RMACC 2018 HPC Symposium

This workshop is an introduction to using high-performance computing (HPC) systems, specifically the RMACC Summit compute resource, effectively. It is intended to give students an introduction to HPC and an overview of the tools needed to start making use of Summit for you research computing needs. If you do not already have a Summit account, we will have temporary account credentials available for use for the duration of the workshop.

By the end of this workshop, attendees will know how to:

• Connect to a cluster
• Use basic UNIX commands for managing files
• Work with the Nano editor
• Write simple shell scripts
• Submit and manage jobs on Summit using the SLURM scheduler
• Manage software through environment modules

This workshop is an introduction to using high-performance computing (HPC) systems, specifically the RMACC Summit compute resource, effectively. It is intended to give students an introduction to HPC and an overview of the tools needed to start making use of Summit for you research computing needs. If you do not already have a Summit account, we will have temporary account credentials available for use for the duration of the workshop.

By the end of this workshop, attendees will know how to:

• Connect to a cluster
• Use basic UNIX commands for managing files
• Work with the Nano editor
• Write simple shell scripts
• Submit and manage jobs on Summit using the SLURM scheduler
• Manage software through environment modules

Exascale systems are coming soon. At the same time the demands of science for supercomputing are changing. Rates of data generation and complexity of science require new approaches. I will discuss the current exascale system plans and some of the challenges expected. The primary focus of the talk will be on how we get science applications ready for exascale through both the DOE computing facilities and the Exascale Computing Project. The result of this process is expected to bring in a new way for science to use HPC systems.

​​​​This summer is a fantastic milestone for Pure as we recently announced
"AIRI" - The industry's First Complete AI-Ready Infrastructure.



AIRI is a first in the industry across multiple fronts, but two specifically:
-1st storage platform to be on the main stage of AI (as referenced by
several Analysts and Press over the past several weeks of public-facing briefings) -1st time that NVIDIA has teamed with another company in this fashion within AI.



The promise of the adoption of AI is vast. It's easy to talk about self-driving cars, but AI will touch everything from the Grocery
industry to Healthcare to Transportation through autonomous vehicles ranging from cars to trucks, to planes, and boats. At the NVIDIA GTC conference, NVIDIA CEO Jensen Huang talked about a 25X increase in GPU (the technology that is increasingly driving the performance and scalability of AI through Deep Learning) performance over the last 5 years. But AI has largely been the domain of science fiction and large research institutions and labs.



The Enterprise has largely been on the outside looking in with AI due
to infrastructure complexity and cost. With AIRI, AI is far more accessible to those Enterprise Customers in every industry that will literally change the world. That accessibility to the Enterprise with
AIRI will allow the vast potential of AI's computational capabilities for mass consumption.



Please join Keith Ober, FlashBlade Systems Engineer, for an AIRI architectural overview.

This work presents program evaluation findings affiliated with gender-based selection bias in an international high performance computing (HPC) student program. The research demonstrates application reviewers from all countries are unconsciously biased against female applicants resulting in significantly lower applicant scores. Additional data that contradicts reviewer bias and supports gender parity among participants in HPC knowledge level and academic achievement are presented. Suggestions for reducing selection bias in light of our findings are discussed. This talk also includes preliminary findings from the expansion of this study to members of underrepresented minorities in HPC.

A.I. has been the hot trend of the last two years with roots stemming from the 1950s and further back. Today we are finally seeing the fruit of computational & mathematical excellence lending to an industry explosion in autonomous everything. We will cover how HPE is driving the market by helping customer in their AI journey but also how Hewlett Packard Labs is driving innovation.

Customers have made significant investments in on-premise HPC facilities. Meanwhile, the cloud has emerged as a new way to deliver HPC. Rescale offers a non-disruptive, unified platform approach to HPC in the cloud. In this talk, Gabriel Broner will discuss the use cases, challenges, and benefits for extending HPC to the cloud. A live demonstration will also be given.

Prerequisites:
Basic Linux command line knowledge a plus (but not required)
 Required: Laptop, modern web browser(Chrome,Firefox, Safari)

This tutorial will first give an overview of Jetstream and various aspects of the system. Then we will take attendees through the basics of using Jetstream via the Atmosphere web interface. This will include a guided walk-through of the interface itself, the features provided, the image catalog, launching and using virtual machines on Jetstream, using volume-based storage, and best practices. We are targeting users of every experience level. Atmosphere is well-suited to both HPC novices and advanced users. This tutorial is generally aimed at those unfamiliar with cloud computing and generally doing computation on laptops or departmental server resources. While we will not cover advanced topics in this particular tutorial, we will touch on the available advanced capabilities during the initial overview.

Prerequisites:
Basic Linux command line knowledge a plus (but not required)
 Required: Laptop, modern web browser(Chrome,Firefox, Safari)

This tutorial will first give an overview of Jetstream and various aspects of the system. Then we will take attendees through the basics of using Jetstream via the Atmosphere web interface. This will include a guided walk-through of the interface itself, the features provided, the image catalog, launching and using virtual machines on Jetstream, using volume-based storage, and best practices. We are targeting users of every experience level. Atmosphere is well-suited to both HPC novices and advanced users. This tutorial is generally aimed at those unfamiliar with cloud computing and generally doing computation on laptops or departmental server resources. While we will not cover advanced topics in this particular tutorial, we will touch on the available advanced capabilities during the initial overview.

Would you like to use distributed resources of the Open Science Grid, or just want to learn how to do large-scale high throughput computing?  The format of this tutorial is a mix of lecture and hands-on exercises, so please bring your laptop and make sure you have an SSH client installed.

 You will be provided with a training account on OSG Connect which, during the tutorial, will be upgraded to a full user account. After the session, you will have full access to the OSG, know how to run and scale up workloads, and manage your data. If time permits, the OSG User Support team will also help get your own workload set up for execution on OSG.

Topics to include:
    Introduction to OSG Connect
    Job Scheduling with HTCondor
    Scaling Up Workloads
    Managing data

Would you like to use distributed resources of the Open Science Grid, or just want to learn how to do large-scale high throughput computing?  The format of this tutorial is a mix of lecture and hands-on exercises, so please bring your laptop and make sure you have an SSH client installed.

 You will be provided with a training account on OSG Connect which, during the tutorial, will be upgraded to a full user account. After the session, you will have full access to the OSG, know how to run and scale up workloads, and manage your data. If time permits, the OSG User Support team will also help get your own workload set up for execution on OSG.

Topics to include:
    Introduction to OSG Connect
    Job Scheduling with HTCondor
    Scaling Up Workloads
    Managing data