The world is increasingly moving to a future based on cloud computing. This Fourth Industrial Revolution will feature major technology advances fuelled by data – the electricity of our age – as its fundamental cornerstone. Microsoft is building a global trusted infrastructure to power this new engine of transformation as part of our mission to empower every person and every organization on the planet to achieve more.
Operating one of the largest public clouds, Microsoft constantly pushes forward to drive innovation. We need to keep pace with tremendous growth in the cloud, support a broad spectrum of workloads including emerging cloud services, and enable easy scaling across global datacenter regions.
To accomplish this, our innovation needs to expand beyond developing new hardware to the transform the model for open source hardware development. Project Olympus is envisioned to usher in a new era of open source hardware development at cloud speed with the OCP community.
We’ll reveal the benefits of this new model to create innovation through an iterative design process and spark a development and manufacturing ecosystem optimized for efficiency and cost.
For Microsoft, innovation + open source hardware = community collaboration at cloud speed.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=665a4e46c2a0a9874e96534ce17da835
In this session we will hear from Yahoo Japan America/ Yahoo Japan and their adoption of OCP, including details about the challenges they’ve faced along the way, the business value they’ve achieved, and their ideas for future enhancements.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=d521b2c8bf48e546e0aff2f610867de7
In 2010, Steve Jobs declared we had entered the post-PC era. Now firmly in the middle of the mobile era, people are primarily interacting with digital information and services using their mobile devices. But what comes next? What will the post-mobile era look like, and what will the implications be for networking infrastructure and the data center? In this future-focused talk, futurist, Steve Brown, will discuss the post-mobile era and what you need to do to be ready for it.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=eaaf69c745b87a58ab3831a5967bfcab
Luis MartinGarcia will talk about how Facebook looks at its network from multiple angles: from the way the network connects people together across the globe, to the individual components of a switch that interconnects servers at a datacenter. The talk will share Facebook’s vision for open hardware and how simple pieces can be used to build networks at a very high scale.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=73304c6de7cc31d6db7d35ad9a7571b1
Mark Shaw, Director, Azure Cloud Hardware Infrastructure, Microsoft Corp. will delve into the innovative designs and details, from critical components to large sub-systems, of Microsoft’s Project Olympus open source hardware initiative. The session will also elaborate on how a full product portfolio and complete ecosystem can be created through community collaboration and the participation of industry partners.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=7f1a969752227328727437484febae00
Based upon the new OCP v2 spec, Hyve has designed a new 48V OCP v2 rack with a single bus bar for the masses. This new rack design, built around the Google/Facebook specification, will result in the highest levels of power efficiency and an overall simplified design. This presentation is designed as an introduction to the new rack design, unique features and more.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=62e3c5c0037b8641b3cef42dac6f36b3
NVMe is an exciting technology and interface. It is one that brings many benefits to both enterprise & client deployments. But it has its own challenges, mainly not all systems are ready to take full advantage of it. We discuss the benefits of the interface and describe different popular form-factors and how large and small enterprise datacenter deployments can maximize utilizing these form-factors.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=bfe64a7548cceb463915c407137a7303
Intel is helping advance the goals of OCP through many innovative collaborations with OCP members to fully leverage the potential of Intel’s broad technology portfolio. Join this session to hear how Intel is collaborating on key OCP designs, and learn how Intel® Rack Scale Design open-standard software is being used by OCP members to solve common system manageability pain points.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=4c929ef89a9e430ff4079ad12961627d
You are an innovator and disruptor. Learn how HPE fits your business to keep you from free fall. We get you to the landing zone first as your open partner.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=9ae0486e2dba422a267751142ec9f3fb
Based upon the new OCP v2 spec, Hyve has designed a new 48V OCP v2 rack with a single bus bar for the masses. This new rack design, built around the Google/Facebook specification, will result in the highest levels of power efficiency and an overall simplified design. This track is designed to be an interactive Q&A discussion panel, moderated by Amir Michael from Salesforce and will include speakers from Hyve Solutions, Google and Facebook
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=312a50089d4930cbca7fcf836b2000fe
Leendert van Doorn, Distinguished Engineer, Microsoft Azure, Microsoft Corp. will address the intersection between innovations in silicon technology and the resultant impacts on power, performance and efficiency. Innovations incorporated in Microsoft’s Project Olympus enable a broad spectrum of advanced workloads and support cutting-edge capabilities in delivering cloud services.
