Saudi Arabian High Performance Computing Conference and Exhibition


KACST, Riyadh. 19-21 April 2016

Register

About the Event



Riyadh - April 2016

The Saudi Arabian High-Performance Computing (SAHPC) Forum is a premier regional event at which HPC industry leaders, application developers, scientific and engineering users, educators, and economic and institutional planners network on the opportunities and challenges of supercomputing. The SAHPC Forum creates national 'critical mass' that enfolds local HPC communities in Saudi Arabia, facilitating exchanges and fostering new collaborations in business, research, and education. It also helps develop a national identity for the Kingdom of Saudi Arabia as an emerging international power in supercomputing.

The main theme of the conference is HPC applications, and the topics up for discussion include:


  • Oil and Gas .
  • Bioinformatics .
  • Meteorology .
  • Security and Defense .

King Abdulaziz city for Science and Technology (KACST) is delighted to organize the forum for this year, 2016, which will take place during 19-20 of April. Saudi Arabia’s HPC community will inspire and enable scientific, economic, and social advances through the development and implementation of computing and data-intensive solutions, while the HPC-trained workforce will foster innovation, attract investment, and create jobs yet unimagined. The SAHPC conference program comprises two days of Keynote speech, talks by HPC experts in above-mentioned themes and plenary discussions. The program also includes a third day (April 21) for a hands-on tutorial on SANAM supercomputer plus the basic principles of parallel programming . However, perhaps the most important parts of the meeting are the breaks and the social events, where networking can be conducted and talks followed up on. During these regular breaks, vendor displays will be available for your perusal. Please take this chance to maximize your opportunities over two rich days of networking. Thank you for joining us!

15+

hours

14+

speakers

25+

technologies

 

Schedule


April 19, 2016
7:30am - 8:30am
Registration & Networking
8:30am - 9:00am
Opening
Prince Turki Al-Saud - President, KACST
Venue: Conference Hall
His majesty prince Turki Al-Saud will give the starting speech
9:00am - 9:30am
Keynote : FAIR – "future accelerator facility and Green IT"
Dr. Volker Lindenstruth - Chair for HPC architecture Goethe University, Germany
Venue: Conference Hall
N/A
9:30am - 10:00am
Keynote: “Solver software infrastructure for exascale applications”
Dr. David Keyes - Extreme Computing Research Center King Abdullah University of Science and Technology
Venue: Conference Hall
At the heart of the International Exascale Software Project (launched in 2009) is the belief that a vast array of applications share a core of tasks that can be abstracted and layered in such a way as to be served by a common software infrastructure, more efficiently than if each community develops their own infrastructure independently. This philosophy is embedded in the “research DNA” of the speaker’s institution, KAUST (also launched in 2009), especially with respect to solver software, represented by the speaker’s Extreme Computing Research Center (ECRC). What is general enough to be leveraged over many applications, however, needs to be tuned for performance to specific evolving hardware. Without the “co-design” of software and hardware, performance of future applications may never exceed today’s few Petaflop/s. The algorithmic adaptations required to migrate today’s successful “bulk synchronous” open source parallel scientific software base to the exascale environment include: (1) reducing synchronization scope and frequency, (2) reducing memory traffic per core, (3) exploiting more instruction-uniform concurrency, and (4) relying more on algorithms for fault tolerance than expensive hardware redundancy. For solvers, attention can be confined to a handful of algorithms that possess optimal complexity: lin-log scaling in problem size. Otherwise weak scaling to a billion cores is not practical. We illustrate with examples from ongoing ECRC research.
10:00am - 10:15am
Break & Networking
10:15 am -11:00 am
Sponsored Talk by Intel (details will be posted soon)
Dr. Alan Gara - chief architect for Exascale Systems for the Data Center Group Intel Corp.
Venue: Conference Hall
N/A
11:00 am – 11:45 am
On the future of HPC Solutions
Dr. Stephen Wheat - Director of the HPC Pursuits team - HPE
Venue: Conference Hall
As shown at www.Top500.org, the performance growth of systems has become slightly less steep. Nevertheless, the performance growth remains phenomenal. That growth is fueled by the myriad and independent technology advancements. The rate and periodicity of introduction of advancements is high and frequent, presenting a flow of continuous improvements as well as constant change. HPE is focusing on developing customer solutions that deliver the value of those improvements while not disrupting the customer’s focus on achieving their objectives. This talk will describe several HPE solutions visions and their motivations. The overall objective is to deliver sustainable customer value through faster adoption of new technologies with greater confidence with an objective towards the customer’s investment protection.
11:45 am – 1:15 pm
Prayer & Lunch
Venue: Conference Hall
N/A
1:15 pm - 1:30 pm
Sponsored Talk by Huawei (details will be posted later)
Venue: Conference Hall
N/A
1:30 pm -

