Slurm), office hours, and opportunities from the national computing community! The next maintenance is scheduled for Tuesday April 8, 2025 (AKA the 2nd Tuesday of the month).

Notable Updates​

  • Operating system - The login and compute node images were updated to address system patches and security updates in the Linux operating system (OS).
  • Hardware - The HPC engineering team used this scheduled time to perform what would have otherwise been disruptive tasks by physically re-arranging equipment within the data center in preparation for a future (GPU) cluster launch. A new rack was also prepared for a bulk expansion of HPC compute node capacity.

New Features and Documentation​

  • Our GPU documentation section has an additional worked example for preparing a software container to utilize Hyak GPUs. Check out the new content.
  • VS Code has been added as an interactive application via Hyak’s Open OnDemand Platform. Go to Open OnDemand(VPN required if off campus) and view the supporting documentation.
  • Have you ever wanted to contribute documentation or tutorials to support your fellow Hyak users? By preparing a GitHub CodeSpace and containerized development environment, we have made contributing to Hyak’s documentation website and repository more accessible. We invite and appreciate your feedback and contributions, no matter how small. Check out our guide for using the development container and submitting your contributions via a Pull Request.

Upcoming Training and Events​

  • The Research Computing Club (RCC) is hosting a Hackathon on Saturday April 19 10am-5pm. The RCC officers have prepared a set of modules for participants to work through together in small groups. Lunch will be provided. Registration is required. Additional details coming soon.
  • GPU Day Thursday May 15 SAVE THE DATE! hosted by UW-IT Research Computing featuring hands-on demonstrations from Cambridge Computer and NVIDIA. Time and venue TBA. Stay tuned for more details and links to register.

New training videos uploaded​

Hyak: Advanced Slurm recording from training on February 25. This tutorial demonstrates these benefits of Slurm and provide you with some template Slurm scripts that you can adapt for your purposes. Click here for walk through tutorial and training materials.

Check out our Research Computing Training Playlist on UW-IT's YouTube channel.

Winter 2025 Office Hours​

If you would like to request 1 on 1 help, please send an email to help@uw.edu with "Hyak Office Hour" in the subject line to coordinate a meeting.

Opportunities​

  • We're Hiring! The Hyak team is hiring an HPC Staff Scientist to join our team. Use your experience to help Hyak users build capacity especially as we launch out new GPU system this fall. Job description HERE.

  • Join the UW AI Community of Practice on MS Teams to get updates from UW-IT's AI team about events and join the discussion around AI in the news, society, and culture.

  • The Accelerated AI Algorithms for Data-Driven Discovery (A3D3) Institute funded by the National Science Foundation (NSF), under the Harnessing the Data Revolution (HDR) program, is seeking postbaccalaureate research fellows to join our interdisciplinary teams of scientists and engineers to develop and deploy artificial intelligence to accelerate science discoveries in particle physics, astrophysics, biology, and neuroscience. APPLY!

  • The Alfred P. Sloan Foundation is providing a postdoctoral fellowship program to support early career researchers in the social sciences and humanities who are interested in building a career in understanding the implications of AI for the science and research ecosystem. APPLY!

  • eScience is seeking current and incoming UW postdocs who are actively involved in developing or utilizing advanced data science/AI tools and techniques in their research at the UW. Applications are due Saturday, March 15th. APPLY!

  • Tania Malik, a lecturer at the School of Informatics and Cybersecurity, and a director of HPC Nexus lab at TU Dublin, Ireland is looking to support one postdoc under the SUSTAIN-FIT Horizon Europe Programme. If you are passionate about Energy-Efficient High-Performance Computing (HPC) or Green Computing for modern HPC platforms feel free to contact tania.malik@tudublin.ie with your CV and a brief research proposal.

If you have any questions about using Hyak, please start a help request by emailing help@uw.edu with "Hyak" in the subject line.

Happy Computing,

Hyak Team

February 2025 Maintenance Details

· 5 min read
Kristen Finch
Kristen Finch
HPC Staff Scientist

Our February maintenance is complete, and Klone is back in operation. The next maintenance is scheduled for Tuesday March 11, 2025 (AKA the 2nd Tuesday of the month). Below we discuss notable updates from this maintenance session, upcoming user training, newly-released training videos, office hours, and opportunities from the national computing community!

Notable Updates​

  • Security updates for firmware on all nodes.
  • Software updates for security on the operating system.
  • Increased resiliency of the job scheduler.