Microsoft has been building and managing datacenters for decades, during which time we’ve been a driving force in innovation across design, operations, and service delivery.
Our open source journey has shared the designs and technologies that power our own world-class cloud infrastructure, to enable an open, community-driven approach to innovation, and develop an industry- wide ecosystem via OCP.
The need to keep pace with tremendous growth in the cloud is clear. Equally important is the ability to deliver unique services and support a broad spectrum of workloads including those that are likely to spike in the coming years, given advances in fields like machine learning and AI.
In addition to keeping pace with the tremendous growth in the cloud, cloud infrastructure needs to balance power, performance and efficiency to deliver a broad spectrum of advanced workloads and anticipate emerging cloud services yet to be developed.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=5e232f151532e2af3b9754af4ca79f0b
Equinix and Flex explain how they work together to take the friction out of large OCP cloud deployments all over the world by standing up data center hardware quickly and efficiently.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=966a63b0cb8a6df2a217fe4f11ef1373
QCT (Quanta Cloud Technology) has been #1 in OCP since day 1, with the industry’s most diverse lineup of products and integrated solutions powered by OCP standards. We will present an overview of the latest additions to our lineup, including:
· the SXM2 GPU platform and Big Basin powering AI applications
· a multi-purpose cubby NVMe storage solution with Lightning Bolt
· Intel’s Rack Scale Design concept for Open Rack v2
· the Carmel 2.0 19” high-density OCP compliant server
Attendees of this session will learn how QCT’s Rackgo X solutions offer optimized choices for many different applications in the data center, all powered by the ultra-efficient 21” Open Rack v2 specification, with a single bus bar design. It’s ideal for cloud service providers and large enterprise datacenters. You’ll learn how the Rackgo X Leopard Cave, Big Basin, and Lightning Bolt solutions are each designed to fill different roles in your OCP datacenter strategy.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=405838c6caeea4b35812341cc4653096
This year, Wiwynn brings new life to the torrential evolution of the data center. Just like the sprouting leaves of a tree, Wiwynn’s latest cutting-edge OCP products, or “leaves,” will once again showcase Wiwynn’s front-charging role in the OCP ecosystem. Spanning high performance hardware and software products in the 19” and 21” worlds, come witness the full breadth of OCP technologies. And finally, Wiwynn introduces a special, successful case study with a well-renowned customer in helping complete the customer’s first OCP migration that demonstrates Wiwynn and its Partners’ commitment to planting the future of OCP.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=a57b13fbca22830ac4d0364d6a42feff
Join us for a discussion around innovation and using open strategies to lower your TCO without sacrificing customer satisfaction and revenue. HPE recognizes the challenges and risks of moving to an open environment. We understand the needs of bold innovators like you and are your trusted partner.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=7f2369a5229ec6755c080dda8096a3b1
As Solid-State Drives (SSDs) become commonplace in data-centers and storage arrays, there is a growing demand for predictable latency. Traditional SSDs, serving block I/Os, fail to meet this demand. They offer a high-level of abstraction at the cost of unpredictable performance and suboptimal resource utilization. We propose that SSD management trade-offs should be handled through Open-Channel SSDs, a new class of SSDs, that give hosts control over their internals. We present our experience building LightNVM, the Open-Channel SSD subsystem. We introduce a new Physical Page Address I/O interface that exposes SSD parallelism and storage media characteristics. LightNVM integrates into traditional storage stacks, while also enabling storage engines to take advantage of the new I/O interface. Our experimental results demonstrate that LightNVM has modest host overhead, that it can be tuned to limit read latency variability and that it can be customized to achieve predictable I/O latencies.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=b64e246e5b62848390c406b68cf23718
