User Updates

The /work filesystem is currently having issues, as of about 3PM CST. We are working on getting it restored.

Lonestar6 has a new queue, "gpu-a100-small" designed for users who do not need a node's entire 3 GPUs in the "gpu-a100" queue. The jobs in this queue consume 1/3 of the full resources of a full node. If your job uses only a single GPU and less than 70GB of memory, you should try running in the "gpu-a100-small" queue. The charge rate for the "gpu-a100-small" queue is 1.5 SUs per node hour. Please submit any questions you may have through the TACC Consulting System or feedback form. https://tacc.utexas.edu/portal/ticketshttps://tacc.utexas.edu/about/help/

CMake with Victor Eijkhout | December 1, 2023 register HERE: https://www.tacc.utexas.edu/use-tacc/training/2023/cmake/This training is virtual / webinar from 10AM CST - 12PM CST. CMake is a build system for packages -- applications and libraries -- written in C++, C, or Fortran. It is also the basis of an eco-system, making it possible for one package to find another, and incorporate it. This short course covers CMake in three, increasingly sophisticated stages: - Help! This software uses CMake! - Help! CMake needs other packages! - Help! I want other people to use my CMake package! -------------Per our policy, attendees must have an active TACC account to participate in training activities on TACC resources. To verify that your login information for an existing account is current and that your account is active, or to create a TACC account, please visit https://accounts.tacc.utexas.edu/ at least one week prior to the start of the training to ensure that TACC staff can review and approve your account request. *If you are creating a new TACC account, we recommend that you use your institution email address (e.g. name@utexas.edu) for prompt account approval and verification.

REGISTER for TACC trainings: https://www.tacc.utexas.edu/use-tacc/training/ *stay tuned for a CMake tutorial on 12/1 -- registration opening soon!Registration is limited and closes Friday 10/20:HPC @ TACC | November 1-2, 2023 (closes at capacity)Virtual only 10AM CST - 4PM CST In this hands-on tutorial, we will explore methods for running containers on HPC systems including advanced tasks like utilizing GPUs for computation and MPI for parallel jobs Docker to perform multi-stage and multi-architecture builds.  - Introduction and motivation to parallel computing, which includes an introductory lab on TACC resources. - Edit and compile code, submit jobs to the queue - MPI: beginner and intermediate - Basic concepts of the Message Passing Interface (MPI) library for large scale parallel programming - Interactive combination of lectures and code walk-throughs - HPC tools specifically designed for complex user environments (Lmod, Sanity Tool) - Tools for workflow management (ibrun, launcher, launcher-GPU, Pylauncher) - Tools for job monitoring and profiling (Remora, TACC-Stat, core_usage, amask) - Other tools per participant interest - MPI with containers  ________________ Machine Learning | November 6-8, 2023 (closes at capacity)Virtual only 10AM CST - 4PM CST TACC’s Machine Learning training is designed to give participants fundamental knowledge of machine learning (ML) and deep learning (DL) techniques as well as experience with hands-on tutorials running ML and DL workflows on HPC systems. Participants will learn the basics of how to start using the computing clusters at TACC for ML and DL analysis. They will also learn fundamentals and theory of ML and DL techniques, and explore example applications of how ML and DL techniques can be applied on TACC’s HPC clusters. This training is virtual on the following days: Nov. 6: Computing cluster and Python environment at TACC Nov. 7: Introduction to ML, supervised learning methods, and their applications Nov. 8: Introduction to Deep Learning and its applications ________________ Per our policy, attendees must have an active TACC account to participate in training activities on TACC resources. To verify that your login information for an existing account is current and that your account is active, or to create a TACC account, please visit https://accounts.tacc.utexas.edu/ at least one week prior to the start of the training to ensure that TACC staff can review and approve your account request. *If you are creating a new TACC account, we recommend that you use your institution email address (e.g. name@utexas.edu) for prompt account approval and verification. 

