Network Functions Virtualization Product Guide
Overview of the Network Functions Virtualization (NFV)
Red Hat OpenStack Platform provides the foundation to build a private or public Infrastructure-as-a-Service (IaaS) cloud on top of Red Hat Enterprise Linux. It offers a massively scalable, fault-tolerant platform for the development of cloud-enabled workloads.
Network Functions Virtualization (NFV) uses virtualization to move network node functions into building blocks that interconnect to create communication services. NFV is a new way to define, create, and manage networks by replacing dedicated hardware appliances with software and automation.
This guide briefly discusses Red Hat’s effort to accelerate NFV deployments using the Red Hat OpenStack Platform.
Chapter 1. Understanding Red Hat Network Functions Virtualization (NFV)
Network Functions Virtualization (NFV) is a software-based solution that helps the Communication Service Providers (CSPs) to move beyond the traditional, proprietary hardware to achieve greater efficiency and agility while reducing the operational costs.
An NFV environment allows for IT and network convergence by providing a virtualised infrastructure using the standard virtualization technologies that run on standard hardware devices such as switches, routers, and storage to virtualize network functions. The management and orchestration logic deploys and sustains these services. NFV also includes a Systems Administration, Automation and Life-Cycle Management that reduces the manual work necessary. It ensures that people use modern tools, for example, devops, to work faster and at a larger scale.
1.1. Advantages of NFV
The main advantages of implementing NFV are as follows:
- Accelerates the time-to-market by allowing you to to quickly deploy and scale new networking services to address changing demands.
- Supports innovation by enabling service developers to self-manage their resources and prototype using the same platform that will be used in production.
- Addresses customer demands in hours or minutes instead of weeks or days, without sacrificing security or performance.
- Reduces capital expenditure because it uses commodity-off-the-shelf hardware instead of expensive tailor-made equipment.
- Reduces operational costs by streamlined operations and automation that optimizes day-to-day tasks and improves employee productivity.
1.2. Supported Configurations for NFV Deployments
Red Hat OpenStack Platform supports NFV deployments with the inclusion of automated OVS-DPDK and SR-IOV configuration. Customers looking for a Hyper-converged Infrastructure (HCI) solution can now colocate the Compute sub-system with the Red Hat Ceph Storage nodes. This hyper-converged model delivers lower cost of entry, smaller initial deployment footprints, maximized capacity utilization, and more efficient management in NFV use cases. Red Hat OpenStack Platform also supports, with limitations, the deployment of Technology Preview services such as OVS hardware offload. For more information on the support scope for features marked as technology previews, see Technology Preview.
You can now use composable roles to create custom deployment roles. Composable roles allow you to add or remove services from each role.
For more information on the Composable Roles, see Composable Roles and Services.
Chapter 2. Software
2.1. ETSI NFV Architecture
The European Telecommunications Standards Institute (ETSI) is an independent standardization group that develops standards for information and communications technologies (ICT) in Europe.
Network functions virtualization (NFV) focuses on addressing problems involved in using proprietary hardware devices. With NFV, the necessity to install network-specific equipment is reduced, depending upon the use case requirements and economic benefits. The ETSI Industry Specification Group for Network Functions Virtualization (ETSI ISG NFV) sets the requirements, reference architecture, and the infrastructure specifications necessary to ensure virtualized functions are supported.
Red Hat offers an open-source based cloud-optimized solution to help the Communication Service Providers (CSP) to achieve IT and network convergence. Red Hat adds NFV features such as SR-IOV and DPDK-OVS to Red Hat OpenStack.
2.2. NFV ETSI Architecture and Components
In general, NFV platform has the following components:
- Virtualized Network Functions (VNFs) - the software implementation of network functions, such as routers, firewalls, load balancers, broadband gateways, mobile packet processors, servicing nodes, signalling, location services, and so on.
- NFV Infrastructure (NFVi) - the physical resources (compute, storage, network) and the virtualization layer that make up the infrastructure. The network includes the datapath for forwarding packets between virtual machines and across hosts. This allows you to install VNFs without being concerned about the details of the underlying hardware. NFVi forms the foundation of the NFV stack. NFVi supports multi-tenancy and is managed by the Virtual Infrastructure Manager (VIM). Enhanced Platform Awareness (EPA) improves the virtual machine packet forwarding performance (throughput, latency, jitter) by exposing low-level CPU and NIC acceleration components to the VNF.
