Nothing Special   »   [go: up one dir, main page]

Ensemble Faqs PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 30

Ensemble provides a unique platform and architecture for supporting network functions

virtualization (NFV). Here are a series of frequently asked questions, and answers.

Ensemble architecture and components............................................................................................................... 3

Q: What is Ensemble? ............................................................................................................................................................. 3

Q: How do the Ensemble components fit into the ETSI NFV architecture? ...................................................... 5

Q: What is Ensemble Connector? ....................................................................................................................................... 6

Q: What are the benefits of Ensemble Connector? ..................................................................................................... 7

Q; What service activation and service assurance features does Ensemble Connector provide? ............ 8

Q: How does Ensemble Connector support service resiliency? ............................................................................. 8

Q: What is Ensemble Orchestrator?................................................................................................................................... 9

Q: How does Ensemble Orchestrator support resiliency? ...................................................................................... 11

Q: What is Ensemble Controller? ...................................................................................................................................... 11

Q: What is Ensemble Director? .......................................................................................................................................... 12

Q: How does Ensemble Director process fault information? ................................................................................ 13

Q: What is Ensemble Analytics? ........................................................................................................................................ 13

Q: What is the Ensemble Harmony Ecosystem? ......................................................................................................... 14

Q: How does Ensemble process software upgrades? .............................................................................................. 15

Ensemble technology ................................................................................................................................................ 16

Rev 1 – November 2017 1


Q: What are NFV and SDN, and what is the difference? ......................................................................................... 16

Q: What are white boxes, and why do people want to use them? ..................................................................... 17

Q: Can NFV meet telco performance requirements? ............................................................................................... 18

Q: What is embedded cloud aka “cloud in a box?” .................................................................................................. 18

Q: Why should I move away from network appliances? ......................................................................................... 19

Q: Does Ensemble Connector support containers? .................................................................................................. 20

Q: What lessons has Ensemble learned from working with service providers? ............................................. 20

Q: Is IT/network convergence realistic? ......................................................................................................................... 20

Q: What developments or initiatives are coming to aid NFVI operation? ....................................................... 21

Ensemble and NFV economics .............................................................................................................................. 23

Q: Does the NFV business case work for service providers? ................................................................................. 23

Q: What is the Ensemble pricing strategy?................................................................................................................... 23

Q: How is Ensemble licensed? ........................................................................................................................................... 24

Ensemble and virtual network functions (VNFs) ............................................................................................... 26

Q: What VNFs does Ensemble provide? ........................................................................................................................ 26

Q: Does Ensemble provide an app store for VNFs? .................................................................................................. 27

Q: What is the availability of VNFs that can run on COTS servers? .................................................................... 27

Q: Can we replace appliances with COTS hardware plus proprietary software? ........................................... 27

Ensemble use cases .................................................................................................................................................. 28

Q: What are the key use cases for Ensemble? ............................................................................................................. 28

Q: Does Ensemble have an SD-WAN solution? .......................................................................................................... 28

Q: Does Ensemble have live deployments?.................................................................................................................. 29

Q: What are the benefits of NFV to the end user? .................................................................................................... 29

2
Ensemble architecture and components

Q: What is Ensemble?
A: Ensemble is both a division of ADVA, as well as suite of software products from that division. The
Ensemble software supports pure-play virtualization, eliminating vendor lock-in while enabling a pure
open software and hardware environment. The Ensemble architecture is shown below.

The Ensemble components are designed to work together using open and standard application
programming interfaces. They can also work separately, allowing service providers to mix and match as
needed, and integrate with other systems. In short, Ensemble enables choice in:

Hardware suppliers
 White box, gray box, branded
 Compute, memory and storage
 Features (NEBS, power, redundancy)

3
Software suppliers
 Ability to change as needed
 Including swapping out Ensemble components
Deployment location
 Cloud / data center
 Central office
 Customer premises
 Any combination of the above

More info:
 Network virtualization overview page
 Ensemble Explained (video)
 Ensemble Connector - Network operating systems and VNF hosting platform
 Ensemble Orchestrator - ETSI MANO NFV orchestrator and VNF manager
 Ensemble Controller - ODL-based SDN controller
 Ensemble Director – FCAPS system
 Ensemble Analytics - Advanced analytical tools to correlate network data with service data
 Ensemble Hypervisor – Virtualization system for optical networks
 Ensemble Harmony Ecosystem – Partner program for NFV hardware, software and services
suppliers

4
Q: How do the Ensemble components fit into the ETSI NFV
architecture?
A: The first diagram below shows the ETSI NFV reference architecture. The second diagram overlays
the architecture with the Ensemble components.

