We are just scratching the surface of new consumer experiences, enterprise services and media applications on an intelligent, programmable and flexible, wireless and wireline network infrastructure. Data-rich 5G and edge use cases based on IoT, visual computing, analytics and enhanced mobility reinforce our mission to move data faster, store more of it and process everything.
Network transformation requires a comprehensive approach to compute, storage, and network that includes connectivity, hardware-enabled security and scalable, agile software to support various networking workloads from the network core to the edge. That transformation begins with adoption of Network Function Virtualization (NFV) technologies into the network.
Today Intel shared details about a groundbreaking data centric portfolio of innovations optimized for digital transformation. Fueled by the new, 2nd generation Intel® Xeon® Scalable processors and complementary platform components, communication service providers (CommSPs) can move, store and process data with confidence across the most demanding workloads: from multi-cloud to the intelligent edge and back.
New 2nd Generation Intel® Xeon® Scalable processors
The 2nd generation Intel Xeon Scalable data-centric platform incorporates advanced compute cores, a new memory hierarchy, connectivity and acceleration designed to excel in performance and infrastructure efficiency across a wide range of network-intensive 5G and NFV workloads. The new processor platform delivers up to 1.58X performance improvement for network workloads than the previous generation of Intel Xeon Scalable processors.1
The transformation of CommSP network infrastructure to Intel architecture is already well under way. Whether deployed in existing 4G or newer 5G networks, the new 2nd generation Intel Xeon Scalable processors are ideally suited to drive greater performance from the cloud data center to the network core to the edge.
Success in 5G will be built upon today’s investment in transformed networks. The explosion of data growth, new devices and things and enhanced and emerging new services requires an intelligent network edge. Intel’s vision of the Next Generation Central Office (NGCO) is a game-changing concept because CommSPs can transform traditional central offices (COs) into edge datacenters to support traditional network workloads and emerging, high-value workloads from a single, unified platform. NGCO rack deployments with the new 2nd generation Intel Xeon Scalable solutions will provide the processing horsepower to host all these network functions and services with plenty of headroom to support a new universe of data-centric, 5G services, like cloud gaming, video encoding, live streaming, AI, virtual reality and mixed reality.
For CommSPs that support enterprise services at the network edge, the new 2nd generation Intel Xeon Scalable delivers advanced performance especially in combination with other complementary platform components. The platform can support up to twice the number of subscribers for the virtualized SD-WAN service and up to five times more virtual network function (VNF) capacity when complemented with Intel® Quick Assist Technology and Intel® Ethernet 800 Series.1
More Performance, Power Efficiency for Virtual Network Functions with NFV SKUs
Intel Xeon Scalable processors deliver the processing, storage, memory, virtualization, security, I/O, acceleration, management and analytics capabilities needed for high performance, efficient, scalable and agile 5G-ready networks built on NFV. Our investments are focused on extending our performance leadership to stay several steps ahead of baseline, industry requirements. The newly optimized NFV processor SKUs position CommSPs to manage unprecedented increases in network traffic and new service opportunities.
Intel developed networking and NFV-specialized SKUs (N SKUs) that support virtualized workload acceleration for maximum performance and power efficiency for network functions. This technology offers a higher frequency and capability to tune SKUs for CommSP workloads. The N SKUs are optimized for NFVI deployments across the network and deliver up to 1.58X improved performance across a number of common VNF workloads when compared to first generation Xeon Scalable processors.
The N SKU also offers Intel® Speed Select Technology-Base Frequency to prioritize a few cores for even higher frequency operation which increases the NFVI packet processing performance. By using this capability with the N SKUs, the performance uplift goes from 1.58X to 1.76X1 compared to prior generation Intel Xeon Scalable processors (demonstrated by VPP router with OVS-DPDK performance). CommSPs can run performance critical VNFs on these cores and increase overall system performance. At the same time companies with this optimized server platform have the option to assign lower priority VNFs for certain workloads to operate at the standard frequency.