- NFV Management and Orchestration (MANO) - the management and orchestration layer focuses on all the service management tasks required throughout the life cycle of the VNF. The main goals of MANO is to allow service definition, automation, error-correlation, monitoring, and life cycle of the network functions offered by the operator to its customers, decoupled from the physical infrastructure. This decoupling requires additional layers of management, provided by the Virtual Network Function Manager (VNFM). VNFM manages the life cycle of the virtual machines and VNFs by either interacting directly with them or through the Element Management System (EMS) provided by the VNF vendor. The other important component defined by MANO is the Orchestrator, also known as NFVO. NFVO interfaces with various databases and systems including Operations/Business Support Systems (OSS/BSS) on the top and the VNFM on the bottom. If the NFVO wants to create a new service for a customer, it asks the VNFM to trigger the instantiation of a VNF, which may result in multiple virtual machines.
- Operations and Business Support Systems (OSS/BSS) - provides the essential business function applications, for example, operations support and billing. The OSS/BSS needs to be adapted to NFV, integrating with both legacy systems and the new MANO components. The BSS systems set policies based on service subscriptions and manage reporting and billing.
- Systems Administration, Automation and Life-Cycle Management - manages system administration, automation of the infrastructure components and life cycle of the NFVI platform.
2.3. Red Hat NFV Components
Red Hat’s solution for NFV includes a range of products that can act as the different components of the NFV framework in the ETSI model. The following products from the Red Hat portfolio integrate into an NFV solution:
- Red Hat OpenStack Platform - Supports IT and NFV workloads. The Enhanced Platform Awareness (EPA) features deliver deterministic performance improvements through CPU Pinning, Huge pages, Non-Uniform Memory Access (NUMA) affinity and network adaptors (NICs) supporting SR-IOV and OVS-DPDK.
- Red Hat Enterprise Linux and Red Hat Enterprise Linux Atomic Host - Create virtual machines and containers as VNFs.
- Red Hat Ceph Storage - Provides the the unified elastic and high-performance storage layer for all the needs of the service provider workloads.
- Red Hat JBoss Middleware and OpenShift Enterprise by Red Hat - Optionally provide the ability to modernize the OSS/BSS components.
- Red Hat CloudForms - Provides a VNF manager and presents data from multiple sources, such as the VIM and the NFVi in a unified display.
- Red Hat Satellite and Ansible by Red Hat - Optionally provide enhanced systems administration, automation and life-cycle management.
2.4. Installation Summary
The Red Hat OpenStack Platform director installs and manages a complete OpenStack environment. The director is based on the upstream OpenStack TripleO project, which is an abbreviation for "OpenStack-On-OpenStack". This project takes advantage of the OpenStack components to install a fully operational OpenStack environment; this includes a minimal OpenStack node called the undercloud. The undercloud provisions and controls the overcloud (a series of bare metal systems used as the production OpenStack nodes). The director provides a simple method for installing a complete Red Hat OpenStack Platform environment that is both lean and robust.
For more information on installing the undercloud and overcloud, see Red Hat OpenStack Platform Director Installation and Usage.
To install the NFV features, you need to configure the following additional steps:
Include SR-IOV, and PCI Passthrough parameters to your
network-environment.yamlfile, update the
post-install.yamlfile for CPU tuning, modify the
compute.yamlfile and run the
overcloud_deploy.shscript to deploy the overcloud.
Install the DPDK libraries and drivers for fast packets processing by polling data directly from the NICs. Include the DPDK parameters to your
network-environment.yamlfile, update the
post-install.yamlfiles for CPU tuning, update the
compute.yamlfile to set the bridge with DPDK port, update the
controller.yamlfile to set the bridge and an interface with VLAN configured and run the
overcloud_deploy.shscript to deploy the overcloud.
For required NFV planning guidelines and configuration, see Network Function Virtualization Planning and Configuration Guide.
Chapter 3. Hardware
See Director Installation and Usage for guidance on hardware selection for OpenStack nodes.
For a list of tested NICs for NFV, see Network Adapter Support. Customer Portal login required.