Upcoming Training​

Hyak: Advanced Slurm - This workshop will be held in person on Wednesday February 26 10am - 11:45am in the eScience Classroom (Address: WRF Data Science Studio, UW Physics/Astronomy Tower, 6th Floor; 3910 15th Ave NE, Seattle, WA 98195). For this tutorial, we'll use this container and a publicly available dataset to practice submitting single and array jobs to Hyak's job scheduler, Slurm. Locator is set of python tools that builds a neural network with TensorFlow to estimate the location of organisms based on their genotype (DNA; or genetic background). The neural network is trained on genotypes from a set of organisms with known location. Performance is assessed with “unknown origin” genotypes by calculating the distance between their predicted and true location. Click here to learn more and register for this event.

New training videos uploaded​

Hyak: Intro to Deep Learning recording from training on December 3. Bernease Herman from the eScience Institute introduces deep learning concepts including data representations, data importance, learning tasks, deep learning frameworks, and computing environments. This training includes a hands-on demonstration of fully connected neural networks and convolutional neural networks with MNIST digit data and a brief discussion about deep learning frameworks on Hyak.

Hyak: Open OnDemand recording from training on January 31. Open OnDemand (OOD) is an open-source web portal for HPC centers to provide users with an easy-to-use web interface to HPC clusters. For the last year, the Hyak team has been adding features to OOD. This workshop demonstrates OOD's main features such as exploring the filesystem, composing jobs, and launching interactive applications like Jupyter, Rstudio, MATLAB, and Virtual Desktop. Training presented by UW Research Computing Intern, Bhavik Soni.

Check out our Research Computing Training Playlist on UW-IT's YouTube channel.

Winter 2025 Office Hours​

If you would like to request 1 on 1 help, please send an email to help@uw.edu with "Hyak Office Hour" in the subject line to coordinate a meeting.

Opportunities​

  • COMPLECS: Linux Tools for Text Processing - Many computational and data processing workloads require pre-processing of input files to get the data into a format that is compatible with the user’s application and/or post-processing of output files to extract key results for further analysis. While these operations could be done by hand, they tend to be time-consuming, tedious and, worst of all, error prone. In this session we cover the Linux tools awk, sed, grep, sort, head, tail, cut, paste, cat and split, which will help users to easily automate repetitive tasks. We conclude by showing how large language models (LLMs) such as ChatGPT could be used to write commands using these tools. Register for this remote training.

  • The Trusted CI Student Program offers undergraduate and graduate students invaluable hands-on experience, mentorship, and opportunities to contribute to a secure and resilient digital future. As cybersecurity becomes an essential component of the scientific community, this program empowers students to gain the skills and knowledge needed to thrive in the field. The Trusted CI Student Program offers a unique opportunity for students eager to advance in the cybersecurity field. Applications for the 2025 cohort open on February 3, 2025. Click here for more details on how to apply and program information.

  • Google Cloud Skills Boost seats are available to the UW community. This platform offers online courses, labs, and certifications to help learners develop cloud skills. If you are interested in using this excellent learning resource, please request access via email to Rob Fatland at rob5@uw.edu.

  • Join the Bridge to Ocean Acoustics and Technology (BOAT) workshop and learn ocean acoustics from theory to data practice with Jupyter notebooks! The workshop is organized as part of the BOAT program which aims to grow the ocean acoustics education and research community through open-source curriculum and is funded by the ONR. Application deadline Thursday, February 20th. Apply HERE!

  • The Nawaz Lab at the UW-Applied Physics Laboratory is currently looking for a student to help apply machine learning to long term sensor data with the goal of correcting for drift. Contact Anuscheh Nawaz at anuscheh@uw.edu for more details.

  • The Microsoft AI for Good Open Call supports individuals and organizations creating positive change in Washington State Program, providing opportunities for scientific collaboration with the Microsoft AI for Good Lab. Application deadline: Monday, February 17th. APPLY

  • Text Mining Student Assistant - This student position, jointly supported by the Open Scholarship Commons and eScience, will provide technical expertise on text mining and natural language processing (NLP) to UW researchers at all levels. Apply NOW

If you have any questions about using Hyak, please start a help request by emailing help@uw.edu with "Hyak" in the subject line.

