Hi [[ session.user.profile.firstName ]]

Leverage the Fastest GPU & Storage Communications for DeepLearning Superclusters

In this session, Supermicro’s Mike Scriber and Rob Davis of NVIDIA will discuss the importance of high-performance storage in a GPU system, and how that storage picture is changing. They will talk about how solutions can range from local direct storage to remote storage using NVMeoF and GPUdirect.
Recorded May 19 2021 40 mins
Your place is confirmed,
we'll send you email reminders
Presented by
Rob Davis, NVIDIA; Mike Scriber, Sr. Director Server Solution Management, Supermicro
Presentation preview: Leverage the Fastest GPU & Storage Communications for DeepLearning Superclusters

Network with like-minded attendees

  • [[ session.user.profile.displayName ]]
    Add a photo
    • [[ session.user.profile.displayName ]]
    • [[ session.user.profile.jobTitle ]]
    • [[ session.user.profile.companyName ]]
    • [[ userProfileTemplateHelper.getLocation(session.user.profile) ]]
  • [[ card.displayName ]]
    • [[ card.displayName ]]
    • [[ card.jobTitle ]]
    • [[ card.companyName ]]
    • [[ userProfileTemplateHelper.getLocation(card) ]]
  • Channel
  • Channel profile
  • The Shape of the Edge in the Data Center Aug 12 2021 4:00 pm UTC 1 min
    Scott M. Fulton III, Contributor, Data Center Knowledge - Vik Malyala, Senior Vice President, Supermicro
    Innovative new and upcoming edge computing applications are re-defining the role of data centers, posing many questions to enterprises and operators alike. Join this webinar to learn about the shape edge computing infrastructure is taking in the data center and what it all means for organizations that have (or will have) applications that need to run at the Edge.
  • 2U 2N Multi-GPU Platform Delivers Unrivaled Energy Efficiency and Flexibility Aug 5 2021 5:00 pm UTC 60 mins
    Eric Frazier, Solutions Manager at Supermicro Fred Hsu, Product Manager at Supermicro
    Supermicro’s new 2U 2-Node system, powered by AMD processors and GPUs is the perfect platform for HPC and AI applications. Lower cost infrastructures can be created for computationally intense workloads. Join this webinar to learn from industry experts about the different use cases and benefits of this Multi-GPU optimized platform including:

    - Delivering up to 10% TCO savings utilizing shared power and cooling
    - Providing excellent serviceability, unlike other existing products in the market
    - Fast and flexible networking capabilities - process massive dataflow for demanding AI/ML applications, deep learning training, and inferencing
  • New Generation Supermicro A+ Servers featuring AMD EPYC™ Processors Aug 5 2021 7:00 am UTC 40 mins
    Dieter Bodemer (System Solution Manager at Supermicro) and Alexey Nechuyatov (Senior Manager at AMD Server Business Unit)
    Join experts from Supermicro and AMD to discuss the performance and efficiency gains for Cloud, AI, and storage workloads, which are being offered by the new generation of servers and storage systems.

    Session highlights include advances in rackmount server technology, featuring an all-new compact model sporting 2 x OCP 3.0 compliant Supermicro Advanced I/O Modules to provide a wide range of networking options and 10 NVMe drive bays enabling vastly more efficient transport of data between storage systems and servers, all in a 1U form factor and perfect for high-efficiency cloud computing environments.

    We will walkthrough Supermicro’s flagship performance multi-node ranges, including the A+ BigTwin and TwinPro for hosting and storage-defined networking clusters; A+ SuperBlade for extreme cost-efficiency in high-density computing, and the A+ FatTwin with front and rear I/O for ultimate serviceability.

    Our technical team will also report on the enormously popular A+ dual-processor GPU servers, which support up to 160 PCI-E 4.0 lanes and contain up to 8 direct-attached GPUs for more performant Artificial Intelligence and Machine Learning frameworks.

    Register now to find out first!

    The presenters will host a Q&A followed by an interactive quiz* with 3 x Apple SE GPS watches up for grabs! DO NOT miss out!