5
Q: What is Ensemble Connector?
A: Award-winning Ensemble Connector enables pure-play virtualization: open software running on
open commercial off-the-shelf servers. It does so by providing an open and high-performance
network operating system.

Ensemble Connector includes standard NFV infrastructure (NFVI) components such as Linux, KVM,
Docker, and OpenStack. It runs on standard COTS servers as well as on ADVA’s enhanced server
platforms, including the FSP 150 ProVMi, FSP 150 ProVMe and OS-V8-M.

A high-level block diagram of Ensemble Connector is shown below.

More info:
 Ensemble Connector product page
 Ensemble Connector data sheet
 Ensemble Connector Explained (video)

6
Q: What are the benefits of Ensemble Connector?
A: Ensemble Connector improves on a standard NFVI platform with these features, as shown below:
1. Improved virtual switching – Ensemble Connector’s forwarding performance is faster, more
efficient and more consistent than open vSwitch (OVS).
2. Virtualized Carrier Ethernet 2.0 (CE 2.0) – Ensemble Connector provides standard CE 2.0
functions in software. This enables a single standard COTS server to host not only standard
VNFs, but also to present a CE 2.0 UNI on any of the server’s Ethernet ports. This eliminates
the need for an external NID.
3. Improved networking – Ensemble Connector can support a variety of advanced networking
applications at Layer 2 or 3, including support for LTE access.
4. Zero touch commissioning (ZTC) – With Ensemble Connector, service providers can ship an
unconfigured server to a customer site and then commission it securely without a technician.
5. Embedded cloud – Embedded cloud places a self-contained OpenStack cloud on the edge
node, enabling cloud-native deployments without the issues created by separating the
OpenStack controller from its agents. See “Q: What is embedded cloud aka cloud in a box?”
for more info.
6. Integrated OS with open interfaces – Ensemble Connector provides a unified platform to
simplify deployment. It also provides open, standard APIs to facilitate integration of 3 rd party
VNFs as well as northbound management platforms, including MANO and OSS/BSS.
7. Device scalability – Ensemble Connector supports a wide array of standard COTS servers,
ranging from low-cost Intel Atom®-based devices all the way up to multi-socket Intel®
Xeon® blade servers.
8. Telco management – Ensemble Connector provides sophisticated deployment features such
as ZTC, embedded cloud, NETCONF/YANG, SNMP, SSH, Y.1731, TWAMP, etc.
9. Security – Ensemble Connector provides for security at multiple levels: commissioning,
virtualization, management and user connections.

7
Q; What service activation and service assurance features does
Ensemble Connector provide?
Ensemble Connector provides native capabilities to support service activation and service assurance,
including:

Service activation

 Smart Layer 2 loopbacks (ingress and egress)


 Y.1731 Ethernet loopback supported

Service Assurance

 Y.1731/802.1ag with up and down MEPs (up to 100)


 Link OAM (802.3ah)
 TWAMP reflector for Layer 3 testing

Connector also provides performance management (PM) reporting via 15-minute binned counters:
 Interface RMON PM
 Interface Rx/TX PMs on Ethernet flows
 Per service CoS queue PM
 Y.1731 service OAM
 Tunnel latency, loss, throughput

Q: How does Ensemble Connector support service resiliency?


High availability requires a redundant uCPE to be deployed at a customer site. The second device
needs to be kept at the same software level, virtual infrastructure, VNFs and identical service chain.
When failure of the primary does occur then a mechanism needs to be in place to trigger the failover.
In this case, we propose that failover can be initiated based on packet loss, which is detected by a
failure of multiple CCMs in the Ethernet OAM.