Happy Computing,

Hyak Team

JuiceFS or using Kopah on Klone

· 7 min read
Nam Pho
Nam Pho
Research Computing

If you haven't heard, we recently launched an on-campus S3-compatible object storage service called Kopah docs that is available to the research community at the University of Washington. Kopah is built on top of Ceph and is designed to be a low-cost, high-performance storage solution for data-intensive research.

warning

From our testing we have observed significant performance challenges for JuiceFS in "single mode" which is demonstrated by this blogpost. We do not recommend JuiceFS as a solution for demanding workflows.

While the deployment of Kopah was welcome news to those who are comfortable working with S3-compatible cloud solutions, we recognize some folks may be hesitant to give up their familiarity with POSIX file systems. If that sounds like you, we explored the use of JuiceFS, a distributed file system that provides a POSIX interface on top of object storage, as a potential solution.

info

Simplistically, object storage often presents using two API keys and data is accessed using a command line tool that wraps API calls, whereas POSIX is what you typically get presented with from the storage when interacting with a cluster via command-line.

Installation​

JuiceFS isn't installed by default so you will need to compile it yourself or download the pre-compiled binary from their release page.

As of January 2025 the latest version is 1.2.3 and you want the amd64 version if using from Klone. The command below will download and extract the binary to your current working directory.

wget https://github.com/juicedata/juicefs/releases/download/v1.2.3/juicefs-1.2.3-linux-amd64.tar.gz -O - | tar xzvf -

I have to move it to a folder in my $PATH so I can run it from anywhere by just calling the binary. Your personal environment varies here.

mv -v juicefs ~/bin/

Verify you can run JuiceFS.

npho@klone-login03:~ $ juicefs --version
juicefs version 1.2.3+2025-01-22.4f2aba8
npho@klone-login03:~ $

Cool, now we can start using JuiceFS!

Standalone Mode​

There are two ways to run JuiceFS, standalone or distributed mode. This blog post explores the former. Standalone mode is meant to only present Kopah via POSIX on Klone. The key points being:

  1. There is an active juicefs process required to run while you want to access it.
  2. It is intended for you to run it only on the node you are running the process from.

If you wanted to run JuiceFS on multiple nodes or with multiple users then we will have another proof-of-concept with distributed mode in the future.

Create Filesystem​

JuiceFS separates the data (placed into S3 object storage) and the metadata, which is kept locally in a database. The command below will create the myjfs filesystem and store the metadata in a SQLite database called myjfs.db in the directory where the command is run. It puts the data itself into a Kopah bucket called npho-project.

juicefs format \
  --storage s3 \
  --bucket https://s3.kopah.uw.edu/npho-project \
  --access-key REDACTED \
  --secret-key REDACTED \
  sqlite3://myjfs.db myjfs

You can rename the metadata file and the filesystem name to whatever you want (they don't have to match). The same goes for the bucket name on Kopah. However, I would strongly recommend having unique metadata file names that match the file system names for ease of tracking alongside the bucket name itself.

npho@klone-login03:~ $ juicefs format \
>   --storage s3 \
>   --bucket https://s3.kopah.uw.edu/npho-project \
>   --access-key REDACTED \
>   --secret-key REDACTED \
>   sqlite3://myjfs.db myjfs
2025/01/31 11:52:47.940709 juicefs[1668088] <INFO>: Meta address: sqlite3://myjfs.db [interface.go:504]
2025/01/31 11:52:47.944930 juicefs[1668088] <INFO>: Data use s3://npho-project/myjfs/ [format.go:484]
2025/01/31 11:52:48.666657 juicefs[1668088] <INFO>: Volume is formatted as {
  "Name": "myjfs",
  "UUID": "eb47ec30-c1f7-4a92-9b17-23c4beae7f76",
  "Storage": "s3",
  "Bucket": "https://s3.kopah.uw.edu/npho-project",
  "AccessKey": "removed",
  "SecretKey": "removed",
  "BlockSize": 4096,
  "Compression": "none",
  "EncryptAlgo": "aes256gcm-rsa",
  "KeyEncrypted": true,
  "TrashDays": 1,
  "MetaVersion": 1,
  "MinClientVersion": "1.1.0-A",
  "DirStats": true,
  "EnableACL": false
} [format.go:521]
npho@klone-login03:~ $

You can verify there is now a myjfs.db file in your current working directory. It's a SQLite database file that will store your file system meta data.

