Loading…

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Software [clear filter]
Tuesday, September 15
 

10:00am CDT

Getting the Most Bang out of your Buck with Compilers on POWER - Archana Ravindar, IBM Systems
Compilers form a critical component of the system stack and form a bridge between the underlying hardware and the higher application program level. In order to get the most of your POWER system, we should know how to use compilers to our advantage and employ its features to run programs most efficiently on the hardware. Compilers are equipped with knobs of various kinds such as flags, builtins and pragmas. Knowing when to use these knobs can make a huge difference performance wise. This presentation will cover these aspects in the context of the IBM XL compiler and open source compilers such as GCC and LLVM on POWER. We shall touch upon the corresponding knobs in each compiler so that the user can pick the knob according to their need. We will also cover how compilers communicate on why they could or could not do certain optimizations. Understanding why can lead to additional performance improvement opportunities and deepen the users understanding of the program and the compiler.

Continue the conversation in Slack

Speakers
AR

Archana Ravindar

Senior Engineer, IBM Systems
Dr. Archana Ravindar is a senior engineer with IBM Systems and works on compiler performance analysis and optimization of industry standard benchmarks and open source workloads on POWER. She has worked on IBM proprietary and open source based compilers such as GCC and LLVM on POWER... Read More →



Tuesday September 15, 2020 10:00am - 10:30am CDT
Track 3
  Software
  • See Session Slides yes

10:35am CDT

OpenFOAM CFD Simulations on Power 9 Platforms - Varun Chebbi, New Horizon College of Engineering
Computational fluid dynamics (CFD) is concept used solve various fluid flow problems.CFD requires intensive computational power and typically needs high-performance computing to reduce potentially long experimentation times.The Cloud based CFD simulation can help many industries in solving their complex fluid flow problems in every field.Of all the CFD tool OpenFOAM is an open source tool for complex fluid flow problems. In the world full of digitalisation,we as a mechanical engineer have eased our R&D projects by running our complex CFD computation using Power 9 cloud servers. IBM Power9 platform incorporates workflow management and multi-cloud implementation across various cloud resources.Here, we present the components of our technology in using it to create a cloud-based version in Analysis CFD software. Our CFD experiment report is mainly on the flow simulation of the heat exchange fluid flow and analyse the thermal flow using couple of test cases using Power9 HPC cloud platform.

Continue the conversation in Slack

Speakers
VC

Varun Chebbi

Research Assistant, New Horizon College of Engineering
Varun Uday Chebbi is an enthusiastic Mechanical Engineering Graduate from Viveswaraiah Technological Univesity, India. He was selected for the Indo-French student exchange from his University and in University of Le Havre Research Laboratory (France) he worked as Research Assistant... Read More →



Tuesday September 15, 2020 10:35am - 11:05am CDT
Track 4
  HPC  Software
  • See Session Slides yes

10:35am CDT

Open Source Power Availability Tool - Mithun HR & Krishna Harsha Voora, IBM India Pvt Ltd
Necessity to find the open source packages on any platform is ineluctable. The Open Source POWER Availability Tool (OSPAT) is a search engine which helps you find open source packages that are available on the IBM POWER architecture. OSPAT engine analyses opensource repositories, custom repositories having information about IBM Ported packages and returns the requested package information to the user. OSPAT engine makes use of container technology to collect the data and analyses package information. The results displayed will provide the package name, version, Linux distribution that supports the package and a link which displays the package information.

Continue the conversation in Slack

Speakers
avatar for Krishna Harsha Voora

Krishna Harsha Voora

Software Developer - Cognitive Systems, IBM India Pvt. Ltd.
Krishna Harsha Voora is a consultant in UNIX® and Linux operating systems with IBM Systems and Development Lab India. He is actively contributing to Hyperledger Fabric on POWER and he evangelizes Blockchain technology in various forums, conferences, and meet-ups. He also actively... Read More →
avatar for Mithun HR

Mithun HR

Software Developer - Cognitive Systems, IBM India Pvt Ltd
Mithun works as a software developer at IBM Systems and Development Lab India.He works on Opensource Performance projects with IBM Linux on Power ISV ecosystem.



Tuesday September 15, 2020 10:35am - 11:05am CDT
Track 3
  Software
  • See Session Slides yes

11:10am CDT

POWER9 Support in coreboot - Michał Żygowski, 3mdeb Embedded Systems Consulting
coreboot is an open source firmware development framework and its history is reaching 1999. Its primary goal was to boot the machine as fast as possible and launch a Linux kernel. OpenPOWER firmware has a similar goal: initialize the hardware and launch a Linux kernel. coreboot has always been in favor of open hardware and open firmware solutions. The best proof is that the first open firmware implementation for RISC-V appeared right here, in coreboot. Now the time comes for OpenPOWER. This talk will describe the plans of porting the POWER9 architecture to coreboot along with Talos II and Talos II Lite machines. With joint cooperation of 3mdeb Embedded Systems Consulting, Insurgo Technologies Libres/Open Technologies and Raptor Computing Systems this plan may become a reality in the near future. In this presentation, the details of the coreboot port for POWER9 will be discussed covering hostboot, skiboot and petitboot and how they fit into coreboot firmware model.

