data center design

Open Networking

Open Networking

In today's digital age, where connectivity is the lifeline of businesses and individuals alike, open networking has emerged as a transformative approach. This blogpost delves into the concept of open networking, its benefits, and its potential to revolutionize the way we connect and communicate.

Open networking refers to a networking model that promotes interoperability, flexibility, and innovation. Unlike traditional closed networks that rely on proprietary systems, open networking embraces open standards, open source software, and open APIs. This approach enables organizations to break free from vendor lock-in, customize their network infrastructure, and foster collaborative development.

Enhanced Agility and Scalability: Open networking empowers businesses to adapt swiftly to changing requirements. By decoupling hardware and software layers, organizations gain the flexibility to scale their networks seamlessly and introduce new services efficiently. This agility is crucial in today's dynamic business landscape.

Cost-Effectiveness: With open networking, businesses can leverage commodity hardware and software-defined solutions, reducing capital expenditures. Moreover, the use of open source software eliminates costly licensing fees, making it an economically viable option for organizations of all sizes.

Interoperability and Vendor Neutrality: Open networking promotes interoperability between different vendors' products, fostering a vendor-neutral environment. This not only frees organizations from vendor lock-in but also encourages healthy competition, driving innovation and ensuring the best solutions for their specific needs.

Data Centers and Cloud Networks: Open networking has found significant applications in data centers and cloud networks. By embracing open standards and software-defined architectures, organizations can create agile and scalable infrastructure, enabling efficient management of virtual resources and enhancing overall performance.

Campus Networks and Enterprise Connectivity: In the realm of campus networks, open networking allows organizations to tailor their network infrastructure to meet specific demands. Through open APIs and programmability, businesses can integrate various systems and applications, enhancing connectivity, security, and productivity.

Telecommunications and Service Providers: Telecommunications and service providers can leverage open networking to deliver innovative services and improve customer experiences. By adopting open source solutions and virtualization, they can enhance network efficiency, reduce costs, and introduce new revenue streams with ease.

Open networking presents a transformative paradigm shift, empowering organizations to unleash the full potential of connectivity. By embracing open standards, flexibility, and collaboration, businesses can achieve enhanced agility, cost-effectiveness, and interoperability. Whether in data centers, campus networks, or telecommunications, open networking opens doors to innovation and empowers organizations to shape their network infrastructure according to their unique needs.

Highlights: Open Networking

**Fostering Innovation**

a) Open Networking refers to a network where networking hardware devices are separated from software code. Enterprises can flexibly choose equipment, software, and networking operating systems (OS) by using open standards and bare-metal hardware. An open network provides flexibility, agility, and programmability.

b) Additionally, open networking effectively separates hardware from software. This approach enhances component compatibility, interoperability, and expandability. In this way, enterprises gain greater flexibility, which facilitates their development.

c) Open networking relies on open standards, which allow for seamless integration between different hardware and software components, regardless of the vendor. This approach not only reduces dependency on single-source suppliers but also encourages a competitive market, fostering innovation and driving down costs.

d) Furthermore, open networking solutions are often built on open-source software, which benefits from the collective expertise of a global community of developers and engineers.

At present, Open Networking is enabled by: 

  • A. Open Source Software 
  • B. Open Network Devices 
  • C. Open Compute Hardware 
  • D. Software Defined Networks 
  • E. Network Function Virtualisation 
  • F. Cloud Computing 
  • G. Automation 
  • H. Agile Methods & Processes 

Defining Open Networking

Open Networking is much broader than other definitions, but it’s the only definition that doesn’t create more solution silos or bend the solution outcome to a buzzword or competing technology.  There is a need for a holistic definition of open networking that is inclusive and holistic and produces the best results. 

As a result of these technologies, hardware-based, specific-function, and proprietary components are being replaced by more generic and more straightforward hardware, and software is being migrated to perform more critical functions.

Open Networking in Practice:

Open Networking is already making its mark across various industries. Cloud service providers, for example, rely heavily on Open Networking principles to build scalable and flexible data center networks. Telecom operators also embrace Open Networking to deploy virtualized network functions, enabling them to offer services more efficiently and adapt to changing customer demands.

**Role of SDN and NFV**

Moreover, adopting software-defined networking (SDN) and network function virtualization (NFV) further accelerates the realization of the benefits of open networking. SDN separates the control plane from the data plane, providing centralized network management and programmability. NFV virtualizes network functions, allowing for dynamic provisioning and scalability. 

A. Use Cases and Real-World Examples: 

Data Centers and Cloud Computing: Open networking has gained significant traction in data centers and cloud computing environments. By leveraging open networking principles, organizations can build scalable and flexible data center networks that seamlessly integrate with cloud platforms, enabling efficient data management and resource allocation.

**Separate Control from Data Plane**

Software-Defined Networking (SDN): SDN is an example of open networking principles. By separating the control plane from the data plane, SDN enables centralized network management, automation, and programmability. This approach empowers network administrators to dynamically configure and optimize network resources, improving performance and reducing operational overhead.