REGISTER for TACC trainings: https://www.tacc.utexas.edu/use-tacc/training/ *stay tuned for a CMake tutorial on 12/1 -- registration opening soon!Registration is limited and closes Friday 10/20:HPC @ TACC | November 1-2, 2023 (closes at capacity)Virtual only 10AM CST - 4PM CST In this hands-on tutorial, we will explore methods for running containers on HPC systems including advanced tasks like utilizing GPUs for computation and MPI for parallel jobs Docker to perform multi-stage and multi-architecture builds.  - Introduction and motivation to parallel computing, which includes an introductory lab on TACC resources. - Edit and compile code, submit jobs to the queue - MPI: beginner and intermediate - Basic concepts of the Message Passing Interface (MPI) library for large scale parallel programming - Interactive combination of lectures and code walk-throughs - HPC tools specifically designed for complex user environments (Lmod, Sanity Tool) - Tools for workflow management (ibrun, launcher, launcher-GPU, Pylauncher) - Tools for job monitoring and profiling (Remora, TACC-Stat, core_usage, amask) - Other tools per participant interest - MPI with containers  ________________ Machine Learning | November 6-8, 2023 (closes at capacity)Virtual only 10AM CST - 4PM CST TACC’s Machine Learning training is designed to give participants fundamental knowledge of machine learning (ML) and deep learning (DL) techniques as well as experience with hands-on tutorials running ML and DL workflows on HPC systems. Participants will learn the basics of how to start using the computing clusters at TACC for ML and DL analysis. They will also learn fundamentals and theory of ML and DL techniques, and explore example applications of how ML and DL techniques can be applied on TACC’s HPC clusters. This training is virtual on the following days: Nov. 6: Computing cluster and Python environment at TACC Nov. 7: Introduction to ML, supervised learning methods, and their applications Nov. 8: Introduction to Deep Learning and its applications ________________ Per our policy, attendees must have an active TACC account to participate in training activities on TACC resources. To verify that your login information for an existing account is current and that your account is active, or to create a TACC account, please visit https://accounts.tacc.utexas.edu/ at least one week prior to the start of the training to ensure that TACC staff can review and approve your account request. *If you are creating a new TACC account, we recommend that you use your institution email address (e.g. name@utexas.edu) for prompt account approval and verification. 

REGISTER for TACC trainings: https://www.tacc.utexas.edu/use-tacc/training/ *stay tuned for a CMake tutorial on 12/1 -- registration opening soon!Registration is limited and closes Friday 10/20:HPC @ TACC | November 1-2, 2023 (closes at capacity)Virtual only 10AM CST - 4PM CST In this hands-on tutorial, we will explore methods for running containers on HPC systems including advanced tasks like utilizing GPUs for computation and MPI for parallel jobs Docker to perform multi-stage and multi-architecture builds.  - Introduction and motivation to parallel computing, which includes an introductory lab on TACC resources. - Edit and compile code, submit jobs to the queue - MPI: beginner and intermediate - Basic concepts of the Message Passing Interface (MPI) library for large scale parallel programming - Interactive combination of lectures and code walk-throughs - HPC tools specifically designed for complex user environments (Lmod, Sanity Tool) - Tools for workflow management (ibrun, launcher, launcher-GPU, Pylauncher) - Tools for job monitoring and profiling (Remora, TACC-Stat, core_usage, amask) - Other tools per participant interest - MPI with containers  ________________ Machine Learning | November 6-8, 2023 (closes at capacity)Virtual only 10AM CST - 4PM CST TACC’s Machine Learning training is designed to give participants fundamental knowledge of machine learning (ML) and deep learning (DL) techniques as well as experience with hands-on tutorials running ML and DL workflows on HPC systems. Participants will learn the basics of how to start using the computing clusters at TACC for ML and DL analysis. They will also learn fundamentals and theory of ML and DL techniques, and explore example applications of how ML and DL techniques can be applied on TACC’s HPC clusters. This training is virtual on the following days: Nov. 6: Computing cluster and Python environment at TACC Nov. 7: Introduction to ML, supervised learning methods, and their applications Nov. 8: Introduction to Deep Learning and its applications ________________ Per our policy, attendees must have an active TACC account to participate in training activities on TACC resources. To verify that your login information for an existing account is current and that your account is active, or to create a TACC account, please visit https://accounts.tacc.utexas.edu/ at least one week prior to the start of the training to ensure that TACC staff can review and approve your account request. *If you are creating a new TACC account, we recommend that you use your institution email address (e.g. name@utexas.edu) for prompt account approval and verification. 