    *Please download the Kahoot! app on your smartphone before the session to access the quiz. Note: No account signup is required.
  • Liquid Cooling Fundamentals for your Datacenter Aug 3 2021 5:00 pm UTC 45 mins
    Dave Martinez, Sandia National Labs.; Mark Ng, Supermicro; Brian Wu, Supermicro; Michael Schulman, Moderator, Supermicro
    Learn how implementing Liquid Cooling can reduce your OPEX by reducing the amount of electricity your data center uses. Get an understanding of the liquid cooling technologies and see how leading organizations are implementing this technology today and preparing for the future.
  • Solution: Red Hat OpenShift Data Foundation & SMCI Servers Recorded: Jul 29 2021 30 mins
    Mayur Shetty, Red Hat; Michael Ocampo & Max Chung Supermicro;
    This presentation will look into how Red Hat OpenShift® Data Foundation (ODF), software-defined storage, is integrated with and optimized for the Red Hat OpenShift Container Platform. We will also look into how OpenShift Data Foundation 4.x is built on Red Hat Ceph® Storage, Rook, and NooBaa to provide container native storage services that support block, file, and object services. OpenShift Data Foundation will deploy anywhere OpenShift does: on-premise or in the public cloud.

    We will also look at how Red Hat ODF supports a variety of traditional and cloud-native workloads, including:

    Block storage for databases and messaging.
    Shared file storage for continuous integration and data aggregation.
    Object storage for archival, backup, and media storage.

    With OpenShift 4.x, Red Hat has rearchitected OpenShift to bring the power of Kubernetes Operators to transform our enterprise-grade Kubernetes distribution with automating complex workflows, i.e., deployment, bootstrapping, configuration, provisioning, scaling, upgrading, monitoring, and resource management. In conjunction, OpenShift Data Foundation and Supermicro servers will transform the storage customer experience by making it easier for Supermicro and Red Hat customers to install, upgrade and manage storage on OpenShift.
  • Leveraging Intel® Optane Technology with VMware Cloud Foundation Recorded: Jul 29 2021 38 mins
    Sridhar Kayathi, Senior Technical Solutions Specialist, Intel®; Sim Upadhyayula, General Manager, Solutions, Supermicro
    Modernize the data center to simplify workload management with VMWare Cloud Foundation and increase workload performance and operational efficiency with Intel Scalable processors and Intel® Optane™ technology.
  • Software Defined Storage for Data Protection Recorded: Jul 29 2021 23 mins
    Shiva Shastri, Sr. Technical Product Manager, Commvault; Paul McLeod, Product Director, Supermicro
    A new networked storage architecture called Software-Defined Storage emerged from the world of cloud computing a few years ago. Called SDS for short, it took the storage world by storm with many startups emerging, followed by investments and acquisitions from the larger storage vendors. But what is SDS? How does it fit into the networked storage landscape? What are its advantages over long time tried and trustworthy solutions like Fibre Channel SANs? Join this session at Supermicro’s Open Storage Summit and find out the answers to these questions and how to deploy SDS with Supermicro solutions.
  • What Enterprises Need to Know When Deploying Container Based Environments Recorded: Jul 29 2021 31 mins
    Eric Burgener, Mayur Shetty, Sim Upadhyayula, Anantha Adiga, Jose Gomez
    We discuss the impact of containerized workflows on server and storage architectures. This session will also talk about the hardware and software selection for converged, hyper converged, and disaggregated storage architectures and the challenges and rewards of SDS for container based workloads (Edge to the Datacenter).

    Eric Burgener, Research Vice President, IDC Analyst
    Mayur Shetty, Principal Solution Architect, Red Hat
    Sim Upadhyayula, General Manager, Solutions, Supermicro
    Anantha Adiga, Cloud Solutions Architect, Intel® Optane Group
    Jose Gomez, Sr. Technical Marketing Engineer – DevOps, Nutanix
  • QuantaStor™ Software-Defined Storage Solution Recorded: Jul 28 2021 28 mins
    Steven Umbehocker, Founder & CEO, OSNEXUS; Sherry Lin, Solution Product Manager, Supermicro
    Overview of a typical scale-out solution using high-density Supermicro nodes
    Overview of a typical scale-up solution using SBB Supermicro nodes
    Overview of pros/cons of using scale-up vs scale-out and what use cases best fit which hardware architecture