In order to maintain resiliency within the service chain, all VNFs in the service chain should support
high availability synchronization. Connector has been designed to provide flexible networking for high
availability synchronization between VNFs of redundant service chain instances, thereby providing
resiliency to the overall service.

8
 If the flow of CCMs is disrupted (due to VNF
 Use link down transfer with 1 second CCM for failure or VNF forwarding) LDT can be
service chain protection. configured to deactivate the physical
 Link down transfer (LDT) allows Connector to WAN/LAN/sync interface(s), thus forcing a
inject CCMs at 1 second intervals into the vPort, failover to occur.
and monitor for those CCMs at the other end  LDT is a feature ADVA has today on other
of the service chain. MaestrOS-based products. However, this
feature will need to be ported into Connector.

Q: What is Ensemble Orchestrator?


A: Our award-winning Ensemble Orchestrator provides a scalable and economical solution for NFV
orchestration. Ensemble Orchestrator is compliant with the ETSI NFV MANO architecture, and it
provides both an NFV orchestrator (NFVO) and a generic VNF manager (VNFM).

Automated service creation


Our Ensemble Orchestrator helps providers and operators achieve greater profitability through policy-
driven automation for quick and reliable service creation, activation and assurance.

9
Single-point management
Ensemble Orchestrator provides a single-point of entry for end-to-end VNF and network service
lifecycle management. It handles VNF onboarding, service design, service deployment and VNF/service
operations and management.

Seamless integration
Ensemble Orchestrator features cloud management, service orchestration, and networking capabilities
as well as seamless operation across the various existing IP and virtual networks when integrated with
our Ensemble Controller.

Ensemble Orchestrator is feature rich:

 Full lifecycle management including VNF onboarding, service design, service deployment, as
well as VNF/service operations and generic/vendor-provided VNF management
 Dynamic tracking of cloud resources and advanced VNF placement algorithms
 Provides programmable and policy-driven workflows
 Supports cross-cloud VNF service chaining
 Multi-tenancy and per-tenant quota management
 Support for multiple cloud types
 Scalable and highly available architecture

More info:
 Ensemble Orchestrator product page
 Ensemble Orchestrator data sheet
 Ensemble Orchestrator Explained (video):
 NFV orchestration demands openness and flexibility
 Orchestration for NFV Needs Scale and Resiliency
 Making MANO Easy

10
Q: How does Ensemble Orchestrator support resiliency?

A single horizontal-scaling architecture


allows for performance and resiliency within
the Ensemble Orchestrator, as shown in the
nearby figure.

The application tier consists of multiple


instances of the Ensemble Orchestrator
Core. Ensemble Orchestrator achieves
scaling by adding instances of the Ensemble
Orchestrator core and then load balancing
across them. All state is stored in the local
instance database, which is configured for
cluster-level replication.

All incoming requests can be served by any


active instance of the Orchestrator Core.
Internally scheduled loads (monitoring, etc.) are automatically amortized across all instances.

The proxy tier provides topology hiding for the back-end Orchestrator Core instances. HTTP load
balancing is applied for both Web UI and REST APIs. The proxy tier may provide virtual IP or other
connectivity guarantee mechanisms. Ensemble Orchestrator supports any open-source or general
commercial HTTP-proxy solutions including HAProxy and keepalived.

The client layer is scale-agnostic and transactional in nature. Clients may use active/standby
connections or may spread requests among multiple active proxies.

Q: What is Ensemble Controller?


A: Ensemble Controller provides a service-aware network controller and management system for
Ensemble Connector applications. Its SDN applications and interfaces are based on the OpenDaylight
(ODL) framework.

Ensemble Controller implements the necessary WAN-SDN capabilities for delivering Connector control
functions, including:

11
 Authentication and zero touch provisioning of Ensemble Connector
 Provisioning of secure management tunnels
 HA clustering and horizontal scaling
 Dynamic network configuration handling
 Integration with northbound orchestration platforms

Ensemble Controller is implemented as a set of applications and plugins running on top of unmodified
ODL binaries. It is interoperable with any third-party commercial distributions of ODL, and it is loosely
coupled to ease migration to other controller frameworks e.g., Open Networking Operating System
(ONOS) if required.