B. Key Open Networking Projects:

Open Network Operating System (ONOS): ONOS is a collaborative project that focuses on creating an open-source, carrier-grade SDN (Software-Defined Networking) operating system. It provides a scalable platform for building network applications and services, facilitating innovation and interoperability.

OpenDaylight (ODL): ODL is a modular, extensible, open-source SDN controller platform. It aims to accelerate SDN adoption by providing developers and network operators with a common platform to build and deploy network applications.

FRRouting (FRR): FRR is an open-source IP routing protocol suite that supports various routing protocols, including OSPF, BGP, and IS-IS. It offers a flexible and scalable routing solution, enabling network operators to optimize their routing infrastructure.

The Role of Transformation

Infrastructure: Embrace Transformation:

To undertake an effective SDN data center transformation strategy, we must accept that demands on data center networks come from internal end-users, external customers, and considerable changes in the application architecture. All of these factors put pressure on traditional data center architecture.

Dealing effectively with these demands requires the network domain to become more dynamic, potentially introducing Open Networking and Open Networking solutions. For this to occur, we must embrace digital transformation and the changes it will bring to our infrastructure. Unfortunately, keeping current methods is holding back this transition.

Modern Network Infrastructure:

In modern network infrastructures, as has been the case on the server side for many years, customers demand supply chain diversification regarding hardware and silicon vendors. This diversification reduces the Total Cost of Ownership because businesses can drive better cost savings. In addition, replacing the hardware underneath can be seamless because the software above is standard across both vendors.

Leaf and Spine Architecture:

Further, as architectures streamline and spine leaf architecture increases from the data center to the backbone and the Edge, a typical software architecture across all these environments brings operational simplicity. This perfectly aligns with the broader trend of IT/OT convergence.  

Working with Open Source Software

Linux Networking

One remarkable aspect of Linux networking is the abundance of powerful tools available for network configuration. From the traditional ifconfig and route commands to the more recent ip command, this section will introduce various tools and their functionalities.

Virtual Switching: Open vSwitch

What is Open vSwitch?

Open vSwitch is a multilayer virtual switch that enables network automation and management in virtualized environments. It bridges virtual machines (VMs) and the physical network, allowing seamless communication and control over network traffic. With its extensible architecture and robust feature set, Open vSwitch offers a flexible and scalable networking solution.

Open vSwitch offers many features, making it a popular choice among network administrators and developers. Some of its key capabilities include:

1. Virtual Network Switching: Open vSwitch can create and manage virtual switches, ports, and bridges, creating complex network topologies within virtualized environments.

2. Flow Control: With Open vSwitch, you can define and control network traffic flow using flow rules. This enables advanced traffic management, filtering, and QoS (Quality of Service) capabilities.

3. Integration with SDN Controllers: Open vSwitch seamlessly integrates with various Software-Defined Networking (SDN) controllers, providing centralized management and control of network resources.

Containers & Docker Networking

Docker networking revolves around containers, networks, and endpoints. Containers are isolated environments that run applications, while networks act as virtual channels for communication. Endpoints, on the other hand, are unique identifiers attached to containers within a network. Understanding these fundamental concepts is crucial for grasping Docker network connectivity.

Docker Networking Fundamentals

Docker networking operates on a virtual network that allows containers to communicate securely. Docker creates a bridge network called “docker0” by default and assigns each container a unique IP address. This isolation ensures that containers can run independently without interfering with each other.

The default bridge network in Docker is an internal network that connects containers running on the same host. Containers within this network can communicate with each other using IP addresses. However, containers on different hosts cannot directly communicate over the bridge network.

Orchestrator: Understanding Docker Swarm

Docker Swarm, a native clustering and orchestration tool for Docker, allows the management of a cluster of Docker nodes as a single virtual system. It provides high availability, scalability, and ease of use for deploying and managing containerized applications. With its intuitive user interface and powerful command-line interface, Docker Swarm simplifies managing container clusters.

Related: For pre-information, you may find the following posts helpful:

  1. OpenFlow Protocol
  2. Software-defined Perimeter Solutions
  3. Network Configuration Automation
  4. SASE Definition
  5. Network Overlays
  6. Overlay Virtual Networking

Open Networking Solutions

Open Networking: The Solutions

Now, let’s look at the evolution of data centers to see how we can achieve this modern infrastructure. To evolve and keep up with current times, you should use technology and your infrastructure as practical tools. You will be able to drive the entire organization to become digital. Of course, the network components will play a key role. Still, the digital transformation process is an enterprise-wide initiative focusing on fabric-wide automation and software-defined networking.

A. Lacking fabric-wide automation:

One central pain point I have seen throughout networking is the necessity to dispense with manual work lacking fabric-wide automation. In addition, it’s common to deploy applications by combining multiple services that run on a distributed set of resources. As a result, configuration and maintenance are much more complex than in the past. You have two options to implement all of this.