Session 1 (Bio-informatics)


1:30 pm – 2:00 pm
High Performance Computing for Bioinformatics: The Saudi Genome Project case
Dr. Mohamed Abouelhoda - Head of bioinformatics team of the Saudi Genome Project, KACST & KFSHRC - KSA
Venue: Conference Hall
The Saudi Genome Project aims at sequencing 100,000 genomes over a period of five years. This produces huge amounts of data that need to be analyzed in an efficient and timely manner. In this talk, we will address the computational challenges related to this task, considering the structure of the sequencing facilities and the versatility of the analysis requirements. We will show how the use of high performance computing systems and layers built on top of the native and cloud hardware could address the requirements in an efficient manner. We will also shed light on the still coming challenges with new sequencing technologies and new medical applications.
2:00 pm – 2:30 pm
Solving Motif Finding Problem on Aziz Using CPU, GPGPU, and MIC Architectures
Dr. Hossam Faheem - HPC manager, Fujitsu Company
Venue: Conference Hall
The Motif Finding Problem MFP is a computationally intensive problem in the bioinformatics domain. Solving such problems on heterogeneous clusters consisting of CPUs, CUDA GPUs, and Intel Many Core (MIC) architectures is considered a challenging problem. This presentation shows how to solve the MFP on Aziz supercomputer where we can find CPU, GPGPU, and MIC architectures. Different scheduling strategies can be used on Aziz. Different parallel computing paradigms such as MPI, OpenMP, CUDA, and MIC can be easily deployed. We can find that the total execution time will be significantly improved when compared to pure CPU-based implementation. The presentation then shows how to modify the code to assign the tasks to the architectures.
2:30 pm – 3:00 pm
High Performance Computing for High Throughput Genomics: experience from Qatar Genome Project
Andrey Ptitsyn - Principal Scientist Sidra Medical and Research Center, Qatar
Venue: Conference Hall
Establishing Sidra Medical and Research Center created a significant challenge for design and implementation of modern computational research infrastructure. With projected staff of 800 doctors and 500 dedicated research personnel Sidra is expected to become one of the largest Biomedical research Centers in the Middle East. The challenge was further augmented by the advent of the National Genome program aiming to acquire WGS for the entire population of Qatari nationals with projected growth up to 350,000. High-Performance Computing facility is major research instrumentation for modern Biomedical Science and a requirement for a large-scale research in Genomics. Working with IBM solution architects we have designed a custom architecture specifically optimized for data-intensive computation in Genomics and Biomedical Big Data Analytics. In this talk I would like to share the experience of building an HPC facility custom designed for Genomics and Health Science Analytics.

- 3:00 pm

End of session 1


7:30 pm -
Gala Dinner
Venue: Conference Hall
N/A
April 20, 2016

8:30am - 9:00am
Keynote: Exploiting Hierarchical Locality for Productive Extreme Computing
Dr. Tarek El-Ghazawi - Professor and IEEE Fellow - The George Washington University
Venue: Conference Hall
Modern high-performance computers are characterized with massive hardware parallelism and deep hierarchies. Hierarchical levels may include cores, dies, chips, and nodes to name a few. Locality exploitation at all levels of the hierarchy is a must as the cost of data transfers can be high. Programmer’s knowledge and the expressivity of locality-aware programming models such as the Partitioned Global Address Space (PGAS) can be very useful. However, locality awareness can come at a high cost. In addition, asking programmers to worry about expressing locality relations at multiple architecture hierarchy levels is detrimental to productivity and systems and hardware must provide adequate support for exploiting hierarchical locality. In this talk I will discuss a framework for understanding and exploiting hierarchical locality in preparation for the next era of extreme computing. The role of system and hardware support will be highlighted will be stressed and examples will be shared.
9:00am -