    Steven Umbehocker, Founder & CEO, OSNEXUS
    Sherry Lin, Solution Product Manager, Supermicro
  • Software-Defined Durability, Protection and Availability at Massive Scale Recorded: Jul 28 2021 33 mins
    Frederik De Schrijver, ActiveScale Product Manager, Quantum Corporation; Paul McLeod, Product Director, Supermicro
    Not all object storage systems are alike. Designing systems to store terabytes to exabytes of data - for the long term - not only requires scalable performance, but also serious considerations that ensure the durability, protection and availability of both the data and the underlying storage platform. In this session, we will discuss how object storage and scale-out architectures are evolving to meet the future demands of incessant unstructured data growth. Join us to learn how:

    • ActiveScale’s software architecture uniquely solves the challenge of scale and durability
    • How rebalancing prevents some systems from meeting your data growth needs
    • How object storage is a key resource in your fight against ransomware
    • How new storage architectures are driving innovation in genomics, HPC, and AI

    Join Frederik De Schrijver, Quantum, and Paul Mcleod, Supermicro, to learn how we are working together to drive new levels of capability.
  • The Speed of RAID Fused with the Scale of Object Recorded: Jul 28 2021 26 mins
    Paul Evans, Solution Advisor, RozoFS; Damien Zhao, Product Manager, Supermicro
    There is a well-deserved love for the speed and simplicity of RAID, but in a world of impossibly-large HDDs and Multi-PB buckets, there is little RAID can do for us at scale. Unfortunately, the 'classic' Erasure Coding used to replace RAID has brought us mathematical complexity, high costs and significant performance impacts. Is there a way to blend both speed and scale and run it on simple and efficient hardware? Indeed there is: Rozo Systems has adapted a mathematical transform (Mojette) to function as an encode/decode engine mimicking the simplicity of RAID, the integrity of erasure coding, and the scalability and distributed access of object storage. Even better: it works beautifully over SMB3, NFS4, and the Rozo DirectSCALE parallel client. Find out how RozoFS is easy on the hardware, but greased lightning for your applications and users.
  • Hybrid Cloud Deployment Recorded: Jul 28 2021 26 mins
    Phil Hooker, Technical Director, Nebulon; William Li, Sr. Product Manager Supermicro
    Get the benefits of the public cloud experience for any on-prem application, core-to-edge with Supermicro server-embedded, infrastructure software delivered as-a-service
  • How Software-Defined Storage Supports IT Infrastructure Modernization Recorded: Jul 28 2021 31 mins
    Eric Burgener, Alan Johnson, Frederik De Schrijver, Steven Umbehocker, Tobias Flitsch, David Larsen, Paul Evans
    Discussion of how enterprises are transforming storage siloes using cloud enabled SDS for increased agility and resiliency. The power of SDS achieve cloud goals. The impact to IT and what concerns customers may have about doing this.

    Eric Burgener, Research Vice President, IDC Analyst
    Alan Johnson, Director of Emerging Technologies, Supermicro
    Frederik De Schrijver, ActiveScale Product Lead, Quantum Corporation
    Steven Umbehocker, Founder & CEO, OSNEXUS
    Tobias Flitsch, Director of Product Management, Nebulon
    David Larsen, Technical Solutions Specialist, Intel
    Paul Evans, Solution Advisor, RozoFS
  • Supermicro SuperBlade Powering Modern Workloads Recorded: Jul 28 2021 45 mins
    Shanthi Adloori, Director of Product Management, Supermicro
    Today's cloud, HPC, and AI workloads require a new level of compute performance. During this webinar, Director of Product Management Shanthi Adloori explores the results and benefits of the Supermicro A+ SuperBlade, featuring 3rd Gen AMD EPYC processors, powering modern workloads.
  • Worlds fastest File System for HPC, AI and Machine Learning Recorded: Jul 27 2021 19 mins
    Shimon Ben-David, CTO, WekaIO; Alvin Chu, Sr Director of Storage and AI Technologies Solutions, Supermicro
    One of the main challenges of a true scale-out system is effectively scaling the metadata to enable practical scale-out without compromising performance. Traditionally, this is solved using a metadata server that becomes a performance bottleneck or by TOKEN sharing between nodes that becomes less effective as the cluster grows. This session will present how WekaIO, a fully software-defined data platform, solves this issue, allowing for effective scale-out to 1000s of servers and increasing performance as more servers are added to the cluster. Furthermore, this unique approach also allows for multiple configurations options from the Supermicro product portfolio.

