---

Abstract

Accelerated workloads are becoming more common and include AI, scientific computing, graphics, and even databases. With more GPUs choices than ever before, it would seem that there is an accelerator for every task. Yet, this also leads to a Paradox of Choice, where it becomes difficult to make a choice due to a large number of options. The goal of this workshop is to bring clarity into the current state of accelerators and to equip attendees with tools helping them evaluate the available choices. Specifically, it will review accelerators from Nvidia and other vendors and compare their performance, performance/Watt and price/performance across AI workloads and beyond. Also, it will show examples of best practices for deployments of accelerated workloads depending on GPU topologies and server internal architectures. Attend to become an expert in accelerated computing.

---

Abstract

Visual analysis is one of the most powerful tools for data exploration and interpretation. It takes advantage of visualization techniques and allows scientists to work with their research data in interactive and intuitive way. In today’s HPC environment and Big Data era, data analysis techniques, together with visualization, gain on importance. However, the amounts of data and the sizes of single datasets impose the need for adequate software tools. In this tutorial we will address this problem by providing participants with strong tool for data processing, visualization and visual analysis – VisNow, an open source generic platform based on data flow paradigm. The goal of this tutorial is to introduce the audience to the concept of visual analysis, show basic ideas of scientific visualization and to go step-by-step through several case studies in hands-on sessions based on our platform. Problems of visualization of common HPC data structures, including 2-D and 3-D, scalar and vector, regular and unstructured data will be covered and adequate elements of the software described to give participants the basics of VisNow usage.

 

Brief agenda

• Introduction to Scientific Visualization and Visual Analysis
• Visualization systems and paradigms
• Generic data structures
• Introduction to VisNow
• Hands-on Session #1 – 2D data visualization
• Hands-on Session #2 – 3D data visualization
• Hands-on Session #3 – Vector data visualization
• Hands-on Session #4 – Unstructured data visualization.

---

Abstract

The tutorial will deliver basic knowledge about the idea, functionality and particularly the practical usage of QCG-PilotJob [1], a flexible tool which allows users in an easy, efficient and unified way to execute a large number of jobs on various types of computing resources. That being said, it should be stressed that the tool is designed to be easily installable and run in the same way regardless if it is an installation on a high-end HPC system , smaller campus resource or a user’s personal laptop.

 

Brief agenda

The tutorial will consists of the following items:

1. Theoretical introduction to QCG-PilotJob
2. Hands-on with Jupyter Notebook Platform:
   1. Registration (it is advised to do this in advance, registration will be available one day before the tutorial)
   2. Installation of QCG-PilotJob
   3. Usage
      1. Batch usage and JSON-based QCG-PilotJob’s request file format
      2. Programmable usage with main API and executor API
   4. Demonstration of practical application of QCG-PilotJob for EasyVVUQ toolkit
3. QCG-PilotJob in its natural environment – an example of a dynamic scenario that was based on QCG-PilotJob in order to execute it on an HPC cluster
4. Q&A

[1] https://qcg-pilotjob.readthedocs.io

 

BIO

The authors of QCG-PilotJob is a team of experienced specialists from the Poznan Supercomputing and Networking Center who for many years are engaged in the development of software supporting complex computations. The team members have extensive knowledge and practical skills in such aspects as tasks scheduling, distributed computing, workflows orchestration as well as design and implementation of user-level tools. These were gained during the course of many national and international projects and on the basis of close cooperation with a variety of stakeholders, including computational scientists on the one hand and representatives of the computing facilities and systems administrators on the other.

---

Abstract

Huawei IdeaHub is the state of the art product line of interactive monitors that promotes creativity as well as efficient and versatile team collaboration. During our workshop I’ll present how simple and inspiring hybrid learning / working could be. With artificial intelligence inside, Huawei IdeaHub turns every room into smart room. Bundles intelligent writing with ultralow 35ms latency, High Definition (HD) video conferencing with speaker tracking, crystal clear audio as well as smooth 4K quality wireless sharing. All these intelligent features will be presented during our workshop.

---

Abstract

For many years, the network of mass storage devices will have a technological island, with more railways. Often dominated by solutions that did not solve the problem of a Fiber Channel solution in the market dismissal process. The amount of data processed and reviews of their publishing terms, such as innovations such as SSD, MVMe or RDMA technology. It has been found that this speed and innovative Ethernet technology connects storage faster and more efficiently than a Fiber Channel network. The presentation show how to use technologies based on Huawei AI technology and Dorado mass storage in a modern way.

---

Abstract

Huawei OceanStor Pacific Next-gen HPDA Storage Introduction

• Huawei had built a full-stack infrastructure solution for HPC.
• focus on the development of IT infrastructure and professional services in HPC,
• offer a variety of professional solutions for customers across different industries.