Undertaking Manual or Automated Approach

First, you can connect these services by manually spinning up the servers, installing the necessary packages, and SSHing to each one. Alternatively, you can go toward open network solutions with automation, particularly Ansible automation with Ansible Engine or Ansible Tower with automation mesh. As automation best practice, use Ansible variables for flexible playbook creation that can be easily shared and used amongst different environments.  

B. Fabric-wide automation and SDN:

However, deploying a VRF or any technology, such as an anycast gateway, is a dynamic global command in a software-defined environment. We now have fabric-wide automation and can deploy with one touch instead of numerous box-by-box configurations. 

We are moving from a box-by-box configuration to the atomic programming of a single entity’s distributing fabric. This allows us to carry out deployments with one configuration point quickly and without human error.

C. Configuration management:

Manipulating configuration files by hand is tedious, error-prone, and time-consuming. Equally, performing pattern matching to make changes to existing files is risky. The manual approach will result in configuration drift, where some servers will drift from the desired state. 

Configuration Drift: Configuration drift is caused by inconsistent configuration items across devices, usually due to manual changes and updates and not following the automation path. Ansible architecture can maintain the desired state across various managed assets.

Storing Managed Assets: Managed assets, which can range from distributed firewalls to Linux hosts, are stored in an inventory file, which can be static or dynamic. Dynamic inventories are best suited for a cloud environment where you want to gather host information dynamically. Ansible is all about maintaining the desired state for your domain.

Challenge: The issue of Silos

To date, the networking industry has been controlled by a few vendors. We have dealt with proprietary silos in the data center, campus/enterprise, and service provider environments. The major vendors will continue to provide a vertically integrated lock-in solution for most customers. They will not allow independent, 3rd party network operating system software to run on their silicon.

Required: Modular & Open

Typically, these silos were able to solve the problems of the time. The modern infrastructure needs to be modular, open, and straightforward. Vendors need to allow independent, 3rd party network operating systems to run on their silicon to break from being a vertically integrated lock-in solution. Cisco has started this for the broader industry regarding open networking solutions with the announcement of the Cisco Silicon ONE. 

The Rise of Open Networking Solutions

New data center requirements have emerged; therefore, the network infrastructure must break the silos and transform to meet these trending requirements. One can view the network transformation as moving from a static and conservative mindset that results in cost overrun and inefficiencies to a dynamic routed environment that is simple, scalable, secure, and can reach the far edge. For effective network transformation, we need several stages. 

**Routed Data Center Design**

Firstly, transition to a routed data center design with a streamlined leaf-spine architecture and a standard operating system across cloud, Edge, and 5G networks. A viable approach would be to do all this with open standards, without proprietary mechanisms. Then, we need good visibility.

**Networking and Visibility**

As part of the transformation, the network is no longer considered a black box that needs to be available and provide connectivity to services. Instead, the network is a source of deep visibility that can aid a large set of use cases: network performance, monitoring, security, and capacity planning, to name a few. However, visibility is often overlooked with an over-focus on connectivity and not looking at the network as a valuable source of information.

**Monitoring at a Flow level**

For efficient network management, we must provide deep visibility for the application at a flow level on any port and device type. You would deploy a redundant monitoring network if you want something comparable today. Such a network would consist of probes, packet brokers, and tools to process the packet for metadata.

**Packet Brokers: Traditional Tooling**

Traditional network monitoring tools like packet brokers require life cycle management. A more viable solution would integrate network visibility into the fabric and would not need many components. This would enable us to do more with the data and aid in agility for ongoing network operations.

Note: Observability: Detecting the unknown

There will always be some requirement for application optimization or a security breach, where visibility can help you quickly resolve these issues. Monitoring is used to detect known problems and is only valid with pre-defined dashboards that show a problem you have seen before, such as capacity reaching its limit.

On the other hand, we have the practices of Observability that can detect unknown situations and are used to aid those in getting to the root cause of any problem, known or unknown: 

Example Visibility Technology: sFlow

What is sFlow?

sFlow is a network monitoring technology that allows for real-time, granular network traffic analysis. By sampling packets at high speeds, sFlow provides a comprehensive view of network behavior, capturing key data such as source and destination addresses, port numbers, and traffic volumes. This invaluable information serves as the foundation for network optimization and security.

Evolution of the Data Center

**Several Important Design Phases**

We are transitioning, and the data center has undergone several design phases. Initially, we started with layer 2 silos, suitable for the north-to-south traffic flows. However, layer 2 designs hindered east-west communication traffic flows of modern applications and restricted agility, which led to a push to break network boundaries.

**Layer 3 Routing & Overlay Networking**

Hence, routing at the top of the rack (ToR) with overlays between ToR is moved to drive inter-application communication. This is the most efficient approach, which can be accomplished in several ways. 

The demand for leaf and spine “clos” started in the data center and spread to other environments. A closed network is a type of non-blocking, multistage switching architecture.

This network design extends from the central/backend data center to the micro data centers at the EdgeEdge. Various parts of the edge network, PoPs, central offices, and packet core have all been transformed into leaf and spine “clos” designs. 