Chapter 4. NFV Data Plane Connectivity
With the introduction of NFV, more networking vendors are starting to implement their traditional devices as VNFs. While the majority of them are looking into virtual machines (VMs), some are also looking at the container-based approach, per design choice. The OpenStack-based solution should be rich and flexible for two primary reasons:
- Application readiness - Network vendors are currently in the process of transforming their devices into VNFs. Different VNFs in the market have different maturity levels; common barriers to this readiness include enabling RESTful interfaces in their APIs, evolving their data models to become stateless, and providing automated management operations. OpenStack should provide a common platform for all.
Broad use-cases - NFV includes a broad range of applications that serve different use cases. For example, Virtual Customer Premise Equipment (vCPE) aims at providing a number of network functions such as routing, firewall, VPN, and NAT at customer premises. Virtual Evolved Packet Core (vEPC), is a cloud architecture that provides a cost-effective platform for the core components of Long-Term Evolution (LTE) network, allowing dynamic provisioning of gateways and mobile endpoints to sustain the increased volumes of data traffic from smartphones and other devices.
These use cases are implemented using different network applications and protocols, and require different connectivity, isolation, and performance characteristics from the infrastructure. It is also common to separate between control plane interfaces and protocols and the actual forwarding plane. OpenStack must be flexible enough to offer different datapath connectivity options.
In principle, there are two common approaches for providing data plane connectivity to virtual machines:
- Direct hardware access bypasses the Linux kernel and provides secure direct memory access (DMA) to the physical Network Interface Card (NIC) using technologies such as PCI Passthrough (denominated SR-IOV PF in OpenStack now) or single root I/O virtualization (SR-IOV) for both Virtual Function (VF) and Physical Function (PF) pass-through.
- Using a virtual switch (vswitch), implemented as a software service of the hypervisor. Virtual machines are connected to the vSwitch using virtual interfaces (vNICs), and the vSwitch is capable of forwarding traffic between virtual machines, as well as between virtual machines and the physical network.
4.1. Fast data path options
Some of the fast data path options are as follows:
- Single Root I/O Virtualization (SR-IOV) is a standard that makes a single PCI hardware device appear as multiple virtual PCI devices. It works by introducing Physical Functions (PFs) that are the full-featured PCIe functions representing the physical hardware ports and Virtual Functions (VFs) that are lightweight functions that can be assigned to the virtual machines. The virtual machines see the VF as a regular NIC that communicates directly with the hardware. NICs support multiple VFs.
- Open vSwitch (OVS) is an open source software switch that is designed to be used as a virtual switch within a virtualized server environment. OVS supports the capabilities of a regular L2-L3 switch and also offers support to the SDN protocols such as OpenFlow to create user-defined overlay networks (for example, VXLAN). OVS uses Linux kernel networking to switch packets between virtual machines and across hosts using physical NIC. OVS now supports connection tracking (Conntrack) and built-in firewall capability. This avoids overhead of Linux bridge with iptables/ebtables. Open vSwitch for Red Hat OpenStack Platform environments offer out-of-box OpenStack Networking (neutron) integration with OVS.
- Data Plane Development Kit (DPDK) consists of a set of libraries and poll mode drivers (PMD) for fast packet processing. It is designed to run mostly in the user-space, enabling applications to perform their own packet processing directly from or to the NIC. DPDK reduces latency and allows more packets to be processed. DPDK Poll Mode Drivers (PMDs) run in busy loop, constantly scanning the NIC ports on host and vNIC ports in guest for arrival of packets.
- DPDK accelerated Open vSwitch (OVS-DPDK) is Open vSwitch bundled with DPDK for a high performance user-space solution with Linux kernel bypass and direct memory access (DMA) to physical NICs. The idea is to replace the standard OVS kernel data path with a DPDK-based data path, creating a user-space vSwitch on the host that uses DPDK internally for its packet forwarding. The advantage of this architecture is that it is mostly transparent to users. The interfaces it exposes, such as OpenFlow, OVSDB, the command line, remain mostly the same.
Chapter 5. NFV Performance Considerations
For an NFV solution to be useful, its virtualized functions must meet or exceed the performance of physical implementations. Red Hat’s virtualization technologies are based on the high-performance Kernel-based Virtual Machine (KVM) hypervisor, common in OpenStack and cloud deployments.
5.1. CPUs and NUMA nodes
Previously, all memory on x86 systems was equally accessible to all CPUs in the system. This resulted in memory access times that were the same regardless of which CPU in the system was performing the operation and was referred to as Uniform Memory Access (UMA).
