University of Hawaii System Seal

High Performance Computing

University of Hawaii System

Information Technology Services – Cyberinfrastructure

Full shutdown to perform core infrastructure update of Mana HPC

On July 7th, 2021, Mana will be inaccessible as we perform our maintenance. We will need to hold and requeue all jobs and power down all Mana systems.   A description of how this is different from our normal monthly maintenance days can be found here: 

http://go.hawaii.edu/LVP

About HPC

Mana is the University of Hawai‘i (UH) high performance computing (HPC) cluster – a collection of many computers called nodes connected together with a network – that solves computational problems too large for standard computers.

Resources & Services

  • High performance computing resources
  • Consulting
  • Individual and group training

Eligibility & Sign-up

Accounts are free of charge for all active UH faculty, staff, students across all ten campuses.

You are required to register for an HPC onboarding session to create your account.

Review onboarding training material ahead of your session.  A link to this material is also provided each time you connect to the HPC.

HPC 101 Learning Resources:

https://hpc-carpentry.github.io/

https://carpentries.org/

Research Data Management

Globus is a non-profit service for secure, reliable research data management.

What can I do with Globus?

  • Transfer files: From kilobytes to petabytes, with Globus you can efficiently, reliably, and securely move data between systems within your site or across an ocean
  • Share files with othersAll you need is an email address to share data with colleagues – Globus manages authentication and access
  • Develop applications and gateways: Our open REST APIs and Python SDK empower you to create an integrated ecosystem of research data services, applications, and workflows

Register for a free Globus Account:

https://docs.globus.org/how-to/get-started/

Need help setting up Globus with the Mana HPC?

Contact us at uh-hpc-help@lists.hawaii.edu

HPC in the Classroom

Use HPC in the classroom by requesting a teaching allocation for an upcoming listed UH course.

*Requests must be made no less than 1 month prior to the start of the term in which the allocation will be used.

 

Resource requirements will be discussed with instructors that include:

      • Software needs and installation
      • Number of nodes and/or reservations required for students in the class
      • Need for shared storage between class members
      • Need for onboarding facilitation
      • Other special arrangements

 

Send your teaching allocation request to: uh-hpc-help@lists.hawaii.edu

Courses powered by HPC

 

COURSE INSTRUCTOR CAMPUS OFFERED
BOTANY 662: High Throughput Sequencing for Ecology and Evolution Anthony Amend Mānoa Spring 2015 – Spring 2017
CEE 696: Deep Learning in Civil and Environmental Engineering and Earth Science Jonghyun Harry Lee Mānoa Fall 2019
CHEM 761: Advanced Biochem Rui Sun Mānoa Spring 2018
CHEM 761: Computational Study of Condensed Matters: from Theories to Applications Rui Sun Mānoa Fall 2019
CHEM 761: Special Topics in Biochemistry: Molecular Understanding of Proteins Ellinor Halglund Mānoa Fall 2020
CS 272: Machine Learning for Data Science Travis Mandel Hilo Spring 2018
ICS 621: Analysis of Algorithms Nodari Sitchinava Mānoa Fall 2017
ICS 632: High Performance Computing Henri Casanova Mānoa Fall 2015, Fall 2016, Fall 2017
ICS 635: Machine Learning Peter Sadowski Mānoa Spring 2019
ICS 636: Info Theory in Machine Learning Susanne Still Mānoa Spring 2019
ICS 683: Advanced Computer Vision Kyungim Baek Mānoa Fall 2018
ICS 691: Deep Learning with Neural Networks Peter Sadowski Mānoa Fall 2018
ICS 421: Database Systems II Lipeow Lim Mānoa Spring 2018
MATH 271: Applied Statistics with R Grady Weyenberg Hilo Fall 2018
MBBE 683: Adv Bioinformatics for Biologists Gernot Presting Mānoa Fall 2018
PEPS 662: Systemics and Phylogenetics Daniel Rubinoff Mānoa Fall 2019
ZOOL 719: Systematics and Evolution Rebecca Chong & Michael Kanter Mānoa Fall 2018

We provide exclusive cyberinfrastructure resources across the ten campus system to support technical research and educational needs. Faculty and staff can purchase or lease compute nodes to power their research and select a long term storage solution for their data.

High Performance Compute Nodes

We offer computational resources that fit the needs of your research. Gain exclusive access to compute nodes that are at your disposal through a condo node purchase or lease nodes for a time period of your choice.

Condo Nodes

Condo nodes ordering will be available from Sept. 1-30, 2021

Nodes include both Infiniband HDR100 and 25GB ethernet networking to the cluster, along with 5-year warranties.

Name Description Price
CPU Nodes
Standard  Node 2 x Intel Xeon 6240R 2.4GHz, 48 core total, 192Gb RAM $9,120*
GPU Nodes
GPU Node 2 x Intel Xeon 6248R 3GHz, 48 core total, 192Gb RAM, 8 x Quadro RTX 5000 w/ 16GB GDDR6 RAM $28,060*

*Please note that prices listed are based on previous orders, and may be subject to change as orders need to go out for bid.