More info:
 Ensemble Controller Explained (video)

Q: What is Ensemble Director?


A: Ensemble Director is Ensemble’s Web 2.0 solutions platform for telecom management (FCAPS). It
facilitates end-to-end management of large networks (native ADVA and third-party elements).

Ensemble Director provides a multi-layer architecture that delivers:


Core components
 Fault management
 Performance monitoring
 Security
 Provisioning
 Inventory
 Nodal discovery
 Topology discovery

Services
 Service provisioning
 End-to-end service management
 NFV/SDN service
 Service fault monitoring
 Service assurance

Application Extensions
 Open API architecture
 Custom and third-party applications

12
Ensemble Director hosts native and third-party, multi-layer network and service management
applications. It features an open, extensible, YANG-based, model-driven architecture providing a
software framework that supports custom application development in Scala/Java and Python.
Ensemble Director features industry standard interfaces, including web, CLI, SNMP, CSV, REST,
NETCONF, and RESTconf.

More info:
 Ensemble Director Explained (video)

Q: How does Ensemble Director process fault information?


Ensemble Director provides a powerful framework to collect faults and alarms from all Ensemble
components, as shown in the nearby image. Displays include:
 Dynamic current alarm count display
 Current alarm summary table
 Powerful alarm filtering/search
 Event browser maintains complete history of
raise/clear events
 Correlation between raise/clear events
 Alarm acknowledgment, latching alarms and
commenting function
 Log browser
 Northbound API and reporting
 Publish subscribe model
 Syslog

Q: What is Ensemble Analytics?


A: Ensemble Analytics leverages big data technologies, and applies advanced analytical tools to
correlate network data with service data and provides actionable intelligence for service orchestration
and assurance. It provides an open, standard API layer to integrate with Ensemble and third-party
analytics applications including:
 Network service lifecycle tracking and performance management
 NFV assurance and auto-scaling
 Capacity management and planning
 Root cause analysis

13
More info:
 Ensemble Analytics Explained (video)

Q: What is the Ensemble Harmony Ecosystem?


A: ADVA's ecosystem of hardware, software, and services partners focuses on accelerating automation
and virtualization initiatives at the metro service edge.

We believe a strong and open ecosystem is crucial for the success of NFV-based services.
Communication service providers are moving quickly out of proofs of concept. Each of the Ensemble
Harmony partnerships is being driven by real-world customer deployments.

The Ensemble Harmony partner program is a pillar of the open environment. We believe this is
essential for the realization of software-defined services. It enables service providers to select best-of-
breed components and avoid single vendor proprietary systems, while also decreasing their project
risks and accelerating deployments.

More Info:
 Ensemble Harmony Ecosystem web page
 Ensemble Harmony Ecosystem Explained (video)

14
Q: How does Ensemble process software upgrades?
Since Ensemble Connector operates as a nominal Linux application, service providers are able to
leverage standard IT automation tools to manage deployment and upgrades. In this context, Ensemble
leverages Ansible as a framework for deploying Ensemble Connector and its embedded instance
OpenStack. Ansible provides an open source, agentless, secure tool that allows operators to abstract
complex tasks into simple YAML.

Ensemble provides reference Ansible


playbooks supporting common tasks like
deployment and in-service upgrades. These
playbooks can be customized by the service
provider IT team or by ADVA professional
services as needed. The reference Ansible tools
provided with the Ensemble solution are used
to manage:
 Ensemble Connector flat file commissioning
 OpenStack provider networks
 Ensemble Connector software packages
 OpenStack compute and controller
containers
 Linux kernel
 Reversion to previous versions in case of
unsuccessful upgrades

Hitless upgrade of the Ensemble Connector OpenStack cloud can be accomplished at an Ensemble
Connector server by launching a new OpenStack instance on the server, migrating the existing VNF
services to the new cloud environment and then turning down the old cloud. The capability can be
productized into the per-release scripting services for use within Ensemble Connector’s automated
change management (ACM) Ansible tools.

Note that that hitless upgrade of OpenStack requires additional allocated CPU and memory to support
operating multiple cloud instances.