Continue the conversation in Slack

Speakers
avatar for Michał Żygowski

Michał Żygowski

Firmware Engineer, 3mdeb Embedded Systems Consulting
Firmware Engineer with networking background. Feels comfortable with low-level development using C/C++ and assembly. Interested in advanced hardware features, security and coreboot. Core developer of coreboot. Maintainer of Braswell SoC, PC Engines, Protectli and Libretrend platforms... Read More →



Tuesday September 15, 2020 11:10am - 11:40am CDT
Track 3
  Software
  • See Session Slides yes

11:45am CDT

Competitive Analysis of the Top Gradient Boosting Machine Learning Algorithms - Sai Ayachit, Shyam R, Anubhav Singh & Vinayak Patil, The National Institute of Engineering
Given the rapid increase in computing power and data-driven approaches to tackle many real-world problems today, ML has become an integral part of many solutions.As per a Kaggle survey in 2019, boosting algorithms are among the top 3 preferred methods used by data scientists. Their popularity is due to their robustness against overfitting, faster training times, ability to handle multimodal data while leaving a small memory footprint. In this paper, we compare four state-of-the-art gradient boosting algorithms XGBoost, CatBoost, LightGBM, and SnapBoost on 4 diverse datasets. We perform this competitive analysis on the IBM PowerAI AC922 server. This platform helps end-users experience faster iterations and training than the standard x86. Finally, we present the accuracy and training times of all the algorithms across the 4 datasets. We perform analysis using two approaches; One with only the baseline algorithms, and the other with systematic Hyperparameter Optimization with HyperOpt.

Continue the conversation in Slack

Speakers
SA

Sai Ayachit

Student, The National Institute of Engineering
Sai is a final year undergraduate student at The National Institute of Engineering. Her recent work experience was with IBM under the Global Remote Mentorship(GRM) Program. She made significant contribution to the proposal "GENDER IDENTIFICATION OF SILKWORM PUPAE USING TRANSFER LEARNING... Read More →
SR

Shyam R

Student, The National Institute of Engineering
Shyam is a final year computer science undergraduate at The National Institute of Engineering. He loves to solve challenging problems and explore various Tech-stacks. He is passionate about  Predictive Modelling, Deep Neural Networks, Adversarial models, and Inferential statistics... Read More →
AS

Anubhav Singh

Student, The National Institute of Engineering
Anubhav is a CS undergraduate from The National Institute of Engineering. His interests lie in the emerging technologies like AI and blockchain and competitive coding. He strives to find an optimized solution for real world use cases. His competitive spirit has helped him excel in... Read More →
VP

Vinayak Patil

Student, The National Institute of Engineering
Vinayak is majoring in Computer Science and Engineering at NIE, Mysuru. He likes to code and learn new technologies. He has expertise in various programming languages, Data Structures, and Algorithms. His work experience spans across various domains like Machine Learning, Deep Learning... Read More →



Tuesday September 15, 2020 11:45am - 12:15pm CDT
Track 4
  AI  Software
  • See Session Slides yes

12:45pm CDT

LinuxKI on POWER: The kernel Performance Toolkit Porting Experience and Challenges - Sergey Kachkin, YADRO
There are a lot of Linux performance tools. Some of them are already working on the OpenPOWER platform, while some are not properly ported yet. The LinuxKI Toolset is an open source performance troubleshooting tool for Linux. It is designed to identify performance issues beyond the typical performance metrics and results in faster root cause identification for many performance issues. In this presentation, I'll share the porting experience of this kernel performance tool on OpenPOWER, as well as address challenges and techniques that can be used to port modules if you are a performance engineer and not a professional programmer.

Continue the conversation in Slack

Speakers
avatar for Sergey Kachkin

Sergey Kachkin

Performance Engineer, YADRO
Sergey is a performance engineer with YADRO since 2016. He has more than 10 years of overall experience in dealing with OS performance problems, mission-critical systems slowdown issues and kernel memory dump analysis in Linux and HP-UX. His past experience also includes tuning and... Read More →



Tuesday September 15, 2020 12:45pm - 1:15pm CDT
Track 3
  Software
  • See Session Slides yes

12:45pm CDT

Towards a Formally-verified Software Toolchain for open ISAs - Boris Shingarov, LabWare
This talk summarizes lessons from several years of experimenting with automatic generation of compiler toolchains from processor description languages (PDL), using approaches ranging from superoptimization via logic programming to SMT-based binary analysis, and presents a new experimental system which builds on these lessons. The new approach takes ISA specifications written in the Sail PDL, as the starting point for software-tools synthesis and verification. Sail is used by practicing engineers to create full-scale formalizations of real ISAs; on the other hand, Sail is based on Lem and is mathematically rigorous. This talk demonstrates how the tool transforms a short sequence of POWER instructions into a proof of an invariant of the final state holding universally over the space of allowed behaviors.

