Meeting the Challenges of the 5G core with the new 3rd Generation Intel® Xeon® Scalable Processor Family
The relentless growth of data around the world is no secret. As 5G networks proliferate worldwide, data traffic continues to expand. In fact, Ericsson estimates that data traffic will reach 226 exabytes per month by 2026, a massive 450 percent increase from 2021.1
Every day, I see communications service providers (CoSPs) working to transform their networks to deal with this avalanche of data. Data volume, though, is merely the first challenge. CoSPs need versatile, flexible, performant platforms to address the security, power optimization, and sustainability challenges of an increasingly distributed cloud-native 5G core. CoSPs are also looking to improve operational efficiency and service agility in their transition to 5G.
We launched our 3rd Generation Intel® Xeon® Scalable processor family earlier in the year. As the use of web APIs for 5G allows for new areas of potential exposure, the commercial solutions on this platform are already demonstrating cutting-edge performance and addressing the COSP challenges mentioned above—including security.
Performance when and where it counts
Network demands and challenges have grown, and so have the capabilities of our processors and platform. The new 3rd Generation Intel® Xeon® Scalable processor family has the capacity to address nearly any challenge 5G network workloads bring its way—offering higher performance, greater throughput, increased memory, enhanced security, and lower power consumption. These advancements provide a critical foundation for CoSPs future-proofing their networks while also lowering total cost of ownership (TCO).
Our new N SKUs offer the best performance per watt of any of Intel’s processors. They deliver greater throughput for virtual network functions (VNFs) and cloud-native network functions (CNFs), along with lower power consumption for dense or constrained physical deployments. How do they do it? First, we increased instructions per clock (IPC) by 20% from the previous generation.2 Next, we increased the maximum number of cores to 40, with a 20–36 core range in network-optimized processors.
Then, we put the processors to the test to see how they performed on network workloads. 3rd Generation Intel® Xeon® Scalable processors deliver on average up to 62% more performance than the prior generation for a range of broadly deployed network and 5G workloads.3 The 5G user plane function (UPF) achieved a 42% generation-on-generation gain.4 Combined with a set of new instructions to accelerate cryptography, we witnessed performance increases of 88% for single core and 95% system-wide.5
An ecosystem that’s ready for prime time
We are seeing similar performance with our ecosystem partners driving solutions in the industry. For example, ZTE was able to reach 462 Gbps on the 5G core UPF.6 Likewise, NEC achieved 640 Gbps by integrating 2-socket servers fitted with 3rd Generation Intel® Xeon® Scalable processors into their 5G core product.7
Thanks to close collaboration with both partners and CoSPs, Intel has had the benefit of a closer view of new age requirements. Together, we’re innovating and enabling breakthrough performance.
Features designed for power optimization
In addition to performance optimization, operators are constantly looking to reduce the TCO of their network. One way to accomplish this: reducing the power consumption of the network infrastructure. While we were thrilled to see the 42% generation-on-generation UPF performance increase, we were just as excited to share that we achieved a 16% performance-per-watt improvement for that same configuration.8 This includes using Dynamic Device Personalization (DDP), which adds intelligence during packet processing with our network adapters.
Moreover, in this latest generation of the Intel® Xeon® Scalable processor, we provided operators with several features to intelligently program and further optimize their network. First, we added cluster power optimization. This enables system administrators to review all the cores in the system, identify those not in use, and disable them. We also offer server-integrated power controls. This allows use policy-based power control to shift the operating frequency and load in the server cluster to save power at known non-peak hours.
Finally, workload-integrated power controls enable the applications to adjust their performance point dynamically when there is lower traffic throughput for the app to process. This replaces the constant polling model we traditionally had with the Data Plane Development Kit (DPDK). All of these features give operators options to lower the aggregate power usage of the data center cluster.
A holistic, versatile, flexible platform for next-gen frameworks
Whether deployed on customer premises or at the edge, protecting traffic at UPFs without sacrificing performance requires a few other steps. Namely, accelerating Vector Packet Processing (VPP), IPsec forwarding, and reducing the impacts of full data encryption. The new instructions inside 3rd Generation Intel® Xeon® Scalable processors accelerate cryptography and improve security while reducing compute cycles. With the accelerators, you can eliminate the performance impacts of full data encryption.
With the understanding that 5G service-based architecture uses public and private key cryptography for the Transport Layer Security (TLS), we also leveraged Intel® Software Guard Extensions (SGX) in our 3rd Gen Xeon. Intel® SGX helps protect data in use via unique application isolation. We use hardened enclaves to store the private keys that once authorized validated network functions to communicate. The result: a minimized attack surface that protects data and code from malware attacks while allowing every infrastructure element to have a strong security posture.
To make it easier for CoSPs to utilize Intel® SGX without new software development, we teamed up with Red Hat and Fortanix. You can learn more about these efforts in the white paper, “Confidential Computing for 5G Networks.”
Processors built for network workloads
Drawing on decades of innovation and critical partnerships with the world’s leading software and solution providers, we built this new generation of processors specifically for network workloads such as 5G UPF. The net result is a series of N SKUs that offer 1.9 times more performance than the prior generation and a consistent platform from the edge to the core.8
I’m excited for our customers to learn more about these SKUs—and how the new 3rd Generation Intel® Xeon® Scalable processors can provide the flexibility, security, and scale needed to meet the demands of today’s and tomorrow’s 5G networks. Discover more about our innovations in the core by reading our new white paper, “Building the 5G Wireless Core.”