The ACM framework can also be used to deliver patches to individual components of the Ensemble
Connector platform – including the Ensemble Connector dataplane, OpenStack containers, and
Ensemble Connector management components.

15
Ensemble technology

Q: What are NFV and SDN, and what is the difference?


A: Software-defined networking (SDN) got its start on campus networks. As researchers were
experimenting with new protocols they were frustrated the need to change the software in the
network devices each time they wanted to try a new approach. They came up with the idea of making
the behavior of the network devices programmable and allowing them to be controlled by a central
element. This led to a formalization of the principle elements that define SDN today:
 Separation of control and forwarding functions
 Centralization of control
 Ability to program the behavior of the network using well-defined interfaces

Network functions virtualization (NFV) is a means to enable telco operators to leverage the benefits of
the cloud, including:
 Replacing closed appliances with software running on open servers, as shown below
 Reduction of CAPEX, OPEX, and power
 Construction of services from components from an ecosystem of hardware and software
suppliers
 New commercial models including usage-based and shared-risk/shared-reward
 New methods of working such as agile development and DevOps

16
The original ETSI NFV white paper included this diagram showing the intersection of SDN and NFV:

More info:
 ETSI NFV white paper
 NFV and SDN: What's the Difference?
 NFV and SDN: What’s the Difference Two Years Later?
 Cloudifying the Communications Network

Q: What are white boxes, and why do people want to use them?
A: White box is a term that denotes a generic and open networking platform. White boxes are
desirable for the following reasons:
 They decouple hardware from software, enabling best-of-breed choices in both
 They allow customers to deploy other enterprise applications in the same computing footprint
as their telco services
 They enable operators to ride the broader hardware innovation curve
 They simplify operations in locations that have onerous import restrictions by allowing for
local procurement of compatible hardware platforms

For NFV, we are specifically interested in white box servers. These servers can be loaded with the
software of our choice, including the preferred Linux/KVM/OpenStack environment for cloud-native
operation.

17
This is in contrast to data center applications that often use white box switches. These switches have an
OpenFlow interface or abstraction software.

More info:
 What is Universal CPE?
 Understanding the Use of Universal CPE
 Why Service Providers Need Universal CPE (page 10)
 Universal Customer Premises Equipment

Q: Can NFV meet telco performance requirements?


A: Yes. With the aid of more efficient VNFs and optimized infrastructure like Ensemble Connector,
operators can deploy virtualized systems that meet both performance and cost requirements.

More Info:
 Open vSwitch is No Match for ADVA Ensemble Connector Performance
 Performance and Low Cost Enable NFV and SDN

Q: What is embedded cloud aka “cloud in a box?”


A: Telco operators see many benefits to network functions virtualization (NFV). In short, the goal is to
bring the advantages of the cloud to telco operators. One of the big benefits is the wide array of
available support software, including Linux, kernel-based virtual machine (KVM), Open vSwitch and
OpenStack. These software packages provide a base of well-known functionality upon which operators
can build innovative new virtualized services.

However, the telco network is not the same as the data center, so cloud-centric solutions may not
transfer directly. In particular, operators are concerned about OpenStack regarding its scalability and
resiliency. How do we enable virtualized services to be deployed at scale when using OpenStack?

One way is to move from a centralized mode to a distributed mode. In a distributed mode, each
compute node has an instance of OpenStack controller creating a “cloud in a box” as shown below.

18
In the case of Ensemble Connector, we run the OpenStack controller in a Docker container and can
limit its footprint to a fraction of a CPU core, if required.

More Info:
 Can Cloud in a Box Address OpenStack Issues?
 What Is Cloud in a Box? (video)

Q: Why should I move away from network appliances?


A: Network appliances include routers, firewalls, SD-WAN endpoints, Wi-Fi controllers and others. They
include hardware and software from a single supplier. As a result, they are closed solutions that
typically provide a single-function from a single-supplier, which is delivered as a physically integrated
solution. If a service provider wants to change software vendors, this typically also requires changing
out expensive network appliance hardware. Additionally, most network appliance vendors will not
allow service provides to run “foreign applications” on the device. This type of vendor lock-in
encompasses all of the attributes that operators are trying to move away from with the push for cloud-
centric solutions.