Session 2 (Oil & Gas)



9:00am - 9:30am
What HPC Can Learn from DevOps: Aramco Use Case
Eng. Walid Shaari and Eng. Ahmed Bukhamsen - HPC Systems Engineers and Solutions Architect, Saudi Aramco
Venue: Conference Hall
What HPC operations can learn from web operations and devops? HPC community can’t escape the inevitable growth, and diversity requirements of HPC infrastructure, the demands of the HPC developers, and users for smarter effective use and deployment of resources. Growth and innovation in large scale infrastructures management is a problem already solved for technology leaders “Google, Amazon, Facebook, Microsoft, and twitter”, having adopted Cloud, Containers, big data technologies as well as web ops, and devops practices, they are changing rapidly the IT, and Enterprise scene. This presentation will try to highlight some of the challenges, insights and lessons learned via a use case of the Aramco HPC team journey with configuration management, and orchestration. highlight some of the differences, and benefits of both communities. The ability to have elastic robust infrastructure addressing future HPC Business demands allowing end users self-service on-demand computing.
9:30am - 10:00pm
:Deep Neural Networks and GPUs
Guy Gueritz - Oil & Gas EMEA Industry Business Development Manager , NVIDIA
Venue: Conference Hall
lthough not a new idea, the interest in deep learning has exploded in the past few years because three key ingredients have come together: huge amounts of data is gathered by internet service providers; researchers are rapidly advancing the techniques used; and the GPU has been recognised as the ideal computing platform to deliver the enormous processing resources needed. We will explain some of the underlying concepts of deep learning, look at a few application areas, and review the hardware and software solutions that NVIDIA offers.
10:00am - 10:30am
Building a case for automation: Positive Business Impact of Automation
Antoine Schonewille - Saudi Aramco
Venue: Conference Hall
Growing HPC environments see an exponentially growing number of problems that might interfere with the normal HPC process (i.e. computation), of which a big part can be automated. Doing this automation the proper way can save time and make life easier while it takes away repetitive or routine tasks. One of the in-house developed systems implemented at the end of 2014 helps in determine failing or about to fail compute resources. This specific automation entirely manages the detection, test flow and status communication. Doing this alone has saved many hours of labor. Although this is a work in progress, results gathered during a year showed an overall improvement in compute resources availability and reliability. A recent addition of automated Job statistics gathering and correlation gives clear insights of the whereabouts of the HPC environment and the focus points. As a result, the total solution dramatically increased cluster reliability and as such computation (job) success rate.

- 10:30am

End of session 2



10:30am - 10:45am
Coffee Break
Venue: Conference Hall
N/A

10:45 am -

Session 3 (Meteorology)