The network overlay

When increasing agility, building a complete network overlay is common to all software-defined technologies. An overlay is a solution abstracted from the underlying physical infrastructure. This means separating and disaggregating the customer applications or services from the network infrastructure. Think of it as a sandbox or private network for each application on an existing network.

Example: Overlay Networking with VXLAN

The network overlay is more often created with VXLAN. The Cisco ACI uses an ACI network of VXLAN for the overlay, and the underlay is a combination of BGP and IS-IS. The overlay abstracts a lot of complexity, and Layer 2 and 3 traffic separation is done with a VXLAN network identifier (VNI).

The VXLAN overlay

VXLAN uses a 24-bit network segment ID, called a VXLAN network identifier (VNI), for identification. This is much larger than the 12 bits used for traditional VLAN identification. The VNI is just a fancy name for a VLAN ID, but it now supports up to 16 Million VXLAN segments. 

Challenge: Traditional VLANs

This is considerably more than the traditional 4094-supported endpoints with VLANs. Not only does this provide more hosts, but it also enables better network isolation capabilities, with many little VXLAN segments instead of one large VLAN domain.

Required: Better Isolation and Scalability

The VXLAN network has become the de facto overlay protocol and brings many advantages to network architecture regarding flexibility, isolation, and scalability. VXLAN effectively implements an Ethernet segment that virtualizes a thick Ethernet cable.

Use Case: – **VXLAN Flood and Learn**

Flood and learn is a crucial mechanism within VXLAN that enables the dynamic discovery of VXLAN tunnels and associated endpoints. When a VXLAN packet reaches a switch, and the destination MAC address is unknown, the switch utilizes flood and learns to broadcast the packet to all its VXLAN tunnels. The receiving tunnel endpoints then examine the packet, learn the source MAC address, and update their forwarding tables accordingly.

Traditional policy deployment

Traditionally, deploying an application to the network involves propagating the policy to work through the entire infrastructure. Why? Because the network acts as an underlay, segmentation rules configured on the underlay are needed to separate different applications and services.

This creates a rigid architecture that cannot react quickly and adapt to changes, therefore lacking agility. The applications and the physical network are tightly coupled. Now, we can have a policy in the overlay network with proper segmentation per customer.

1. Virtual Networking & ToR switches

Virtual networks and those built with VXLAN are built from servers or ToR switches. Either way, the underlying network transports the traffic and doesn’t need to be configured to accommodate the customer application. Everything, including the policy, is done in the overlay network, which is most efficient when done in a fully distributed manner.

2. Flexibility of Overlay Networking

Now, application and service deployment occurs without touching the physical infrastructure. For example, if you need to have Layer 2 or Layer 3 paths across the data center network, you don’t need to tweak a VLAN or change routing protocols.  Instead, you add a VXLAN overlay network. This approach removes the tight coupling between the application and network, creating increased agility and simplicity in deploying applications and services.

**Key Point: Extending from the data center**

Edge computing creates a fundamental disruption among the business infrastructure teams. We no longer have the framework where IT only looks at the backend software, such as Office365, and OT looks at the routing and switching product-centric elements. There is convergence.

Therefore, you need many open APIs. The edge computing paradigm brings processing closer to the end devices, reducing latency and improving the end-user experience. It would help if you had a network that could work with this model to support this. Having different siloed solutions does not work. 

3. Required: Common software architecture

So the data center design went from the layer 2 silo to the leaf and spine architecture with routing to the ToR. However, there is another missing piece. We need a standard operating software architecture across all the domains and location types for switching and routing to reduce operating costs. The problem remains that even on one site, there can be several different operating systems.

I have experienced the operational challenge of having many Cisco operating systems on one site through recent consultancy engagements. For example, I had an IOS XR for service provider product lines, IOS XE for enterprise, and NS OX for the data center, all on a single site.

4. Challenge: The traditional integrated vendor

Traditionally, networking products were a combination of hardware and software that had to be purchased as an integrated solution. Conversely, open networking disaggregates hardware from software, allowing IT to mix and match at will.

With Open Networking, we are not reinventing how packets are forwarded or routers communicate. With Open Networking solutions, you are never alone and never the only vendor. The value of software-defined networking and Open Networking is doing as much as possible in software so you don’t depend on delivering new features from a new generation of hardware. If you want a new part, it’s quickly implemented in software without swapping the hardware or upgrading line cards.

5. Required: Move intelligence to software.

You want to move as much intelligence as possible into software, thus removing the intelligence from the physical layer. You don’t want to build in hardware features; you want to use the software to provide the new features. This is a critical philosophy and is the essence of Open Networking. Software becomes the central point of intelligence, not the hardware; this intelligence is delivered fabric-wide.

As we have seen with the rise of SASE, customers gain more agility as they can move from generation to generation of services without hardware dependency and without the operational costs of constantly swapping out the hardware.

