August, 2018
Agenda Introduction Network functions virtualization (NFV) promise and mission cloud native approach Where do we want to go with NFV? 2
Miroslaw Walukiewicz I m from Gdansk, Poland. 25 years in networking industry - 18 years working for Intel Now a system architect in Intel Programmable Solutions Group (formerly Altera) I like looking for solutions in unusual places. I m married and I have two adult sons. My hobby is history of art = galleries, ruins, and ruins... 3
NFV: Market Transformation Network Infrastructure Physical Appliances Flexible Cloud Infrastructure Commercial Off-the-Shelf (COTS) Servers vbng vepc vipsec vlb NFV Built with a combination of highperformance and embedded CPUs, ASICs, ASSPs, NPUs, and FPGAs Uniform, scalable, and programmable control + data plane 4
Need for Network Functions Virtualization (NFV) Single application on dedicated hardware and proprietary management Firewall EPC Router Enables efficient scaling, faster TTM, Lower TCO, Flexibility Decoupled software on standard x86 server hardware solution agnostic management Firewall VM vepc vrouter Management & Orchestration > _ HW Accelerators TEM/OEM Proprietary OS ASIC, DSP, FPGA, ASSP Difficult to scale with growing demands Compute Storage Networking Chipset Acceleration FPGA AI Chip GPU
NFV: Market Transformation Network Infrastructure Physical Appliances vepc vbng Optional slide to replace slide 5 if you don t have rights to use the images. Flexible Cloud Infrastructure Commercial Off-the-Shelf (COTS) Servers vipsec vlb NFV Built with a combination of highperformance and embedded CPUs, ASICs, ASSPs, NPUs, and FPGAs Uniform, scalable, and programmable control + data plane 6
Data Center Considerations Performance & Capabilities Operations Per Second Low Latency Scalable Total Cost of Ownership Compute Efficiency Power Consumption Utilization Management 7
Toward Cloud Native Solution Scalability Power or core optimization for example, daily traffic change and vepc scaling in/out Massive Internet of Things (IoT) with large number of attachments and Mobility Management Entity (MME) or NRF scaling High Availability Fast failure detection and recovery mechanism Stateless 8
Evolve Packet Core (EPC) SGSN S4 Control Plane IP traffic growing at up to 30-50% per year S1-MME S3 MME S6a S11 HSS Gx/S7 PCRF Rx Data Plane User Data Network Signaling Gi-LAN, IMS, UE enodeb S1-U Serving S5/ S8 PDN Gi Operator IP Gateway Gateway services Internet S2a/c S2b epdg Trusted non-3gpp IP Access Un-Trusted non- 3GPP IP Access The Evolved Packet Core provides a common subscriber anchor for mobility, billing, policy, and charging. IP-based core network infrastructure that provides packet data services to support Long Term Evolution (LTE) Radio Access Network (RAN). EPC consists of several integrated functions that allow for convergence of legacy licensed radio technologies (2G/3G/4G), unlicensed (WiFi) and integration with wireline or other alternative networks to enable IP-based communications and services over both wireless or wireline networks. 9
VNF Virtual Links Between VNF and VNFC for EPC 10
Virtual Network Design Inside Data Center, L2/LAG Virtual Link Spine Spine Spine IP tunnel VxLAN/GRE/NSH Leaf Leaf Leaf Leaf ToR ToR ToR ToR App 1 App2 App 3 vnic vnic vnic vnic SDN vswitch SDN vswitch vtep vtep Virtual Networking IP Net 1 IP Net 3 IP Net 2 host host host IP Net n host App 1 App2 App 3 vnic vnic vnic vnic SDN vswitch SDN vswitch vtep vtep Virtual Networking Rack 1 Rack 2 Rack 3 Rack n 11
Virtual Network Design Inside Data Center, Router Virtual Link BGP/IGP IPV6 network Spine Spine Spine BGP Leaf Leaf Leaf Leaf IP tunnel VxLAN/GRE/NSH vnic is L2 overlay (EVPN) or L3 (L3VPN) ToR ToR ToR ToR App 1 App2 App 3 vnic vnic vnic vnic Router Router vtep vtep Virtual Networking IP Net 1 IP Net 3 IP Net 2 host host host IP Net n host App 1 App2 App 3 vnic vnic vnic vnic Router Router vtep vtep Virtual Networking Rack 1 Rack 2 Rack 3 Rack n 12
Intel FPGA improves vepc performance Key Benefits 1. Performance Enhancement CPU cache optimization Core load balancing through VNF workload offload 2. Scalable Cloud Friendly Solution Scalable networking solution + VNF workload offload 3. Scalable high Density IO Cloud Native Solution 13
Intel FPGA Accelerates NFV Adoption Higher Performance Example: Up to 3X performance improvement as demonstrated for vbng (1) Higher throughputs allowing to exceed system capabilities of a single-server architecture HW and SW Programmable Enable higher throughput, low latency and high bandwidth applications Efficiency Example: Up to 50% lower total power as demonstrated for vbng (1) Optimize data plane performance to achieve lower costs while maintaining a high degree of flexibility. Lower Total Cost of Ownership Versatility Growing ecosystem of Accelerator Functions: vepc, FW, LB, IPSEC High level of customization with easy migration to future networking platforms Support of end to end Industry standard and open source tools Adapt Quickly to Evolving Workloads and Standards Ecosystem Enabling NFV ecosystem partners such as TEMs, VNF vendors, System Integrators and Telcos to bring scalable and highperformance solutions to market. Accelerating NFV Adoption (1) WP: https://h20195.www2.hpe.com/v2/getdocument.aspx?docname=a00019574enw (2) Configuration : The measurements were performed on an HPE DL380*[6] server with three Intel Arria 10 FPGA-based SmartNICs as shown in Figure 6. The DL380 server is a dualsocket server with an E5-2660V3 processor per socket and 256 GB of RAM. Refer to performance notices and disclaimers slide for detailed SW and 14 HW configurations
Legal Notices and Disclaimers Intel technologies features and benefits depend on system configuration and may require enabled hardware, software or service activation. Learn more at intel.com, or from the OEM or retailer. No computer system 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 http://www.intel.com/performance. Tests document performance of components on a particular test, in specific systems. Differences in hardware, software, or configuration will affect actual performance. Consult other sources of information to evaluate performance as you consider your purchase. For more complete information about performance and benchmark results, visit http://www.intel.com/performance. Cost reduction scenarios described are intended as examples of how a given Intel- based product, in the specified circumstances and configurations, may affect future costs and provide cost savings. Circumstances will vary. Intel does not guarantee any costs or cost reduction. Results have been estimated or simulated using internal Intel analysis or architecture simulation or modeling, and provided to you for informational purposes. Any differences in your system hardware, software or configuration may affect your actual performance. Intel does not control or audit third-party benchmark data or the web sites referenced in this document. You should visit the referenced web site and confirm whether referenced data are accurate. Intel, the Intel logo and others are trademarks of Intel Corporation in the U.S. and/or other countries. *Other names and brands may be claimed as the property of others. 2018 Intel Corporation. 15