10:45 am – 11:15 am
Algorithmic Challenges and Scientific Opportunity for High-Fidelity Earth System Modeling as We Approach the Exascale Horizon
Dr. Richard MillsHPC Earth System Models Architect, Intel Corp.
Venue: Conference Hall
Numerical weather, climate, and Earth system simulations play an increasingly vital role in society, informing decisions affecting economics (e.g., how much fuel is required to fly a given airplane route) to health and safety (e.g., when to issue evacuation orders ahead of hurricanes) to policy (e.g., climate change mitigation and adaptation). Although the accuracy of these simulations has increased steadily over the past several decades, forward models still suffer from long-standing sources of systematic errors--due to limitations in model resolution as well as incomplete representation of coupled physical and biogeochemical processes--that limit their usefulness for many important applications. Increasing computational power as we approach the exascale horizon promises to enable scientists to address some of these shortcomings, but doing so is complicated by both fundamental challenges in modeling coupled multiscale, multiphysics phenomena, as well as algorithmic issues arising from the increasing complexity of high-performance computing (HPC) architectures, which demand that codes be able to efficiently use increasingly high degrees of parallelism and deepening hierarchies for data movement (multiple types of user-addressable memory, many NUMA domains and caches, on node vs. off node, local vs. wide neighborhoods on the interconnect, and even within the I/O system). We will discuss some of the capabilities being considered for next-generation Earth system models, the possible interplay with hardware architectural trends, and some potential approaches for engineering (or reengineering) Earth science codes efficiently utilize next-generation HPC architectures.
11:15 am – 11:45 am
On the Role of High Performance Computing in Dust Storms Predictions over the Arabian Peninsula
Dr. Ashraf Farahat - Assistant Professor - KFUPM
Venue: Conference Hall
While High Performance Computing (HPC) has a significant role to play in predicting the weather. With a high degree of accuracy, scientists can now predict temporal and spatial atmospheric conditions using HPC. Dust storms represent one of the most extreme weather conditions affecting the Arabian Peninsula as it contains one of the largest sand deserts in the world, Al Rub’ Al Khali, it is also the home of the An-Nafud desert, who sand dunes often exceed a height of 30 meters. Major dust events can cause accidents, affecting ground and air traffic, financial market, and causing respiratory health problems. No one can prevent extreme dust storms, but they can prepare for it – and improved prediction is critical to improving preparation. Current dust storm models could handle smaller length scales down to few meters, however smaller grid analysis is equivalent to longer simulation times. Researchers are limited in how much computational power they have with which to run dust storms simulations. Some processes simply cannot be included. More computational capability could allow researchers to simulate the actual behavior of the dust storms activities. In this talk, we will present how HPC technology could integrate dust storms data collected from satellites and ground based sensors with dust computational models for more accurate predictions of dust storms sources and tracks.

- 11:45 am

End of session 3


11:45 am – 1:15 pm
Lunch & Networking
Venue: Conference Hall
N/A
1:15 pm -

Session 4 (Defense & Security)



1:15 pm – 1:45 pm
Development of CFD Using HPC in Aerospace Applications; Software and Hardware Trends
Dr. Khaled Alhussan - Director of Space and Aeronautics Research Institute - KACST
Venue: Conference Hall
This presentation includes comprehensive discussion on various simulation results obtained using HPC for Computational Fluid Dynamics. Beginning with an introduction to HPC, the presentation moves on to consider how modern supercomputers can be used to solve complex CFD challenges such as Turbulent Flow Modeling, Shock Waves and discontinuities analysis, Design and Optimization in Aeronautics, keeping in view the key aspects while using HPC i.e. scalability, reliability and accuracy, which is discussed in detail with reference to the software and hardware’s trends.

1:45 pm – 2:15 pm
The Convergence of HPC and Big Data – Challenges and Opportunities
Dr. Herbert Cornelius - Technical Director Advanced Computing, Intel EMEA
Venue: Conference Hall
N/A
2:15 pm – 2:45 pm
Applications of HPC in Defense Imagery
Eng. Ayman Alharbiy - Advanced Sensors and Electronics Defence Researcher, KACST
Venue: Conference Hall
In the world of defense, every bit of information counts. This information is used to detect threats around us. For that, defense systems utilize a lot of sensors. Imaging sensors is drawing attention due to its uniqueness and richness. Unfortunately, image processing algorithms are time consuming; and in defense, every second has a price. HPC, with its high processing power, can provide an optimal platform for many image-based applications. The power of HPC can be exploited by heavy tasks.

- 2:45am

End of session 4



2:45 pm – 3:15pm
End of the event
Venue: Conference Hall
N/A
April 21, 2016

The tutorial is aimed to introduce the attendees to some HPC concepts and usage with some practical examples. It will cover the benefits of using HPC over using ordinary computing. The target audience of this tutorial can either be: (i) a programmer who will write code for HPC architectures and models, (ii) a user who needs to use some utilities which have been developed for HPC machines to parallelize the executions. So the tutorial will cover topics related to both types of users. The topics that will be covered include: concepts and terminology of parallel computing, HPC architecture and task scheduling.