**SDN Network Design Options**

We have both controller and controllerless options. With a controllerless solution, setup is faster, agility increases, and robustness in single-point-of-failure is provided, particularly for out-of-band management, i.e., connecting all the controllers.

SDN Controllerless & Controller architecture:

A controllerless architecture is more self-healing; anything in the overlay network is also part of the control plane resilience. An SDN controller or controller cluster may add complexity and impede resiliency. Since the network depends on them for operation, they become a single point of failure and can impact network performance. The intelligence kept in a controller can be a point of attack.

So, there are workarounds where the data plane can continue forward without an SDN controller but always avoid a single point of failure or complex ways to have a quorum in a control-based architecture.

We have two main types of automation to consider: day 0 and days 1-2. First and foremost, day 0 automation simplifies and reduces human error when building the infrastructure. Days 1-2 touch the customer more. This may include installing services quickly, e.g., VRF configuration and building Automation into the fabric. 

A. Day 0 automation

As I said, day 0 automation builds basic infrastructures, such as routing protocols and connection information. These stages need to be carried out before installing VLANs or services. Typical tools that software-defined networking uses are Ansible or your internal applications to orchestrate the building of the network.

Fabric Automation Tools

These are known as fabric automation tools. Once the tools discover the switches, the devices are connected in a particular way, and the fabric network is built without human intervention. It simplifies traditional automation, which is helpful in day 0 automation environments.

  • Configuration Management: Ansible is a configuration management tool that can help alleviate manual challenges. Ansible replaces the need for an operator to tune configuration files manually and does an excellent job in application deployment and orchestrating multi-deployment scenarios.  
  • Pre-deployed infrastructure: Ansible does not deploy the infrastructure; you could use other solutions like Terraform that are best suited for this. Terraform is infrastructure as a code tool. Ansible is often described as a configuration management tool and is typically mentioned along the same lines as Puppet, Chef, and Salt. However, there is a considerable difference in how they operate.

Most notably, the installation of agents. Ansible automation is relatively easy to install as it is agentless. The Ansible architecture can be used in large environments with Ansible Tower using the execution environment and automation mesh. I have recently encountered an automation mesh, a powerful overlay feature that enables automation closer to the network’s edge.

Ansible ensures that the managed asset’s current state meets the desired state. It is all about state management. It does this with Ansible Playbooks, more specifically, YAML playbooks. A playbook is a term Ansible uses for a configuration management script that ensures the desired state is met. Essentially, playbooks are Ansible’s configuration management scripts. 

B. Day 1-2 automation

With day 1-2 automation, SDN does two things.

Firstly, installing or provisioning services automatically across the fabric is possible. With one command, human error is eliminated. The fabric synchronizes the policies across the entire network. It automates and disperses the provisioning operations across all devices. This level of automation is not classical, as this strategy is built into the SDN infrastructure. 

Secondly, it integrates network operations and services with virtualization infrastructure managers such as OpenStack, VCenter, OpenDaylight, or, at an advanced level, OpenShift networking SDN. How does the network adapt to the instantiation of new workloads via the systems? The network admin should not even be in the loop if, for example, a new virtual machine (VM) is created. 

A signal that a VM with specific configurations should be created should be propagated to all fabric elements. When the virtualization infrastructure managers provide a new service, you shouldn’t need to touch the network. This represents the ultimate agility as you remove the network components. 

Summary: Open Networking

Networking is vital in bringing people and ideas together in today’s interconnected world. Traditional closed networks have their limitations, but with the emergence of open networking, a new era of connectivity and collaboration has dawned. This blog post explored the concept of open networking, its benefits, and its impact on various industries and communities.

What is Open Networking?

Open networking uses open standards, open-source software, and open APIs to build and manage networks. Unlike closed networks that rely on proprietary systems and protocols, open networking promotes interoperability, flexibility, and innovation. It allows organizations to customize and optimize their networks based on their unique requirements.

Benefits of Open Networking

Enhanced Scalability and Agility: Open networking enables organizations to scale their networks more efficiently and adapt to changing needs. Decoupling hardware and software makes adding or removing network components easier, making the network more agile and responsive.

Cost Savings: With open networking, organizations can choose hardware and software components from multiple vendors, promoting competition and reducing costs. This eliminates vendor lock-in and allows organizations to use cost-effective solutions without compromising performance or reliability.

Innovation and Collaboration: Open networking fosters innovation by encouraging collaboration among vendors, developers, and users. Developers can create new applications and services that leverage the network infrastructure with open APIs and open-source software. This leads to a vibrant ecosystem of solutions that continually push the boundaries of what networks can achieve.

Open Networking in Various Industries

Telecommunications: Open networking has revolutionized the telecommunications industry. Telecom operators can now build and manage their networks using standard hardware and open-source software, reducing costs and enabling faster service deployments. It has also paved the way for the adoption of virtualization technologies like Network Functions Virtualization (NFV) and Software-Defined Networking (SDN).

Data Centers: Open networking has gained significant traction in the world of data centers. Data center operators can achieve greater agility and scalability using open standards and software-defined networking. Open networking also allows for better integration with cloud platforms and the ability to automate network provisioning and management.