- “Ericsson Mobility Report,” Ericsson, November 2020.
- 20% IPC improvement: 3rd Gen Intel® Xeon® Scalable processor: 1-node, 2x 28-core 3rd Gen Intel® Xeon® Scalable processor, Wilson City platform, 512GB (16 slots /32GB /3200) total DDR4 memory, HT on, ucode=x270, RHEL 8.0, Kernel Version4.18.0-80.el8.x86_64. Test by Intel on 3/30/2021. 2nd Gen Intel® Xeon® Scalable processor: 1-node, 2x 28-core 2nd Gen Intel® Xeon® Scalable processor, Neon City platform, 384GB (12 slots/32GB/2933) total DDR4 memory, HT on, ucode=x2f00, RHEL 8.0, Kernel Version 4.18.0-80.el8.x86_64. Test by
Intel on 3/30/2021. SPECrate2017_int_base (est). Tests at equal frequency, equal uncore frequency, equal compiler.
- 1.62x average network performance gains: geomean of Virtual Broadband Network Gateway, 5G User Plane Function, Virtual Cable Modem Termination System, Vector Packet Processing - Forward Information Base 512B, DPDK L3 Forward 512B, CDN-Live, Vector Packet Processing - IP Security 1420B.
- 1.42x 5G User Plane Function: 1-node, 2(1 socket used)x Intel Xeon Gold 6338N on Whitley Coyote Pass 2U with 128 GB (8 slots/ 16GB/ 2666) total DDR4 memory, ucode 0x261, HT on, Turbo off, Ubuntu 18.04.5 LTS, 4.15.0-134-generic, 1x Intel 810 (Columbiaville), FlexCore 5G UPF, Jan’ 2021 MD5 checksum: c4ad7f8422298ceb69d01e67419ff1c1, GCC 7.5.0, 5G UPF228 Gbps / 294 Gbps, test by Intel on 3/16/2021.
- 1.94x Vector Packet Processing - IP Security 1420B: 1-node, 2(1 socket used)x Intel Xeon Gold 6338N on Intel* Whitley with 128 GB (8 slots/ 16GB/ 2666) total DDR4 memory, ucode 0x261, HT on, Turbo off, Ubuntu 20.04 LTS (Focal Fossa), 5.4.0-40-generic, 1x INTEL* 240G SSD , 1x E810-2CQDA2 (Chapman Beach), v21.01-release, Gcc 9.3.0, VPPIPSEC(24c24t) test by Intel on 3/17/2021.
- “ZTE's High Performance 5G Core Network UPF Implementation Based on 3rd Generation Intel® Xeon® Scalable Processors.” Intel White Paper, March 2021.
- “NEC’s UPF Maximizes 5G Value with High Performance and Flexibility in Containerized, Virtualized or Physical Deployments.” NEC, June, 2021.
8.Average performance based on Geomean of est SPECrate®2017_int_base 1-copy, est SPECrate®2017_fp_base 1-copy, est SPECrate®2017_int_base, est SPECrate®2017_fp_base, STREAM Triad, Intel distribution of LINPACK, Virtualization and OLTP Database workloads. Results have been estimated or simulated. SPECcpu_2017, STREAM, LINPACK Performance: 1-node, 4x 3rd Gen Intel® Xeon® Platinum 8380H processor (pre-production 28C, 250W) on Intel Reference Platform (Cooper City) with 768 GB (24 slots / 32 GB / 3200) total memory, microcode 0x87000016, HT on for SPECcpu, off for STREAM, LINPACK), Turbo on, with Ubuntu 19.10, 5.3.0-48-generic, 1x Intel 240GB SSD OS Drive, est SPECcpu_2017, STREAM Triad, Intel distribution of LINPACK, test by Intel on 5/15/2020. HammerDB OLTP Database Performance: New: 1-node, 4x 3rd Gen Intel® Xeon® Platinum 8380H processor (pre-production 28C, 250W) on Intel Reference Platform (Cooper City) with 768 GB (24 slots / 32 GB / 3200) total memory, microcode 0x700001b, HT on, Turbo on, with Redhat 8.1, 4.18.0-147.3.1.el8_1.x86_64, 1x Intel 240GB SSD OS Drive, 2x6.4T P4610 for DATA, 2x3.2T P4610 for REDO, 1Gbps NIC, HammerDB 3.2, Popular Commercial Database, test by Intel on 5/13/2020. Virtualization Performance: New: 1-node, 4x 3rd Gen Intel® Xeon® Platinum 8380H processor (pre-production 28C, 250W) on Intel Reference Platform (Cooper City) with 1536 GB (48 slots / 32 GB / 3200 (@2933)) total memory, microcode 0x700001b, HT on, Turbo on, with RHEL-8.1 GA, 4.18.0-147.3.1.el8_1.x86_64, 1x Intel 240GB SSD OS Drive, 4x P4610 3.2TB PCIe NVMe, 4 x 40 GbE x710 dual port, Virtualization workload, test by Intel on 5/20/2020.
Notices & Disclaimers
Performance varies by use, configuration, and other factors. Learn more at www.intel.com/PerformanceIndex.
Performance results are based on testing as of dates shown in configurations and may not reflect all publicly available updates. See backup for configuration details. No product or component can be absolutely secure.
Your costs and results may vary.
Intel technologies may require enabled hardware, software, or service activation.
© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.