Continue the conversation in Slack

Speakers
avatar for Boris Shingarov

Boris Shingarov

Open-source VM designer, LabWare
Boris Shingarov designs open-source dynamic language VMs at LabWare. Currently Boris works on formal proof of correctness of JIT compilers, particularly for open-source ISAs. In the past, Boris has contributed to the creation of large-scale open-source software ecosystems through... Read More →


Tuesday September 15, 2020 12:45pm - 1:15pm CDT
Track 1

1:20pm CDT

Quality Assurance and Coding Standards for Parallel Software on POWER systems - Manuel, Appentra Solutions
The automation of testing is critical in software development to improve quality assurance (QA), but today 80% of testing is manual (Gartner) and $32b is spent annually on QA (IDC/Nelson Hall). Coding standards in automotive and cybersecurity (e.g. CWE, MISRA) provide developers with rules and recommendations to prevent faulty code patterns. Parallel programming is challenging and, thus, QA and coding standards are critical for HPC programmers, particularly for industry and novice programmers. The ever-increasing complexity of HPC software and hardware pushes the developers to critically reevaluate testing methods, but, there is no coding standard for parallel programming yet. We will present an ongoing effort to collect the best practices of parallel programming experts and to develop new static code analysis tools specialized in parallelism to ensure code compliance on POWER systems.

Continue the conversation in Slack

Speakers
avatar for Dr. Manuel Arenaz

Dr. Manuel Arenaz

Arenaz, Appentra Solutions
Dr. Manuel Arenaz is the CEO of APPENTRA Solutions and professor at the University of A Coruña (Spain). He holds a PhD in Computer Science from the University of A Coruña (2003) on advanced compiler techniques for parallelisation of scientific codes. He is passionate about technology... Read More →



Tuesday September 15, 2020 1:20pm - 1:50pm CDT
Track 5B
  HPC  Software
  • See Session Slides yes

1:20pm CDT

How Easy is it to Run on ppc64le? - Toshaan Bharvani, VanTosh
This presentation is about how easy it is to migrate workloads on OpenPOWER systems, mainly looking at the ppc64le architecture. Firstly we will look at some examples of how difficult is is to migrate existing open source software to ppc64le. In the second part, some optimizations will be looked at, and how these optimizations can lead to better performance and the effort to this is not as difficult as expected.

Continue the conversation in Slack

Speakers
TB

Toshaan Bharvani

CTO, VanTosh
Toshaan Bharvani is a IT consultant, currently self-employed at VanTosh, with a interest in Open Source Software and Open Source Hardware. He started his IT interest at a very early age, when his father gave him his first own PC components. Ever since he has been interested in IT... Read More →



Tuesday September 15, 2020 1:20pm - 1:50pm CDT
Track 3
  Software
  • See Session Slides yes

1:55pm CDT

E4S: Extreme-scale Scientific Software Stack - Sameer Shende, University of Oregon
The DOE Exascale Computing Project (ECP) Software Technology focus area is developing an HPC software ecosystem that will enable the efficient and performant execution of exascale applications. The Extreme-scale Scientific Software Stack (E4S) [https://e4s.io], is a coherent software stack that will enable application developers to write parallel applications that can target diverse exascale architectures. E4S provides both source builds through the Spack platform and a set of containers that feature a broad collection of HPC software packages. E4S exists to accelerate the development, deployment, and use of HPC software, lowering the barriers for HPC users. It provides container images, build manifests, and turn-key, from-source builds of popular HPC software packages developed as Software Development Kits (SDKs). It will introduce the E4S containers are being deployed at the HPC systems at DOE national laboratories using Singularity, Shifter, and Charliecloud container runtimes.