This presentation highlights the achievements made in the OCP networking group over the last four years, overview of that latest advancements in Open Networking, and a glimpse into the future of Open Networking.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=412f0b957b68d17cf8291bc82e1fbf97
Technology leaders from Flex and Nokia will discuss the unique needs of telecommunications data centers in the context of OCP and the work necessary to deploy these data centers quickly and successfully all over the world.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=139cf99e61dcbe4708a7f545b1e78ed9
Escalating semiconductor development costs and protracted development cycles have forced Ethernet to evolve at a snail’s pace, creating a one-size-fits-all solution in a cadence of two years. Hence, the migration from 3.2T to 12.8T is a six year-long cycle. Contrary to that, modularity offers mix and match options at the silicon level, disaggregating the capacity from the I/O and accelerating time to deployment. Our first modular solution based on port extender technology will significantly reduce cost and power, while minimizing operational costs. Customers will regain choice and design optimization, and can craft a modular and effective configuration for each port density need.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=f9f7bda0baef79258a3162ef8d8eb570
OCP has been the driving force behind open networking for servers, storage, and switches. Open collaboration has driven innovative designs for disaggregated servers and storage platforms. Through close collaboration with the community on open networking initiatives such as Cumulus, Switch Abstraction Interface (SAI), and Microsoft SONiC, Mellanox has helped bring the vision of Open Ethernet to a reality. This empowers users to unlock performance and unleash innovation with an open platform. In this talk we will showcase some of the key innovations that we continue to bring to OCP with other open networking vendor. The session will provide updates on Open-source network operating systems for 25, 50, and 100G switches and highlight advanced networking solutions for new server and CPU architectures that will help the community achieve OCP Networking Done Right!
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=ac2edf57d5bb66d3db16849cd06f2f12
The AMD “Naples” System-on-Chip delivers real innovation to better address the needs of existing and emerging data center workloads. With industry leading Core Count, Memory Bandwidth and unprecedented I/O “Naples” sets a new standard of performance, scalability and balance for the modern Data Center. Come experience how “Naples” is helping drive innovation through open source hardware design leveraging open ecosystem.
Intel® Rack Scale Design (Intel® RSD) is a logical architecture whose open standard APIs offer infrastructure manageability that is compatible with the hardware designs being contributed and adopted by OCP members. Come hear a technical deep dive about how Intel® RSD supports the explosion of workload types and use cases in the modern data center and benefits the OCP community.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=2f1fa6139b5c93b426fc019632e5bf80
Dolly Wu, GM of Inspur US Region and Sales, will review the leading standards in Open Data Center Projects today (OCP, ODCC/Scorpio and Open 19) and announce Inspur OCP ORv2 product line specs, availability and schedule. The session will also focus on Inspur’s participation in all Open Data Center communities and driving convergence in design to bring value, optimization and Open Designs for the Hyperscale Datacenters and Cloud Deployments.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=d9a1bcdc7d28b7230df6d3e9ad4df233
Running a data center is now an economic choice, thanks to high quality public infrastructure offerings from global cloud players. For those who wish to operate bare metal themselves, it is essential to raise the efficiency of operation - in terms of the cost of skills, the speed of change execution, and the precision of the changes implemented. Automation of every aspect of hardware commissioning, allocation, deployment, configuration and monitoring is key to bringing data center operations in line with public cloud efficiency. In this presentation Mark will outline the convergence of network and compute provisioning and operations, as well as the state of the art of open source data center operations APIs.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=f409cc1fc0db4c4e2c339d79be56db3e
OCP HAPPY HOUR
Connect with your OCP community!
When: Wednesday, March 8, from 5:30pm – 7:00pm
Where: Sponsor Expo Halls A & B
What: Music, cocktails, hors d’oeuvres & networking!