Enterprise Networks: Enterprises are increasingly embracing open networking to gain more control over their networks and reduce costs. Open networking solutions offer greater flexibility regarding hardware and software choices, enabling enterprises to tailor their networks to meet specific business needs. It also facilitates seamless integration with cloud services and enhances network security.

Open networking has emerged as a powerful force in today’s digital landscape. Its ability to promote interoperability, scalability, and innovation makes it a game-changer in various industries. Whether revolutionizing telecommunications, transforming data centers, or empowering enterprises, open networking connects the world in ways we never thought possible.

sd-wan3

VIPTELA SD-WAN – WAN Segmentation

 

 

VIPTELA SD-WAN

Problem Statement

WAN edge networks sit too far from business logic and are built, and designed with limited application and business flexibility. On the other hand, applications sit closer to business logic. It’s time for networking to bridge this gap using policies and business logic-aware principles. For additional information on WAN, challenges proceed to this SD WAN tutorial.

 

Traditional Segmentation

Network segmentation is defined as a portion of the network that is separated from the rest. Segmentation can be physical or logical. Physical segmentation involves complete isolation at a device and link level. Some organizations require the physical division of individual business units for security, political, or other reasons. At a basic level, logical segmentation begins with VLAN boundaries at Layer 2. VLAN consists of a group of devices that communicate as if they were connected to the same wire. As VLANs are logical and not physical connections, they are flexible and can span multiple devices.

While VLANs provide logical separation at Layer 2, virtual routing and forwarding (VRF) instances provide separation at Layer 3. Layer 2 VLAN can now map to Layer 3 VRF instances. However, every VRF has a separate control plane and configuration completed on a hop-by-hop basis. Individual VRFs with separate control planes from individual routing protocol neighbor relationships, hamper router performance.

VIPTELA VRF
VRF Separation

MPLS/VPNs overcome hop-by-hop configurations to allow segmentation. This enables physically divided business units to be logically divided without the need for individual hop-by-hop VPN configurations throughout the network. Instead, only the PE edge routers carry VPN information. This supports a variety of topologies such as hub and spoke, partial mesh, or any to any connectivity. While MPLS/VPNs have their benefits, they also introduce a unique set of challenges.

MPLS Challenges

MPLS topologies, once provisioned, are difficult to modify. This is due to the fact that all the impacted PE routers have to be re-provisioned with each new policy change. In this way, MPLS topologies are similar to the brick foundation of a house. Once the foundation is laid, it’s hard to make changes to the original structure without starting over.

Most modifications to VPN site topologies must go through an additional design phase. If a Wide Area Network (WAN) is outsourced to a carrier, it would require service provider intervention with additional design & provisioning activities. A simple task such as mapping application subnets to new or existing RT may involve onsite consultants, new high-level design, and configuration templates which would have to be applied by provisioning teams. Service provider provisioning and design activities are not free and usually have long lead times.

Some flexible Edge MPLS designs do exist. For example, community tagging and matching. During the design phase, the customer and service provider agree on predefined communities. Once these are set by the customer (attached to a prefix) they are matched by the provider to perform a certain type of traffic engineering (TE).

While community tagging and matching do provide some degree of flexibility and are commonly used, it remains a fixed, predefined configuration. Any subsequent design changes may still require service provider intervention.

Applications Must Fit A Certain Topology

The model forces applications to fit into a network topology that is already built and designed. It lacks the flexibility for the network to keep up with changing business needs. What’s needed is a way to map application requirements to the network. Applications are exploding and each has a variety of operational and performance requirements, which should be met in isolation.

Viptela SD-WAN & Topologies Per Application

By moving from hardware and diverse control planes to software & unified control planes at the WAN edge, SD-WAN evolves the fixed network approach. It abstracts the details of WAN; allowing application-independent topologies. SD-WAN provides segmentation between traffic sets and could be a good way to help create on-demand applications. Essentially, creating multiple on-demand topologies per application or group of applications. Each application can have its own topology and dynamically changing policies which are managed by a central controller.

The application controls the network design.

SD-WAN - Application flows
SD-WAN – Application flows

In SD-WAN, the central controller is hosted and managed by the customer, not a service provider. This enables the WAN to be segmented for each application at the customer’s discretion. For example, PCI traffic can be transported using an overlay specifically designed for compliance via Provider A. Meanwhile, ATM traffic can travel over the same provider network but using an overlay specifically designed for ATM. Meanwhile, each overlay can have different performance characteristics for path failover so that if the network does not reach a certain round-trip time (RTT) metric, traffic can reroute over another path. The customer has complete control over what application goes where and has the power to change policies on the fly.

The SDN controller, which acts as the brain, can set different applications to run over different paths based on business requirements and performance SLAs, not on a fixed topology. Each topology is segmented from the others and applications can share or have independent on-demand topologies. SD-WAN dramatically accelerates the time it takes to adapt the network to changing business needs.