REGISTER HERE: https://www.tacc.utexas.edu/use-tacc/training/HPC @ TACC | November 1-2, 2023 (registration closes at capacity and/or on 10/25/23) This training is virtual / webinar from 10AM CST - 4PM CST In this hands-on tutorial, we will explore methods for running containers on HPC systems including advanced tasks like utilizing GPUs for computation and MPI for parallel jobs Docker to perform multi-stage and multi-architecture builds. - Introduction and motivation to parallel computing, which includes an introductory lab on TACC resources.- Edit and compile code, submit jobs to the queue - MPI: beginner and intermediate - Basic concepts of the Message Passing Interface (MPI) library for large scale parallel programming - Interactive combination of lectures and code walk-throughs - HPC tools specifically designed for complex user environments (Lmod, Sanity Tool)- Tools for workflow management (ibrun, launcher, launcher-GPU, Pylauncher) - Tools for job monitoring and profiling (Remora, TACC-Stat, core_usage, amask)- Other tools per participant interest - MPI with containers  ________________ Machine Learning | November 6-8, 2023 Virtual only / webinar from 10AM CST - 3PM CST (registration closes at capacity) TACC’s Machine Learning training is designed to give participants fundamental knowledge of machine learning (ML) and deep learning (DL) techniques as well as experience with hands-on tutorials running ML and DL workflows on HPC systems. Participants will learn the basics of how to start using the computing clusters at TACC for ML and DL analysis. They will also learn fundamentals and theory of ML and DL techniques, and explore example applications of how ML and DL techniques can be applied on TACC’s HPC clusters. This training is virtual on the following days: Nov. 6: Computing cluster and Python environment at TACC Nov. 7: Introduction to ML, supervised learning methods, and their applications Nov. 8: Introduction to Deep Learning and its applications  ________________ Per our policy, attendees must have an active TACC account to participate in training activities on TACC resources. To verify that your login information for an existing account is current and that your account is active, or to create a TACC account, please visit https://accounts.tacc.utexas.edu/ at least one week prior to the start of the training to ensure that TACC staff can review and approve your account request. *If you are creating a new TACC account, we recommend that you use your institution email address (e.g. name@utexas.edu) for prompt account approval and verification. https://www.tacc.utexas.edu/use-tacc/training/

TACC ticketing systems will be down for maintenance on Friday, October 6 beginning at 2 pm CDT. User submitted requests may experience delays. 

Frontera will be reserved for Texascale Days - very large scale runs of at least half system size (3500 - 4096 nodes) or full system size (7000 - 8192 nodes) next week (10/11/23-10/16/23. All production queues with the exception of the rtx and nvdimm queues will be reserved for Texascale Days until Monday, October 16th at 09:00 (CDT).Please submit any questions you may have via the TACC User Portal.https://tacc.utexas.edu/about/help/

Corral mounts on all TACC HPC systems (Frontera, Stampede 2 and Lonestar 6) will be unavailable from 8:00 AM to 5:00 PM (CDT) on Tuesday, September 26th while system maintenance is performed. All other access to Corral via data.tacc.utexas.edu, web-based access, and other services will remain in production during this time.Please submit any questions you may have via the TACC User Portal.https://tacc.utexas.edu/portal/tickets

The last chance to register for the TACC Symposium for Texas Researchers (TACCSTER) is Friday September 8th. TACCSTER will be held on September 27-29, 2023, in person in the ACB building on the J.J. Pickle Research Campus (PRC) of The University of Texas at Austin.TACCSTER is a meeting of scientists, engineers and scholars from across the state who use, and would like to use, the Texas Advanced Computing Center (TACC) to advance their research goals. Researchers from across the country are welcome to attend. This meeting will be an opportunity for you to:Hear invited speakers from a broad range of disciplinesVisit 40+ posters from institutions all over TexasNetwork with your fellow TACC users and researchersLearn new and useful computing skills from a variety of tutorialsFind out information about the latest TACC systems and servicesTo register and for more information please visit: https://www.tacc.utexas.edu/taccsterPlease contact taccster@tacc.utexas.edu if you have any questions.