Continue the conversation in Slack

Speakers
avatar for Sameer Shende

Sameer Shende

Professor, University of Oregon
Sameer Shende serves as the Director of the Performance Research Laboratory at the University of Oregon and the President and Director of ParaTools, Inc. in the USA, ParaTools, SAS in France, and ParaTools, Ltd. in the UK. He is the lead developer of the TAU Performance System(R... Read More →



Tuesday September 15, 2020 1:55pm - 2:25pm CDT
Track 5B
  HPC  Software
  • See Session Slides yes

1:55pm CDT

Setting up the PCIe hotplug in Kernel for Flexible Setups - Sergey Miroshnichenko, YADRO
It is relatively easy to tune the current Linux kernel via command line arguments to handle hot-added PCI devices if the expected topology is known and limited: simply reserve enough address space and bus numbers per hotplug slot, so new devices can fit in there. However, if the system demands online, substantial and frequent reshaping of a current topology, PCIe's constrained nature can apply severe restrictions. In this presentation, we'll describe how Linux can reallocate resources by pausing the affected drivers and instructing them to update the changes, even in challenging situations like hot-adding an array full of diverse devices in the middle of an existing PCIe tree. Proposed improvements allow switching device clusters via PCIe without any special prerequisites: no need to reboot the host OS, while a Hot-Plug Controller is no longer required to reserve bus numbers and memory regions by firmware/BIOS/bootloader. These device clusters can consist of bridges full of other bridges, NVME drives, GPUs, and many other devices.

Continue the conversation in Slack

Speakers
avatar for Sergey Miroshnichenko

Sergey Miroshnichenko

Software Engineer, YADRO



Tuesday September 15, 2020 1:55pm - 2:25pm CDT
Track 3
  Software
  • See Session Slides yes

3:30pm CDT

High-Performance MPI and Deep Learning with Introspection on OpenPOWER Platform - Dhabaleswar K (DK) Panda, X-ScaleSolutions and The Ohio State University & Donglai Dai, X-ScaleSolutions
This talk will focus on high-performance and scalable middleware for Message Passing Interface (MPI) and Deep Learning on OpenPOWER platform with NVIDIA GPGPUs and RDMA-enabled interconnects (InfiniBand and RoCE). The focus will be on two packages with commercial support being available from X-ScaleSolutions. The first package will focus on the OSU MVAPICH2 MPI libraries and their capabilities for high-performance computing with both CPUs (OpenPOWER) and GPUs (NVIDIA). The second package will focus on tight integration between the OSU MVAPICH2-GDR MPI library and the Horovod stack to provide high-performance and scalable Deep Learning (DL) with deep introspection (DI) capabilities for DL frameworks like TensorFlow, PyTorch and MXNet. The DI capabilities allow DL users and runtime developers to easily optimize their DL applications on modern systems. Performance results from the ORNL SUMMIT system (#2nd) and Lassen (#14th) with thousands of GPUs and POWER9 CPUs will be presented.

Continue the conversation in Slack

Speakers
DK

Dhabaleswar K (DK) Panda

Professor and University Distinguished Scholar; Founder and CEO (X-ScaleSolutions), X-ScaleSolutions and The Ohio State University
DD

Donglai Dai

Chief Engineer, X-ScaleSolutions
Dr. Donglai Dai is a Chief Engineer at X-ScaleSolutions and leads company’s R&D team. His current work focuses on developing scalable efficient communication libraries and performance analysis tools for distributed and parallel HPC and deep learning applications on HPC systems... Read More →



Tuesday September 15, 2020 3:30pm - 4:00pm CDT
Track 4
  AI  Software
  • See Session Slides yes

3:30pm CDT

Parallel Biological Sequence Comparison in Large GPU Clusters - Alba Cristina Magalhaes Alves de Melo, University of Brasilia, Brazil
Pairwise Biological Sequence Comparison is an important Bioinformatics application, which is executed many times daily all over the world. In order to accelerate Biological Sequence Comparison applications, GPUs have been used for more than a decade, with very good results. In this talk, we present CUDAlign, a fine-grained multi-GPU parallel strategy to compare huge DNA sequences in hundreds of GPUs. CUDAlign uses parallelogram-shaped parallelism, overlapping of computation and communication, an innovative speculation technique and has pruning capabilities. We show that CUDAlign is able to attain the best performance in the literature for comparison tools with GPUs. We also discuss CUDAlign's energy consumption. We present ongoing work on pruning with multiple GPUS and show some results in an IBM Power9 + NVidia Volta platform. Finally, we discuss open challenges and research directions.

Continue the conversation in Slack

Speakers
avatar for Alba Cristina Magalhaes Alves de Melo

Alba Cristina Magalhaes Alves de Melo

Professor, University of Brasilia, Brazil
Alba Cristina Magalhaes Alves de Melo obtained her PhD in Computer Science from the Institut National Polytechnique de Grenoble (INPG), France, in 1996. Since 1997, she works at the Department of Computer Science at the University of Brasilia, Brazil, where she is now Full Professor... Read More →



Tuesday September 15, 2020 3:30pm - 4:00pm CDT
Track 5B
  HPC  Software
  • See Session Slides yes
 
  • Timezone
  • Filter By Venue Virtual Event
  • Filter By Type
  • AI
  • Break
  • Cloud
  • Community
  • Cores
  • HPC
  • ISA
  • Keynote Session
  • OpenCAPI
  • Silicon Dev
  • Software
  • Use Case
  • See Session Slides