Node Leasing

Node Type Time Period Price
Community Node 1 to 12 months $110 per month per node*

*subject to indirect cost

 

Node Ordering Process

How to purchase condo nodes:

Instructions on how to order nodes will be available when node sales open.

How to lease nodes:

UH faculty and staff are eligible to lease nodes by completing the Standard Node Lease Form

Storage

Accelerate your research and store your data on the cluster for faster access.

Long Term Storage

Long Term Storage (LTS) is a Network Attached Storage (NAS) system that can be leased by UH faculty and staff. Contracts are typically for a one year period, but longer contracts can be made.

 

Storage Size Time Period Price
1 TB
(2 TB minimum order)
1 year $20 per TB*

*subject to indirect cost

Storage Ordering Process

Use the Long Term Storage Order Form to place your order.

Once we receive a completed form, the following information is required to fulfill the order:

  • group name that we can use to limit access to your purchased storage
    • group name should end with _group or _lab and be at least 9, alphanumeric characters, which may also include underscores
    • Example: kite_group
  • UH email address of those who should have access to the newly created LTS volume

If you have questions please contact us at uh-hpc-help@lists.hawaii.edu.

We’re here to support, train, and help. Contact us with HPC challenges, software requests, refresher training, or schedule a consultation.

uh-hpc-help@lists.hawaii.edu.

Video Tutorials

Learn HPC basics with our video tutorials.

Resource Documentation & User Guide

Learn more about Mana by reading our resource documentation and user guide. Commonly asked questions can be found in the FAQ section.

Mana Monthly Maintenance

Please note: The first Wednesday of each month is reserved for HPC maintenance: jobs running on the cluster may have to stop and be restarted.

Mana Constraints & GPU Options

A list of constraints and GPU options for Mana can be found at the following link:

Constraints and GPU options

Campus Champions

Need additional help? Connect with a Campus Champion.

Campus Champions help their institution’s researchers, educators and scholars (faculty, postdocs, graduate students, undergraduates, and professionals) with their computing-intensive and data-intensive research, education, scholarship and/or creative activity, including but not limited to helping them to use advanced digital capabilities to improve, grow and/or accelerate these achievements.

Campus Champions is an Extreme Science and Engineering Discovery Environment (XSEDE) program with a group of over 600 experts at more than 300 colleges, universities and research-focused institutes that help scientists use research computing. 

HPC Specifications:

  • 345 compute nodes
  • 120 GPUs
  • 8,452 cores
  • 62.9 TB of memory
  • 1 PB of long term storage
  • 61 TB of flash scratch storage
  • 150 TB of standard scratch storage

Policies & Procedures:

User accounts:

The “User Account Management Policy” describes in detail the who can use the UH HPC and how the life cycle of cluster users accounts will be managed. [ http://go.hawaii.edu/0SG ]

Unauthorized access is prohibited by law in accordance with Chapter 708 of the Hawaii Revised Statutes all use is subject to University of Hawaii Executive Policy EP 2.210.(http://go.hawaii.edu/JYj)

User quotas and usage of common systems:

The “Resource Management and Usage Policy” details the allowed usages of the disk and compute resources on new systems. Specifically, these include login nodes, date transfer nodes (dtn) usage, openvpn usage (experimental), home directories, group directories, scratch filesystem usage and scheduler policies. [ http://go.hawaii.edu/GSY ]

User account security:

The “Access and Security Policy” addresses the use of DUO multi-factor authentication, ssh connection timeouts and password failure lockouts. [ http://go.hawaii.edu/WSG ]

On and off warranty condo nodes:

The “Procurement/purchasing of compute nodes and compute time Policy” describes the life cycle of condo nodes, continued availability of leased nodes and the end of the service unit model for users to purchase time on the cluster.[ http://go.hawaii.edu/GK0 ]

Central storage:

The “Procurement/purchasing of Storage” document describes the paid option for storage on the xCAT cluster called Long Term Storage (LTS).  [http://go.hawaii.edu/YKG ]

 

Mana is also subject to the additional University of Hawaii policies:

UH researchers use our advanced computing resources to make discoveries across multiple disciplines of science. 

 

Current Research on UH HPC

The User Research section provides a list by year of publication of User Research projects that take advantage of our systems, software, services, and staff.

Acknowledge ITS-CI

If you use ITS-CI systems or services, including the HPC cluster, please let us know of any published results, including student theses, conference papers, journal papers, presentations and web pages.  Please acknowledge resources provided by ITS-CI and the University of Hawaii in the following manner:

The technical support and advanced computing resources from the University of Hawaii Information Technology Services – Cyberinfrastructure are gratefully acknowledged.

Also, please submit copies of dissertations, reports, preprints, reprints and/or URLs in which UH ITS-CI is acknowledged to: itsci@hawaii.edu.