Parallel Track
08:30 am - 11:30 am
Introduction to HPC (HPC 101)
Eng. Ibrahim Al-Fuwaires
Venue:
KACST Conference Hall - Main Auditorium
This tutorial introduces fundamental HPC principles and techniques to design and build High Performance Computing Systems (HPC) based on the Linux operating System. It sheds light on the building blocks of HPC, such as the high-speed interconnects, the HPC scheduler, Parallel Programming techniques (OpenMP, MPI, OpenAAC, GPU programming, ), HPC monitoring and benchmarking, and HPC benchmarking techniques .

11:30 am - 11:45 am
An overview about SANAM
Eng. Fahad Alzannan
Venue:
KACST Conference Hall - Main Auditorium
This short presentation will briefly introduce the audience to the KACST HPC machine (SANAM).
Parallel Track
8:30am - 11:30am
Containers workshop: Docker, and its ECO system for distributed systems
Eng. Walid Shaari

Limited to 20 Sets - confirmations will be sent - closed

Venue:
Training Center - Lab 126
This workshop will cover the Theory and hands-on of Docker containers, and Its eco system. The foundations of the Docker platform, including an overview of the platform system components, images, containers and repositories, installation , using Docker containers from repositories e.g. dockerhub, how to create a container using Dockerfile, containers development life cycle. The strategy is to demonstrate through "live demo, and shared exercise" the reuse and customization of components to build a distributed system case service gradually

11:30 am - 12:45 pm
Prayer break


12:45pm - 02:45pm
Introduction to Parallel Programming (MPI, OpenMP and GPGPU Programming using OpenAcc)
Dr. Majdi Baddourah.
Venue:
KACST Conference Hall - Main Auditorium
Parallel computing is very essential to solve real complicated computational problems. Parallel programing uses state of the art high performance computers with many CPU’s working together to solve very large problem. Writing a parallel program requires the use of special tools/libraries. There are many tools/libraries used in parallel programing such as MPI, OpenMP and OpenACC. MPI (Message Passing Interface) is a library that is used to perform the needed inter-process communication. OpenMP and OpenAcc allows programs to use simple directive to achieve parallel programs. The tutorial will provide an introduction to parallel programing, how to use MPI, OpenMP, OpenACC to write a parallel program with the use of simple applications.

Speakers


  • Portfolio Item

    Dr .Khalid Alhussan

    Director of Space and Aeronautics Research Institute - KACST

    Portfolio Item

    Dr. Mohamed Abouelhoda

    Head of bioinformatics team of the Saudi Genome Project, KACST & KFSHRC - KSA

    Portfolio Item

    Dr . Hossam Faheem 

    HPC Manager

  • Portfolio Item

    Eng. Ayman Alharbiy

    Researcher at Advanced Sensors and Electronics Defense - KACST

    Portfolio Item

    Ashraf Farahat

    Assistant Professor - KFUPM

    Portfolio Item

    Andrey Ptitsyn

    Principal Scientist - Sidra Medical and Research Center - Qatar

  • Portfolio Item

    Eng. Walid Shaari

    HPC Systems Engineer and Solutions Architect - Saudi Aramco

    Portfolio Item

    Dr. Stephen Wheat

    Director of the HPC Pursuits team - HPE

    Portfolio Item

    Antoine Schonewille

    Saudi Aramco

  • Portfolio Item

    Dr. Richard Mills

    HPC Earth System Models Architect - Intel Corporation

    Portfolio Item

    Dr . Tarek El-Ghazawi

    Professor and IEEE Fellow - The George Washington University

    Portfolio Item

    Dr. David Keyes

    Director, Extreme Computing Research Center King Abdullah University of Science and Technology

  • Portfolio Item

    Dr. Volker Lindenstruth

    Chair for HPC architecture- Goethe University, Germany

    Portfolio Item

    Dr. Herbert Cornelius

    Technical Director Advanced Computing - Intel EMEA

    Portfolio Item

    Timothy Lanfear

    NVIDIA

  • Portfolio Item

    Eng. Ahmed BuKhamsin

    Saudi Aramco – High Performance Computing Group

    Portfolio Item

    Dr. Alan Gara

    chief architect - Intel

committee


Committee


Abdulqadir Al-Aqili's photo

Abdulqadir Alaqeeli

Director of the National Center for Computing Technology and Applied Math - KACST