In Non-Uniform Memory Access (NUMA), system memory is divided into zones called nodes, which are allocated to particular CPUs or sockets. Access to memory that is local to a CPU is faster than memory connected to remote CPUs on that system. Normally, each socket on a NUMA system has a local memory node whose contents can be accessed faster than the memory in the node local to another CPU or the memory on a bus shared by all CPUs.
Similarly, physical NICs are placed in PCI slots on the Compute node hardware. These slots connect to specific CPU sockets that are associated to a particular NUMA node. For optimum performance, connect your datapath NICs to the same NUMA nodes in your CPU configuration (SR-IOV or OVS-DPDK).
The performance impact of NUMA misses are significant, generally starting at a 10% performance hit or higher. Each CPU socket can have multiple CPU cores which are treated as individual CPUs for virtualization purposes.
OpenStack Compute makes smart scheduling and placement decisions when launching instances. Administrators who want to take advantage of these features can create customized performance flavors to target specialized workloads, including NFV and High Performance Computing (HPC).
Background information about NUMA is available in the following article: What is NUMA and how does it work on Linux?
5.2. NUMA node example
The following diagram provides an example of a two-node NUMA system and the way the CPU cores and memory pages are made available:
Remote memory available via Interconnect is accessed only if VM1 from NUMA node 0 has a CPU core in NUMA node 1. In this case, the memory of NUMA node 1 acts as local for the third CPU core of VM1 (for example, if VM1 is allocated with CPU 4 in the diagram above), but at the same time, it acts as remote memory for the other CPU cores of the same VM.
5.3. CPU pinning
CPU pinning is the ability to run a specific virtual machine’s virtual CPU on a specific physical CPU, in a given host. vCPU pinning provides similar advantages to task pinning on bare-metal systems. Since virtual machines run as user space tasks on the host operating system, pinning increases cache efficiency.
See Configure CPU Pinning with NUMA for further information.
Physical memory is segmented into contiguous regions called pages. For efficiency, the system retrieves memory by accessing entire pages instead of individual bytes of memory. To perform this translation, the system looks in the Translation Lookaside Buffers (TLB) that contain the physical to virtual address mappings for the most recently or frequently used pages. If a mapping being searched for is not in the TLB, the processor must iterate through all the page tables to determine the address mappings. This results in a performance penalty. It is therefore preferable to optimise the TLB so as to ensure the target process can avoid the TLB misses as much as possible.
The typical page size in an x86 system is 4KB, with other larger page sizes available. Larger page sizes mean that there are fewer pages overall, and therefore increases the amount of system memory that can have its virtual to physical address translation stored in the TLB. As a result, this lowers the potential for TLB misses, which increases performance. With larger page sizes, there is an increased potential for memory to be wasted as processes must allocate in pages, but not all of the memory is likely required. As a result, choosing a page size is a trade off between providing faster access times by using larger pages and ensuring maximum memory utilization by using smaller pages.
Chapter 6. Finding More Information
The following table includes additional Red Hat documentation for reference:
The Red Hat OpenStack Platform documentation suite can be found here: Red Hat OpenStack Platform Documentation Suite
Table 6.1. List of Available Documentation
Red Hat Enterprise Linux
Red Hat OpenStack Platform is supported on Red Hat Enterprise Linux 7.4. For information on installing Red Hat Enterprise Linux, see the corresponding installation guide at: Red Hat Enterprise Linux Documentation Suite.
Red Hat OpenStack Platform
To install OpenStack components and their dependencies, use the Red Hat OpenStack Platform director. The director uses a basic OpenStack installation as the undercloud to install, configure and manage the OpenStack nodes in the final overcloud. Be aware that you will need one extra host machine for the installation of the undercloud, in addition to the environment necessary for the deployed overcloud. For detailed instructions, see Red Hat OpenStack Platform Director Installation and Usage.
For information on configuring advanced features for a Red Hat OpenStack Platform enterprise environment using the Red Hat OpenStack Platform director such as network isolation, storage configuration, SSL communication, and general configuration method, see Advanced Overcloud Customization.
For more details on planning and configuring your Red Hat OpenStack Platform deployment with SR-IOV and OVS-DPDK, see Network Function Virtualization Planning and Configuration Guide.