5G Requirements Require an Architectural Shift
With targeted latency reductions up to 50x, targeted throughput increases up to 100x, and targeted increases in the number of connections up to 100x, the underlying networking and computing infrastructure must transform. By their very nature, network nodes have fixed constraints around power, space, and form factor, yet they face increasing demands for the real-time processing capabilities needed to handle events driven by IoT and other emerging applications.
Meeting these requirements within these constraints will require an architectural shift to a more agile, software-defined network solution, with a significant increase in the compute densities throughout the network to optimize where, when, and how the network traffic is processed. Additionally, emerging latency-sensitive applications and massive amounts of data that require local processing will further drive the demand for compute capability within those fixed constraints.
To balance these competing needs for compute resources, infrastructure will require architecturally compatible solutions optimized for different workloads. The Arm ecosystem enables a range of products, from optimized silicon to flexible systems, that scale from data center and core to edge to access.
Data Center Transformation
Today’s traditional data centers will need to provide services to millions, if not billions, of end users, all while delivering a better level of cost and power efficiency. Core computing will require intelligent offload engines to maximize the general-purpose billable compute cycles available from data center compute racks.
At the same time, many of those services will become increasingly latency sensitive or data processing hungry, pushing the compute closer to the edge, where the data is collected, and the results needed. This shift in where the compute happens means the data center is no longer solely in the center. Or perhaps more appropriately, it expands to the edge where space and power constraints become a severe limitation.
So, new “data center” platforms must offer higher performance, greater compute density, and the ability to scale better in response to elastic demand—all while being optimized to serve their users’ particular business needs.
That’s why data centers seek value in hardware that is energy efficient, optimized for improved performance under a variety of workloads, and can be deployed and managed with maximum efficiency.
Arm IP empowers a broad and varied range of silicon-based solutions that leverage a common software ecosystem to deliver compelling compute density, flexibility, and efficiency versus legacy architectures.
Broad Ecosystem for Infrastructure
The Arm ecosystem encompasses silicon, system, and software elements to provide the capability to solve both current and future needs of these markets. The Arm ecosystem is driving innovation and choice for infrastructure, from data center to network core to edge to access. AIDC is a global directory of solution providers delivering Arm-based infrastructure building blocks and reference solutions. Our community works through open source Linux organizations like Linaro and the Linux foundation, with communities like Open Platform for NFV, and with initiatives like WorksOnArm.com
Major Linux distributions like RHEL, SLES, and Ubuntu are available for Arm-based systems. Key middleware pieces like Java and container technologies like Docker are also being deployed. The Arm approach to standardization is to leverage and influence industry standards like UEFI, ACPI, and DPDK, and to collaborate on the development of new standards within the Arm ecosystem like the Server Base System Architecture.
Semiconductor solutions based on the Arm architecture are available from market-leading companies providing a range of silicon, from high performance server chips to optimized SoCs, to FPGAs, ASICs, and even CoT-based paths to solutions. These are then adopted by OEMs and ODMs who provide systems and software solutions ready for deployment throughout data center, cloud, and network providers.
The current solutions that form the infrastructure edge come from an incredibly diverse ecosystem that is transitioning fast to meet a new set of demands. To help address this transition, Arm has announced Project Cassini, an industry initiative focused on ensuring a cloud-native experience across a diverse and secure edge ecosystem.
Together with our ecosystem partners, Project Cassini will develop platform standards and reference systems upon which cloud-native software stacks can seamlessly be deployed within a standardized Platform Security Architecture (PSA) framework now extended to the infrastructure edge.
Architecting Performance, Efficiency, and Choice in HPC
Arm is changing the landscape of HPC today. Working with silicon partners to bring to market exciting new chips and platform choices with designs targeting the latest and most prevalent HPC applications. Through on-going ecosystem activities, Arm is supporting new production deployments and application optimization initiatives.
Applications such as machine learning and scientific computing are demanding orders-of-magnitude improvements in capacity, capability and efficiency to achieve Exascale computing for next generation deployments. To address this challenge Arm is taking a co-design approach with HPC ecosystem partners — from silicon to system design to application development. Arm offers cross-platform HPC tools, technologies, and expertise to enable, optimize and future proof your designs. Learn more about the Arm HPC ecosystem here.
How The Emerging Compute Continuum Drives Edge Computing
This Forrester Predictions report identifies 2020 as a tipping point for edge computing, driven by the rise of 5G cellular networking, maturing cloud models and relentless improvements in hardware and software as part of a maturing Internet of Things (IoT) infrastructure that now supports any application.
Learn why it’s time to step up investment in Edge Computing.
Accelerating the transformation to a scalable cloud to edge infrastructure.
Empowering the infrastructure to meet next generation throughput demands.
The trusted tools for developing HPC applications on any platform.
- AI will push infrastructure to its limit, but also prove a perfect solution
- Neoverse N1 makes debut in new AWS cloud instances
- Paving the way for AI-enabled supercomputing
- Arm TechCon 2019: Arm Neoverse Then and Now
- Arm TechCon 2019: Showcasing the New Era of Total Compute
- Read More Infrastructure Blogs
- The New Compute Continuum is Powering the Future of IoT
- A New Ecosystem for a New Era in Compute
- Cost Optimize NGINX Plus with Amazon EC2 A1
- Data Will Swamp the Internet, Unless We Think Differently
- How to Manage One Trillion Devices on the Edge
- Empowering the Network Edge with Cisco and Arm
- Improve Performance and ROI with an Arm-Based uCPE Platform