    Shimon Ben-David , CTO, Wekaio
    Alvin Chu, Sr Director of Storage and AI Technologies Solutions, Supermicro
  • Solving I/O Bottleneck with DAOS and PMEM Recorded: Jul 27 2021 35 mins
    Andrey Kudryavtesev, DAOS Product Management, Intel®; Jerry Dien, Director of Server Products, Supermicro
    Intel® has been building an entirely open-source software ecosystem for data-centric computing, fully optimized for Intel® architecture and non-volatile memory (NVM) technologies, including Intel® Optane™ DC persistent memory and Intel® Optane™ DC SSDs. Distributed Asynchronous Object Storage (DAOS) is the foundation of the Intel Exascale storage stack. DAOS is an open-source software-defined scale-out object store that provides high bandwidth, low latency, and high I/O operations per second (IOPS) storage containers to HPC applications. In addition, it enables next-generation data-centric workflows that combine simulation, data analytics, and AI.
  • Software Defined Storage Demystified Recorded: Jul 27 2021 42 mins
    Rob Davis, VP of Storage, NVIDIA Networking; Mike Scriber, Sr. Director Server Solution Management, Supermicro
    A new networked storage architecture called Software Defined Storage emerged from the world of cloud computing a few years ago. Called SDS for short, it took the storage world by storm with many startups emerging followed by investments and acquisitions from the larger storage vendors. But what is SDS? How does it fit into the networked storage landscape? What are its advantages over long time tried and true solutions like Fibre Channel SANs? Join this session at Supermicro’s Open Storage Summit and find out not only the answers to these questions, but also how to deploy SDS with Supermicro solutions.
  • Orchestrating Data Storage to Enable New Performance & Cost/Performance Profiles Recorded: Jul 27 2021 20 mins
    Curtis Anderson, Software Architect, Panasas; Josh Grossman, Product Manager, Supermicro
    Along with the growth in unstructured data, the workloads being applied to that data are becoming less structured. This is because clouds have (re)taught us what HPC sites already knew, that compute should be shared for highest utilization. Shared compute requires shared storage and results in many storage workloads being applied at the same time. The explosion of AI/ML brings an entirely new workload that needs to be supported without buying a new storage system.
    No one wants to buy special purpose storage. Instead, storage should serve all your needs without having to sync data across different systems for different uses.
  • How Hardware Enhancements Drive Software-Defined Storage Improvements Recorded: Jul 27 2021 31 mins
    Eric Burgener, IDC; Liran Zvibel, WekaIO; Randy Kreiser, Supermicro; Andrey Kudryavtesev, Intel; Colin Presly, Seagate
    Discussion of the enabling technologies that customers running software-defined storage (SDS) are most interested in (Performance)
    • New storage technology PMEM, NVMe, Dual Actuator HDDs, NVMeOF, Composable, Computational storage,/Smart media, and new offload strategies like DPU
    • New performance milestones
    • Challenges delivering value (rapid ROI)

    Eric Burgener, Research Vice President, IDC Analyst
    Liran Zvibel, CEO and Founder, Wekaio
    Randy Kreiser, Storage Specialist/FAE, Supermicro
    Andrey Kudryavtesev, HPC Storage Architect, Intel.
    Colin Presly, Senior Director, Office of the CTO, Seagate
  • Innovations and Software-Defined Storage Solutions Recorded: Jul 27 2021 42 mins
    Charles Liang, Supermicro; Alper Ilkbahar, Intel®; Rajiv Ramaswami & Tarkan Maner, Nutanix
    The Open Storage Summit will commence with keynote presentations from Nutanix, Intel and Supermicro senior executives on the latest software-defined storage technology developments for enterprise, data center, and multi-cloud infrastructure, and an overview of Supermicro’s expansive storage system product lines.

    Charles Liang, CEO, Supermicro
    Alper Ilkbahar, VP & GM, Data Center Memory and Storage Solutions, Intel®
    Rajiv Ramaswami, CEO, Nutanix
    Tarkan Maner, CCO, Nutanix
Live webcasts on end-to-end green computing solutions.
Supermicro provides end-to-end green computing solutions for the data center, cloud computing, enterprise IT, big data, high performance computing and embedded computing.

Embed in website or blog

Successfully added emails: 0
Remove all
  • Title: Leverage the Fastest GPU & Storage Communications for DeepLearning Superclusters
  • Live at: May 19 2021 5:00 pm
  • Presented by: Rob Davis, NVIDIA; Mike Scriber, Sr. Director Server Solution Management, Supermicro
  • From:
Your email has been sent.
or close