Learn more about the AMD “Naples” System-on-Chip that sets a new standard of performance, scalability and balance for the Software-Defined Data Center.
Switch Abstraction Interface (SAI) provides simple, consistent APIs for the upper network stack to consume the underlying complex, heterogeneous hardware easily and faster. Multiple popular NOS/network control stack are running on top of SAI, as a fundamental element of disaggregation. We are excited to announce SAI 1.0 release. Architects from major silicon companies will be on the stage to talk about advanced capabilities added/will be added to SAI and use cases lit up through these APIs.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=f555cbdbdc24c9f68d3c5b86b6ae4138
Dr. Know, a new EIA 19” 3U high performance compute accelerator, allows connection from one to four servers. For the 3U four drawers architecture, there are 16 high performance compute accelerator in 4 removable drawers and are compatible with the latest compute acceleration architectures. The system provides higher computing performance and reliability to complete a large number of computing tasks. In addition to efficiency and reliability, Dr. Know also shows application flexibility in computing power sharing and various rack supporting. Furthermore, its tool-less concept eases maintenance of GPU cards, fan modules and PSUs.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=6551fc367e20df0d099ca15f569065b1
Running one of the largest clouds in the world, Microsoft has gained a lot of insight into building and managing a global, high performance, highly available and secure network. We built the these intelligence into Software for Open Networking in the Cloud (SONiC), which thrive for network switch operations and management. Dave Maltz, Distinguished Engineer from Microsoft, will deep dive into the design philosophy underneath SONiC, e.g. containerization, orchestration and how they enable the fast evolution of hyperscale networking.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=726804ae4b1309d606159b65a09b4124
Radisys DCEngine Hyperscale NFV Platform, based on the CG-OpenRack-19 specification, provides datacenter compute and storage infrastructure for carrier networks. The DCEngine Platform supports Intel Rack Scale Architecture, Redfish API and open source based rack management. We would be interested in sharing our work and on future collaboration.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=52c1aafb70058c20f76f3418143a1ca1
http://files.opencompute.org/oc/public.php?service=files&t=232c44b30b059930eba1b1b3c810cf7b
Network security and validating the software running on data center systems has never been more important.
The central components for launching trusted operating system environments are the family of technologies described by the Universal Extensible Firmware Interface (UEFI) Secure Boot and the Trusted Computer Group (TCG) Trusted Platform Module (TPM) specifications.
This talk discusses the plans of the Open Network Install Environment (ONIE) boot loader project to address the challenges of deploying secure technologies on network switches.
The goal is to combine the easy to use image discovery features of ONIE with the industry standard security components of UEFI and TCG.
The benefits are:
Maintains the established image discovery methods of ONIE
Build on industry standard specifications and technologies
Support for x86_64 and AARCH64 UEFI firmwares
Enables all UEFI capabilities
Wiwynn shall present the leading OCP and EIA 19” NVMe storage solutions integrated the NVMe JBOF - Thunder. Design wise, Thunder supports up to 2 hosts with total 24 hot-pluggable U.2 NVMe SSDs, tool-less drawer pull-out function, PSU & cool redundancy, and front (or rear) I/O access. These features make infrastructure design more flexible and easy to maintain. Through the OCP concept design, the SSD, PSUs, Fans and Cables of Thunder are easily accessible and replaceable.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=e91ad18e68dd11a116e8dd52a5f721d4
This presentation will cover Nokia experience in HW management, focusing on OCP HW. We will discuss our backgrounds and incentives, what choices we have made and why. We also elaborate HW management interfaces and related improvement needs.
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=00f83cc5cf4ac62d8d618687a0eefb9a
With support from the OCP Storage Lead, Google, Microsoft and Facebook have been working together through the OCP collaboration principles to set the foundation and process that can accelerate the implementation and adoption of Next Generation HDDs in the Datacenter. (55mins)
Link to slides: http://files.opencompute.org/oc/public.php?service=files&t=ec4e0030cbdf0f2e45362d1bc3c6326a