SD-WAN Topologies

The network topologies can be depicted either physically or logically. Common topologies such as Star, Mesh, Full, and Ring are categorized under a centralized or decentralized function. In a physical world, these topologies are fixed and cannot be automatically changed. Logical topologies may also be hindered by physical device footprints.

In contrast, SD-WAN fully supports the coexistence of multiple application topologies regardless of existing physical footprints. For example, Lync message and video subscriptions may require different path topologies with separate SLAs. Messages may travel over low-cost links while video requires lower latency transports.

SD-WAN can flexibly cater to the needs of any type of application. In retail environments, store-specific applications may require a hub and spoke topology for authentication or security requirements. Surveillance systems may require full mesh topology. Guest Wi-fi may require local Internal access compared to normal user traffic that is scrubbed via a hub site. This per-application topology gives designers better control over the network. Viptela SD-WAN endpoints support multiple logical segments (regardless of the existing physical network), each of which can use a unique topology (full mesh, hub and spoke, ring) and be managed via its own policy.

Viptela SD-WAN: Predictable Application Performance

Obtaining predictable services is achieved by understanding the per-application requirements and routing appropriately.

Traditional MPLS WANs, offer limited fabric visibility. For example, providers allow enterprises to perform traceroute and pings, but bits-per-second is a primitive and unreliable method for measuring end-to-end application performance. Instead, WANs should be monitored at multiple layers, not just at the packet layer.

If a service provider is multiple autonomous systems (AS) away and experiencing performance problems, these cannot be addressed and detected using traditional distance-vector methods. This makes it impossible to route around problems and detect transitory oscillations. If errors on a transit path exist, a way must exist to penalize those paths.

Currently, there’s no way to detect when a remote network on the Internet is experiencing brownouts. Since the routing protocol is still operating, the best path does not change as neighbors might still be up. Routing should exist at the transport and application layer, and monitor both application flows and transactions. SD-WAN provides this function and delivers visibility at the device, transport, and application layer for insight into how the network is performing at any given time. This makes it possible to react to transitory failures before they can impact users.

“This post is sponsored by Viptela, an SD-WAN company. All thoughts and opinions expressed are the authors”

 

viptela1

Viptela Software Defined WAN (SD-WAN)

 

viptela sd wan

Viptela SD WAN

Why can’t enterprise networks scale like the Internet? What if you could virtualize the entire network?

Wide Area Network (WAN) connectivity models follow a hybrid approach, and companies may have multiple types – MPLS and the Internet. For example, branch A has remote access over the Internet, while branch B employs private MPLS connectivity. Internet and MPLS have distinct connectivity models, and different types of overlay exist for the Internet and MPLS-based networks.

The challenge is to combine these overlays automatically and provide a transport-agnostic overlay network. The data consumption model in enterprises is shifting. Around 70% of data is; now Internet-bound, and it is expensive to trombone traffic from defined DMZ points. Customers are looking for topological flexibility, causing a shift in security parameters. Topological flexibility forces us to rethink WAN solutions for tomorrow’s networks and leads towards Viptela SD-WAN.

 

Before you proceed, you may find the following helpful:

  1. SD WAN Tutorial
  2. SD WAN Overlay
  3. SD WAN Security 
  4. WAN Virtualization
  5. SD-WAN Segmentation

 

Solution: Viptela SD WAN

Viptela created a new overlay network called Secure Extensible Network (SEN) to address these challenges. For the first time, encryption is built into the solution. Security and routing are combined into one solution. Enables you to span environments, anywhere-to-anywhere in a secure deployment. This type of architecture is not possible with today’s traditional networking methods.

Founded in 2012, Viptela is a Virtual Private Network (VPN) company utilizing concepts of Software Defined Networking (SDN) to transform end-to-end network infrastructure. Based in San Jose, they are developing an SDN Wide Area Network (WAN) product offering any-to-any connectivity with features such as application-aware routing, service chaining, virtual Demilitarized Zone (DMZ), and weighted Equal Cost Multipath (ECMP) operating on different transports.

The key benefit of Viptela is any-to-any connectivity product offering. Connectivity was previously found in Multiprotocol Label Switching (MPLS) networks. They purely work on the connectivity model and not security frameworks. They can, however, influence-traffic paths to and from security services.

Viptela sd wan

 

Ubiquitous data plane

MPLS was attractive because it had a single control plane and a ubiquitous data plane. As long as you are in the MPLS network, connection to anyone is possible. Granted, you have the correct Route Distinguisher (RD) and Route Target (RT) configurations. But why can’t you take this model to Wide Area Network? Invent a technology that can create a similar model and offer ubiquitous connectivity regardless of transport type ( Internet, MPLS ).

 

Why Viptela SDN WAN?

The business today wants different types of connectivity modules. When you map service to business logic, the network/service topology is already laid out. It’s defined. Services have to follow this topology. Viptela is changing this concept by altering the data and control plane connectivity model using SDN to create an SDN WAN technology.

