2018 OCP Regional Summit

Advice on best practices, methods, and organizational philsophies for adopting Open Source Hardware. Presented by Aaron Sullivan, who has spent considerable time in Enterprise, Telco, and Hyperscale infrastructure environments, and has been part of the OCP Community since its early days.

Autonomous vehicles are one of today’s most exciting applications. Guided by AI, they have the potential to make roads safer, improve fuel economies, and provide a more relaxing driving experience. That’s the view from the driver’s seat. For data center operators, the widespread use of autonomous vehicles will create unprecedented challenges. High-performance compute, storage and network resource, such as HPE Cloudline portfolio, are needed to process massive volumes of data at low latency to ensure safe and efficient vehicle operation.

As an OCP Platinum solution provider, Wiwynn has been a driving player in the OCP ecosystem since its inception.
Wiwynn continues this commitment by offering the latest 21" OCP products that are igniting the next generation of data centers.
Come learn more about Wiwynn, its explosive growth, and its cutting-edge line of 21" OCP products & more in this exciting talk!

The Project Olympus building blocks provide an excellent foundation to develop and configure server solutions for specific server workloads (compute, cold storage, performance storage, and AI). In addition, working with “known good” building blocks significantly improves the time to market of optimized server solutions. If further HW customization is required, the development of new platforms is streamlined within Project Olympus building blocks. This session provides the details on the 3U PCIe expansion system ZT developed (with the Project Olympus Building Blocks) for NVIDIA GPUs, AMD GPUs, FPGAs, and future PCIe add-in cards.

Hardware specialization is the most effective path to higher performance and efficiency, which is attractive for large scale data centers.  Supporting many different hardware instances to get that specialization for different workloads, however, is expensive.  Each system must be separately qualified, different spares must be stocked, volumes are lower for each type of hardware increasing costs, and so on.  Microsoft achieves the benefits of specialization with common hardware by including an FPGA in every Azure and Bing server.  In this talk, I describe how Microsoft converged on a single FPGA platform that supports cutting edge performance on very different acceleration workloads, specifically accelerated software defined networking for Azure and deep learning inference for Bing.

The core value of open computing is to reduce the technical cost and lower technical threshold, so as to have stronger competitiveness and vitality. The development of open computing is gradually changing the needs of data centers—from traditional standardization to open technology-based customization. Inspur’s vast experience in serving CSPs and our deep understanding of customer application scenarios is augmented by a willingness to share technology and co-design utilizing our JDM model to identify precise customer needs and achieve customized solutions through rapid innovation and productization. During this session Inspur will launch new advanced computing, storage and GPU nodes, optimized software management and superior resource pooling solutions, join this session to know more detail.

The data sphere continues to grow at an exponential pace and will accelerate as new data-intensive applications such as IoT, 5G and autonomous driving become mainstream. Legacy storage architectures lack the scalability and efficiency to support future capacity requirements. New standards, such as NVMe over Fabrics and NF1 form factor, offer a solution here: NVMe over Fabrics unifies network and drive protocols for vastly improved efficiency and when combined with space-optimized NF1 form factor SSDs, enables a scalable high-density, low-latency storage solution.