We can also verify the npho-project bucket was created on Kopah to store the data itself.

npho@klone-login03:~ $ s3cmd -c ~/.s3cfg-default ls                                      
2025-01-31 19:48  s3://npho-project
npho@klone-login03:~ $

You should run juicefs format --help to view the full range of options and customize the parameters of your file system to your unique needs but just briefly:

  • Encryption: When you create the file system and format it you can see it has encryption by default using AES256. You can over ride this using the --encrypt-algo flag if you prefer chacha20-rsa or you can use key file based encryption and provide your private key using the --encrypt-rsa-key flag.
  • Compression: This is not enabled by default and there is a computational penalty for doing so if you want to access your files since it needs to be de or re encrypted on the fly.
  • Quota: By default there is no block (set with --capacity in GiB units) or inode (set with --inodes files) quota enforced at the file system level. If you do not explicitly set this, it will be matched to whatever you get from Kopah. This is still useful for setting explicitly if you wanted to have multiple projects or file systems in JuiceFS that use the same Kopah account and have some level of separation.
  • Trash: By default, files are not deleted immediately but moved to a trash folder similar to most desktop systems. This is set with the --trash-days flag and you can set it to 0 if you want files to be deleted immediately. The default here is 1 day after which the file is permanently deleted.

Mount Filesystem​

Running the command below will mount your newly created file system to the myproject folder in your home directory. It does not need to previously exist.

juicefs mount sqlite3://myjfs.db ~/myproject --background
warning

The SQLite database file is critical, do not lose it. You can move its location around afterwards but it contains all the meta data about your files.

This process occurs in the background.

warning

Where you mount your file system the first time is where it will be expected to be mounted going forward.

npho@klone-login03:~ $ juicefs mount sqlite3://myjfs.db ~/myproject --background
2025/01/31 11:57:01.652279 juicefs[1690855] <INFO>: Meta address: sqlite3://myjfs.db [interface.go:504]
2025/01/31 11:57:01.654920 juicefs[1690855] <INFO>: Data use s3://npho-project/myjfs/ [mount.go:629]
2025/01/31 11:57:02.156898 juicefs[1690855] <INFO>: OK, myjfs is ready at /mmfs1/home/npho/myproject [mount_unix.go:200]
npho@klone-login03:~ $

Use Filesystem​

Now with the file system mounted (at ~/myproject) you can use it like any other POSIX file system.

npho@klone-login03:~ $ cp -v LICENSE myproject 
'LICENSE' -> 'myproject/LICENSE'
npho@klone-login03:~ $ ls myproject 
LICENSE
npho@klone-login03:~ $

Remember, you won't be able to see it in the bucket because it is encrypted before being stored there.

Recover Deleted Files​

If you enabled the trash can option then you can recover files up until the permanent delete date.

First delete a file on the file system.

npho@klone-login03:~ $ cd myproject 
npho@klone-login03:myproject $ rm -v LICENSE 
removed 'LICENSE'
npho@klone-login03:myproject $

Verify the file is deleted. Go to recover it from the trash bin.

npho@klone-login03:myproject $ ls          
npho@klone-login03:myproject $ ls -alh          
total 23K
drwxrwxrwx  2 root root 4.0K Jan 31 12:54 .
drwx------ 48 npho all  8.0K Jan 31 13:08 ..
-r--------  1 npho all     0 Jan 31 11:57 .accesslog
-r--------  1 npho all  2.6K Jan 31 11:57 .config
-r--r--r--  1 npho all     0 Jan 31 11:57 .stats
dr-xr-xr-x  2 root root    0 Jan 31 11:57 .trash
npho@klone-login03:myproject $ ls .trash  
2025-01-31-20
npho@klone-login03:myproject $ ls .trash/2025-01-31-20 
1-2-LICENSE
npho@klone-login03:myproject $ cp -v .trash/2025-01-31-20/1-2-LICENSE LICENSE
'.trash/2025-01-31-20/1-2-LICENSE' -> 'LICENSE'
npho@klone-login03:myproject $ ls                     
LICENSE
npho@klone-login03:myproject $

As you can see, we can recover files that are tracked by their delete date. You would need to copy the file back out to recover it.

Unmount Filesystem​

When you are done using the file system you can unmount it with the command below.

npho@klone-login03:~ $ juicefs umount myproject
npho@klone-login03:~ $

Remember, the file system is only accessible in standalone mode so long as a juicefs process is running. Since we ran it in the background you will need to explicitly unmount it.