SDN is all about taking intensive network algorithms out of the hardware. Previously, in traditional networks, this was in individual hardware devices using control plane points in the data path. As a result, control points may become congested (for example – OSPF max neighbors reached). Customers lose capacity on the control plane front but not on the data plane. SDN is moving the intensive computation to off-the-shelf servers. MPLS networks attempt to use the same concepts with Route-Reflector (RR) designs.

They started to move route reflectors off the data plane to compute the best-path algorithms. Route reflectors can be positioned anywhere in the network and do not have to sit on the data path. Controller-based SDN approach, you are not embedding the control plane in the network. The controller is off the path. Now, you can scale out and SDN frameworks centrally provision and push policy down to the data plane.

Viptela can take any circuit and provide the ubiquitous connectivity MPLS provided, but now, it’s based on a policy with a central controller. Remote sites can have random transport methods. One leg could be the Internet, and the other could be MPLS. As long as there is an IP path between endpoint A and the controller, Viptela can provide the ubiquitous data plane.

 

Viptela SD WAN and Secure Extensible Network (SEN)

Managed overlay network

If you look at the existing WAN, it is two-part: routing and security. Routing connects sites, and security secures transmission. We have too many network security and policy configuration points in the current model. SEN allows you to centralize control plane security and routing, resulting in data path fluidity. The controller takes care of routing and security decisions.

It passes the relevant information between endpoints. Endpoints can pop up anywhere in the network. All they have to do is set up a control channel for the central controller. This approach does not build excessive control channels, as the control channel is between the controller and endpoints. Not from endpoint to endpoint. The data plane can flow based on the policy in the center of the network.

Viptela SD WAN

 

Viptela SD WAN: Deployment considerations

Deployment of separate data plane nodes at the customer site is integrated into existing infrastructure at Layer 2 or 3. So you can deploy incrementally, starting with one node and ending with thousands. It is so scalable because it is based on routed technology. The model allows you to deploy, for example, a guest network and then integrate it further into your network over time. Internally they use Border Gateway Protocol (BGP). One the data plane, they use standard IPSec between endpoints. It also works over Network Address Translation (NAT), meaning IPSec over UDP.

When an attacker gets access to your network, it is easy for them to reach the beachhead and hop from one segment to another. Viptela enables per-segment encryption, so even if they get to one segment, they will not be able to jump to another. Key management on a global scale has always been a challenge. Viptela solves this with a propitiatory distributed manager based on a priority system. Currently, their key management solution is not open to the industry.

 

SDN controller

You have a controller and VPN termination points i.e data plane points. The controller is the central management piece that assigns the policy. Data points are modules that are shipped to customer sites. The controller allows you to dictate different topologies for individual endpoint segments. Similar to how you influence-routing tables with RT in MPLS.

The control plane is at the controller.

 

Data plane module

Data plane modules are located at the customer site. They connect this data plane module, which could be a PE hand-off to the internal side of the network. The data plane module must be in the data plane path on the customer site. Internal side, they discover the routing protocols and participate in prefix learning. At Layer 2, they discover the VLANs. Their module can either be the default gateway or perform the router neighbor relationship function. WAN side, data plane module registers uplink IP address to WAN controller/orchestration system. The controller builds encrypted tunnels between the data endpoints. The encrypted control channels are only needed when you build over untrusted third parties.

If the problem occurs with controller connectivity, the on-site module can stop being the default gateway and usually participate in Layer 3 forwarding for existing protocols. It backs off from being the primary router for off-net traffic. It’s like creating VRF for different businesses and default routes for each VRF with a single peering point to the controller; Policy-Based Routing (PBR) for each VRF for data plane activity. The PBR is based on information coming from the controller. Each control segment can have a separate policy (for example – modifying the next hop). From a configuration point of view, you need an IP on the data plane module and the remote controller IP. The controller pushes down the rest.

 

  • Viptela SD WAN: Use cases

For example, you have a branch office with three distinct segments, and you want each endpoint to have its independent topology. The topology should be service driven, and the service should not follow existing defined topology. Each business should depict how they want their business to connect to the network team should not say this is how the topology is, and you must obey our topology.

From a carrier’s perspective, they can expand their MPLS network to areas they do not have a physical presence. And bring customers with this secure overlay to their closest pop where they have an MPLS peering. MPLS providers can expand their footprint to areas where they do not have service. If MPLS has customers in region X and wants to connect to the customer in region Y, they can use Viptela. Having those different data plane endpoints through a security framework would be best before entering the MPLS network.

Viptela allows you to steer traffic based on the SLA requirements of the application, aka Application-Aware Routing. For example, if you have two sites with dual connectivity to MPLS and Internet, data plane modules (located at customer sites) nodes can steer traffic over either the MPLS or Internet transport based on end-to-end latency or drops. They do this by maintaining the real-time loss, latency, and jitter characteristics and then applying policies on the centralized controller. As a result, critical traffic is always steered to the most reliable link. This architecture can scale to 1000 nodes in a full mesh topology.

 

viptela sd wan