More Info:
 Don’t Use Hardware Appliances for SD-WAN
 Cloudifying the Communications Network

19
Q: Does Ensemble Connector support containers?
A: Yes. Ensemble Connector is built using a standard Linux distribution. It supports Docker containers,
which are used for hosting an OpenStack controller for the embedded cloud model. Ensemble
Connector will also support containerized VNFs when they are available.

Q: What lessons has Ensemble learned from working with


service providers?
A: We are fortunate to have worked with operators on NFV deployments for over two years. As a
result, we have increased our knowledge and experience with NFV. Our products now have valuable
features and ease of deployment.

Specific examples:
 Our products now have improved features for deployability, such as zero touch, security, LTE
and support for a wide range of white box / COTS servers
 We have addressed many of the issues of OpenStack with our embedded cloud model
 We can provide support for platform scalability, from small Intel Atom®-based servers at the
low end to multi-socket Intel® Xeon® blade servers at the high end
 We have a variety of commercial models for licensing and support, including perpetual, time-
based, and success-based, with options for different speeds and feature sets
 We have broad and deep experience working with a wide range of VNF suppliers

More Info:
 Real-World NFV, Real Lessons Learned
 Ensemble Continues to Make NFV Easy – and Cloudy

Q: Is IT/network convergence realistic?


IT/network convergence is realistic; it is happening and NFV is the path. NFV was invented by service
providers to bring the benefits of the cloud to the telco network. NFV is often characterized as being
about replacing appliances with software running on COTS servers, but it is much more than that.
Here’s what operators are doing to fully realize the benefits of the cloud:
 They are moving from single-vendor solutions (whether appliances or single-vendor
applications running bare metal) to multi-vendor systems

20
 They are moving from static implementations to orchestrated VNFs on standard NFV
infrastructure, i.e. COTS server, Linux, KVM, OpenStack, Docker, etc.
 They are embracing new methods of working, such as agile development and DevOps. This
includes partnering with both suppliers and customers to accelerate the development cycle
 They are seeking to provide their customers with converged cloud/connectivity solutions. This
requires virtualized networking and security solutions
 They are looking to treat the network as a platform with fungible resources so as to enable a
new class of dynamic services and new technologies such as 5G and IoT
 They are looking to radically increase automation to reduce time to service and human errors,
as well as to enable automatic resolution of most network failures

The barriers and risks include:


 Acquiring the necessary skills through training and/or hiring
 Changing organizations to reflect the new realities of operation, i.e. breaking down internal
silos
 Building out new virtual infrastructure while maximizing the use of the existing network
 Driving new revenues quickly to justify the capital outlay
This migration is just getting underway, but operators see it as necessary to support their future
strategies. It will enable operators to move much more quickly, drive new services and cut costs. The
risks are manageable given leadership and vision.

More Info:
 Staffing and Organizing for Telco Innovation – Part 1
 Staffing and Organizing for Telco Innovation – Part 2
 Staffing and Organizing for Telco Innovation – Part 3
 Network Modernization Means Operator Innovation (Page 10)
 Service Providers are Changing for the Cloud
 Network Virtualization Simplified
 Cloudburst (page 32)

Q: What developments or initiatives are coming to aid NFVI


operation?
A: We see the following:
 Adoption of technologies such as CloudInit and standard VNF descriptors to aid onboarding
 Adoption of standard interfaces such as those being defined by the MEF LSO initiative

21
 Moving to containers to give an option for lightweight deployments
 Availability of ARM-based servers to provide an option for lower cost deployments

More Info:
 ARM Wrestling in the Cloud Arena: Winners Stay Pure

22
Ensemble and NFV economics

Q: Does the NFV business case work for service providers?


A: Yes. The NFV hardware and software market has matured greatly over the last few years. Now,
virtualized services can be deployed on white box COTS servers at a capital cost well below that of the
equivalent appliances. Additionally, using software-centric functions has significantly lowered
operational expenses while enabling innovation, all without changing the deployed hardware. ADVA
has developed tools to help service providers evaluate the potential total cost of ownership and return
on investment that can be achieved by investing in NFV. We are happy to share these tools upon
request.