Questions?​

Hopefully you found this proof-of-concept useful. If you have any questions for us, please reach out to the team by emailing help@uw.edu with Hyak somewhere in the subject or body. Thanks!

January 2025 Maintenance Details

· 5 min read
Kristen Finch
Kristen Finch
HPC Staff Scientist

Our January maintenance is complete, and Klone is back in operation. The next maintenance is scheduled for Tuesday February 11, 2025 (AKA the 2nd Tuesday of the month).

Notable Updates​

  • Slurm update: We updated to version 24.11.0. If any user software was built against the older Slurm libraries (e.g., openmpi), then users may experience errors, and it may be necessary to rebuild their software against the newer Slurm libraries.
  • Compute node updates: The compute node OS images were updated to address any security patches in underlying core packages.
  • Mathematica module installed using the license maintained by the UW Physics Department. Learn how to launch Mathematica on Hyak.
  • MATLAB application has been added to Hyak Open OnDemand Beta. Learn how to launch MATLAB with Open OnDemand.

Upcoming Training​

Hyak: Open OnDemand – Open OnDemand (OOD) is an open-source web portal for HPC centers to provide users with an easy-to-use web interface to HPC clusters. For the last year, the Hyak team has been adding features to OOD. This workshop will demonstrate OOD's main features such as exploring the filesystem, composing jobs, and launching interactive applications like Jupyter. This workshop will be held in person 10am - 11:30am on Friday January 31, 2025 in CSE2 (Gates Center) Room 371 (3800 E Stevens Way NE, Seattle, WA 98195). Click here to learn more and register for this event.

Winter 2025 Office Hours​

If you would like to request 1 on 1 help, please send an email to help@uw.edu with "Hyak Office Hour" in the subject line to coordinate a meeting.

Opportunities​

The eScience Institute offers the annual Winter School to students and lecturers interested in developing basic skills and knowledge of the tools used in data science. Gaining literacy in topics such as Python, R, Jupyter, and reproducible environments can be beneficial beyond STEM, including areas like global or public health, public policy, social sciences, social work, international relations, and business management. Apply by January 24, 2025. Learn more!

Summer Internship Opportunity at Purdue - The Rosen Center for Advanced Computing (RCAC) is seeking students for Research Experience for Undergraduates (REU) paid internships for an 11-week onsite summer REU program. This program aims at developing the next generation workforce in advanced computing and cyberinfrastructure technologies. It offers students from diverse backgrounds the opportunity to gain the knowledge and skills necessary to build and support advanced research computing systems and scientific applications. As part of RCAC's decade long successful student apprentice program, the REU students will learn by doing, working on the National Science Foundation funded Anvil system in a team environment and mentored by cyberinfrastructure professionals. Open to undergraduate students from all backgrounds and undergraduate programs within and beyond Purdue. Each student will present their work to Purdue staff, faculty, students, collaborators, and researchers at the end of the program. Students may present at a national conference as part of the program. This onsite program at Purdue University in West Lafayette, Indiana runs from May through August. Learn more and apply NOW! Interviews start in January!

April 2-3, 2025, for AI Unlocked: Empowering Higher Education through Research and Discovery, a workshop designed for individuals across all disciplines and career stages in higher education. This workshop provides access to cutting-edge computational resources and expertise to researchers, students, faculty, and practitioners, especially those from Minority-Serving Institutions (MSIs). Hosted by the National Artificial Intelligence Research Resource (NAIRR) Pilot User Experience Working Group, this event introduces AI fundamentals, hands-on experience with pre-configured AI tasks, and guidance for tailoring AI models to your projects. Travel support is available for selected participants to the Westin Denver Downtown, and virtual attendance options are also offered. Apply now to advance your AI skills and collaboration in higher education. Apply by January 31, 2025. Acceptances will be notified by the end of February.

The University of Alaska Fairbanks (UAF) Alaska Center for Energy and Power (ACEP) summer internship is a 10-week program for students to gain hands-on research experience and skill development in the energy industry. Our program offers two internship strands: AUSI and REU. Regardless of strand, all interns will receive:

  • A specific research project with 1:1 mentorship from an ACEP researcher
  • Collaborative workspace at ACEP
  • Travel to and from Alaska
  • Field trips related to energy in Alaska

Applications due January 24, 2025.

If you have any questions about using Hyak, please start a help request by emailing help@uw.edu with "Hyak" in the subject line.

Happy Computing,

Hyak Team