We are pleased to announce the sixth annual TACC Symposium for Texas Researchers (TACCSTER) being held on September 27-29, 2023, in person in the ACB building on the J.J. Pickle Research Campus (PRC) of The University of Texas at Austin.TACCSTER is a meeting of scientists, engineers and scholars from across the state who use, and would like to use, the Texas Advanced Computing Center (TACC) to advance their research goals. Researchers from across the country are welcome to attend. This meeting will be an opportunity for you to:Showcase your ongoing research involving TACC resourcesLearn new and useful computing skills from a variety of tutorialsNetwork with other TACC users / researchersHear information about the latest TACC systems and servicesTo register and for more information please visit: https://www.tacc.utexas.edu/taccsterPlease contact taccster@tacc.utexas.edu if you have any questions.

TACC network infrastructure will not be available from 8 AM to 2:00 PM (CDT) on Sunday, 20 August 2023. Network maintenance will be performed during this time.Please submit any questions you may have via the TACC User Portal. https://portal.tacc.utexas.edu/tacc-consulting

The Corral and Corral-secure file system mounts will be unavailable from Lonestar 6 and Stampede2 on Tuesday, July 18 8:00 AM to 5:00 PM (CDT) for system updates. Access through Corral-specific data transfer nodes, and other HPC systems should be unaffected.Please submit any questions you may have via the TACC User Portal. https://portal.tacc.utexas.edu/tacc-consulting

TACC Institutes are in-person and on-site at TACC. Expect robust content, one-on-one interaction with instructors, networking opportunities, and certificates of completion. Breakfast, lunch and snacks will be provided. ***During July 10-14, attendees will have the option to visit sessions in any of the following Institutes at no additional cost: Scientific Visualization Institute 9:00 (breakfast) - 4:00 PM CST Spend a week with TACC's visualization experts learning how to generate rich, informative visualizations and increase the impact of your work. Video (7 min) "What is Scientific Visualization?" Watch: https://youtu.be/RKRHDS5-tOURegister HERE: https://cvent.utexas.edu/d/q0q1hl/ Reproducible Science Institute 9:00 (breakfast) - 4:00 PM CST In this immersive, hands-on workshop, attendees will learn how to make your scientific workflows more reproducible, automated, and scalable using state-of-the-art tools and technologies. More details and registration here: https://cvent.utexas.edu/d/m0q1n7/ Applied Parallel Programming Institute 9:00 (breakfast) - 4:00 PM CST With TACC's high performance computing experts, learn to write parallel applications using OpenMP and MPI. More details and registration here: https://cvent.utexas.edu/d/j0q1h3/

TACC's VPN will be undergoing maintenance from 5:00 PM to 7:00 PM (CDT) on Friday, June 16 2023. Access to all VPN accessed TACC systems will be unavailable during this time. Jobs will continue to run, but users will have no VPN access to TACC services and systems until the maintenance is complete.

Frontera is currently reserved for Texascale Days - very large scale runs of at least half system size (3500 - 4096 nodes) or full system size (7000 - 8192 nodes) this week. All production queues with the exception of the rtx and nvdimm queues are reserved for Texascale Days until Sunday, June 18th at 09:00 (CDT).Please submit any questions you may have via the TACC User Portal. https://tacc.utexas.edu/about/help/

Registration Closing TODAY 6/5: BootUp & ReBoot: Coding/Hacking BOOTCAMP @ TACC Our Institutes are in-person and on-site. Expect robust content, one-on-one interaction with instructors, networking opportunities, certificates of completion and breakfast tacos. See the instructors in action HERE: https://youtu.be/q6rbPJXYZOk BootUp & Reboot @ TACC*  06/12/23 - 06/16/23  07/31/23 - 08/04/23 *2-part series, 1 week each BootUp/Reboot @ TACC is a dual bootcamp focused on immersing participants into the research life cycle. Attendees will solve a computational scientific model, generate data, verify and validate the data, update the model, revisit the data... and repeat until the model accurately predicts a real-life phenomenon. This is the research life cycle. Part training, part code-a-thon, part hackathon, BootUp and ReBoot is an incredible learning experience and a lot of fun. REGISTER HERE: https://cvent.utexas.edu/lLlyr1

Frontera will not be available from 8 AM to 8 PM (CDT) on Tuesday, 13 June 2023. System maintenance will be performed during this time.Please submit any questions you may have via the TACC User Portal. https://portal.tacc.utexas.edu/tacc-consulting