Abdulrahman AlRubaish's photo

Abdulrahman AlRubaish

Senior Exploration Systems Consultant - Saudi ARAMCO

Abdullah Alfadhly's photo

Abdullah Alfadhly

Assistant Research Prof. - KACST

Jamal Alhiyafi's photo

Dr. Jamal Alhiyafi

Chairman and Assistant Professor, CS Department, College of CS and IT - University of Dammam

Neeraj Bajpai's photo

Neeraj Bajpai

Senior IT Architect - IBM Middle East

Adle Fadhl Noor Ahmed's photo

Prof. Adel Fadhl Noor Ahmed

Dean College of Computer Science and Engineering - KFUPM

Ahmad Alabduljabbar's photo

Ahmad Alabduljabbar

HPC and Commercial lead, Saudi Arabia - Intel Corp.

Ahmad Al-Jeshi's photo

Ahmed Al-Jeshi

META Regional HPC Solutions Architect - Intel

Robert Eades's photo

Dr. Robert Eades

Research Advisor and Executive for Education and HPC - IBM Middle East and Africa (MEA)

Raed Al-Shaikh 's photo

Dr. Raed Al-Shaikh

IT Appl/Sys Specialist - Saudi ARAMCO

David Keyes's photo

Dr. David Keyes

Director of the Extreme Computing Research Center - KAUST

Elmootazbellah N. Elnozahy's photo

Prof. Elmootazbellah N. Elnozahy

Dean and Professor of Computer Science – KAUST

Guy Gueritz's photo

Guy Gueritz

IBD Oil & Gas EMEA - NVIDIA

Saad Alamri's photo

Dr. Saad Alamri

CIO – Dammam University

Timothy Lanfear's

Timothy Lanfear

Manager of the European solution architecture and engineering team - NVIDIA’s Enterprise Solutions Group

Abdulqadir Al-Aqili's photo

Dr.Abdullah Alfadhly

Abdulrahman AlRubaish's photo

Dr. Abdullah Alrajeh

Abdullah Alfadhly's photo

Dr. Waleed Alsanie

Jamal Alhiyafi's photo

Eng. Ahmed Alnafessah

Neeraj Bajpai's photo

Eng. Riman Bin Sulaiman

Adle Fadhl Noor Ahmed's photo

Eng. Fahad Alzannan

Abdulrahman AlRubaish's photo

Naif Almutayri

Adle Fadhl Noor Ahmed's photo

Eng. Ahmed Alharthi

Abdulqadir Al-Aqili's photo

Dr.Abdullah Alfadhly

Jamal Alhiyafi's photo

Eng. Ahmed Alnafessah

Abdullah Alfadhly's photo

Eng. Sultan Algamdi

Abdulrahman AlRubaish's photo

Naif Almutayri

Abdulqadir Al-Aqili's photo

Eng. Khaled Almoghamis

Jamal Alhiyafi's photo

Eng. Abdulaziz Alayed

Abdullah Alfadhly's photo

Abdullah Alqhtani

Neeraj Bajpai's photo

Saeed Algamdi

Adle Fadhl Noor Ahmed's photo

Faisal Alyousef

Adle Fadhl Noor Ahmed's photo

Abdulrahman Almuqeren

Name:
Job Title:
Organization:
Email:

To register in tutorilas please follow this link Register

Sponsors


HPE

Diamond Sponsors


cray

Platinum Sponsors


Gold Sponsors


Exhibit Sponsors


Accommodation


Address

Al Mashreq Boutique Hotel
Riyadh, Al Uruba st. & Prince Turki Alawal Road intersection
Home offices bld.
Riyadh

Phone

+966 11 - 2834777

Twitter

AlMashreqHotel


*You can get a discount by communicating with Mr: Faisal Alkhodr - Reservation Manager - Mob:+966551472104

Contact


You have any questions? Here how you can reach us

Address

Conference Hall,
4083 King Abdullah Ibn Abdulaziz Rd,
Riyadh

Phone

+966 11 4813768
+966 11 4813765

Twitter

SA_HPC