Intel® Xeon® D-1600 Processors: High-Density, Low Power Envelope for Edge Deployments
The new Intel® Xeon® D-1600 processor is a high-density, highly-integrated system-on-a-chip (SoC) for network environments with low-power and form-factor constraints. This workload-optimized SoC is packed with technologies aimed squarely at network edge, control plane, security and mid-range storage solutions where per core performance is a priority. The Intel Xeon D-1600 processor offers two to eight cores and up to a 29 percent performance improvement over the prior generation of Intel® Xeon® D-1500 processors, network series.2
Delivering Larger, Lower-Cost Memory Subsystems than DRAM
Intel® Optane™ DC persistent memory is a revolutionary memory technology that supports a new class of performance for server systems, which benefit from larger memory subsystems that DRAM alone cannot achieve cost effectively. One networking use case is a Content Delivery Network (CDN) where “hot” live/linear streaming content can now be buffered in server memory using Intel Optane DC persistent memory instead of “cooler” storage. When used in combination with a small amount of classic DRAM, a server can support a higher quality of service and reduce memory costs. This new memory technology is also well-suited for other network applications, including big data network analytics, distributed memory for containerized cloud native infrastructure and resilient NFVI that can be rebooted much faster without reloading memory.
Early Ecosystem Support
Industry transformation requires a dynamic ecosystem of companies to accelerate innovation with a broad range of hardware, software and services. We are honored to announce early support from a number of our key partners to integrate Intel’s new, data-centric technologies into their solutions, including:
- Dell* plans to offer the new 2nd generation Intel Xeon Scalable processors and N SKUs in its PowerEdge Rack Servers;
- Ericsson* shared plans to incorporate the 2nd generation Intel® Xeon® Scalable processors into the Ericsson Software Defined Infrastructure (SDI) system, which is the foundation for its NFVI solutions; and
- Hewlett Packard Enterprise* will integrate new 2nd generation Intel Xeon Scalable processors and N SKUs on HPE ProLiant DL360, DL380, DL560 and DL580 servers, and HPE Synergy 480 and 660 compute modules ; and
- Nokia* will use in its NFV and SDN solutions the new 2nd generation Intel Xeon Scalable processors and has confirmed interest in N SKUs.
Delivering Advanced Performance, NFV Optimizations for Intel® Select Solutions
As part of our ecosystem enablement efforts a number of partners have committed to refresh their respective Intel® Select Solutions for NFVI with the new, 2nd generation Intel® Xeon® Scalable processors and complementary platform components, including the N SKUs, in the coming months. Read more about our updated Intel Select Solutions for NFVI and Intel® Select Solutions for Visual Cloud Delivery Network in a new blog by Renu Navale, Senior Director, Edge Services and Industry Enabling, Network Platforms Group.
We have also developed Network Transformation Experience Kits that provide a variety of best practice guidelines and collateral enabling consumption of 2nd generation Intel Xeon Scalable platforms in next generation multi-vendor, multi-service, enterprise, cloud and CommSP networks.
Preparing for Next-Generation, Data Centric Networks
The new data centric platforms with NFV optimizations and memory advancements deliver performance to scale the most demanding workloads anywhere in the network. We are working with a broad industry ecosystem to deliver a consistent processor architecture across the network to support more seamless software deployments and reduce time-to-market and R&D costs for new services. With the 2nd generation Intel® Xeon® Scalable processors we are exceeding network performance requirements, such as 200 Gbps for 5G mobile core throughput we demonstrated at MWC Barcelona. We will continue to work with the industry to advance the Intel architecture, complementary platform solutions and optimizations to stay ahead of the compute, storage and network requirements for emerging network workloads.
1 Performance results are based on testing by Intel as of 2/04/2019 and may not reflect all publicly available security updates. See configuration disclosure for details. No product or component can be absolutely secure. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit www.intel.com/benchmarks.
2 Compute (SPECrate*2017): 1x Intel® Xeon® D-1623N processor, Platform: Echo Canyon, 2 x 32GB DDR4 2400 ECC(64GB Total Memory ) ,OS: Red Hat 7.6 with Kernel: 3.10.0-957.5.1.el7.x86_64, Bios: GNVDTRL1.86B.0010.D75.1902060802, uCode: 0xe00000c, Benchmark: SPECrate*2017_int_base (Estimated), Compiler: ICC 184.108.40.206, Storage: Intel® SSD D3-S4510 Series 1.92TB, Score: 19.7 (Estimated) compared to 1x Intel® Xeon® D-1513N processor, Platform: Echo Canyon, 2 x 32GB.