More Info:
 Hard Truths about Software Licensing for NFV
 Optimizing Profitability with Pure-Play NFV
 The Business Value of Software-Defined Networking (page 18)
 The Real Reason to Deploy NFV: New Revenue

Q: What is the Ensemble pricing strategy?


A: The Ensemble components have been part of live deployments for over two years. One result is that
we have a very good understanding of the varying commercial models needed by service providers.
The Ensemble team has defined a flexible set of pricing models to meet these needs. These models
include:
 Perpetual licensing, which allows the service provider to deploy a feature forever, for a given
revision, performance, etc.
o Perpetual licensing typically includes some combination of:
 Feature licenses
 Speed-based licenses
 Support and maintenance
o ADVA’s strategy allows customers who maintain active support agreements to
perform feature constant software upgrades at no additional cost
 Subscription licensing provides all of the aforementioned perpetual licensing benefits but for a
single low monthly fee – typically, subscription models will include some minimum level of
subscription duration

23
 Both of the above models are built around a “pay as you grow” strategy that minimizes up-
front charges and defers licensing fees based on time and usage

In short, we can define a model that makes sense for both the service provider and Ensemble.

Q: How is Ensemble licensed?


Ensemble Connector simplifies the solution licensing into a single structure. Ensemble Connector
operating system (OS) and embedded cloud are licensed through a common license structure. As a
predominantly software-based solution, it is imperative that ADVA has the ability to manage the
distribution and usage of its software products to ensure that all deployed instances are covered
under appropriate commercial terms. The license structure is intended to be simple and non-service
impacting.

When first deployed, Ensemble Connector instances will operate in a fully functional mode for 60 days,
even when no licensing has been applied to the system. After that time – and when a license expires –
Ensemble Connector will still continue to function with no impairment to traffic flows; however, the
system will generate alarms and will prevent any configuration changes from being made to the
system until new license files are applied. Ensemble Connector supports several warning and
notification periods prior to license expiry to ensure that the operator is given ample opportunity to
provide up-to-date license files on the platform.

Operators configure Ensemble Connector instances with the desired features and bandwidth. This
entitlement configuration can be done at the device staging or through the zero touch configuration
file that the Ensemble Controller downloads after Ensemble Connector authenticates onto the
network.

There are three options for handling licensing: manual, automated through the customer OS, and
automated through a local license server.

In the manual method, the service provider applies the license through the Connector CLI. This method
is suitable for low volumes, such as in a lab or proof of concept. The service provider reads the device
ID from the Ensemble Connector and provides this information to ADVA, which generates a license for
that instance, which the operator then manually applies. The Ensemble Connector license is node-
locked to the device ID, derived from the server chassis serial number.

24
In volume deployment, this can be automated through the customer OS. In this method, the customer
OS reads the device ID of the Ensemble Connector instance. The customer OS then provides it to the
ADVA portal and obtains a pre-purchased license through portal APIs opened to the customer OS. The
customer OS then applies the node-locked license to the Ensemble Connector instance.

More Info:
 Hard Truths about Software Licensing for NFV

25
Ensemble and virtual network functions (VNFs)

Q: What VNFs does Ensemble provide?


A: None. Ensemble provides an open architecture for NFV that enables third-party VNFs to be easily
deployed. The figure below shows a subset of the currently onboarded VNFs. This list includes both
commercial and open source VNFs.
Third-party vendors

Open source VNFs Integrations

26
Q: What does onboarding mean?
A: Onboarding is the process of integrating a VNF with the NFV infrastructure (Ensemble Connector)
and MANO (Ensemble Orchestrator). This process can be performed by the Ensemble team, but it can
also be done by the operator or third-party contractor.

Q: Does Ensemble provide an app store for VNFs?


A: Not today. Most of our customers want to acquire VNFs directly from the suppliers. However,
Ensemble does provide for ease of use with its Harmony program and large portfolio of onboarded
VNFs.

More Info:
 Ensemble Harmony Ecosystem web page
 Ensemble Harmony Ecosystem Explained (video)

Q: What is the availability of VNFs that can run on COTS


servers?
A: There is currently a wide variety of software VNFs that are able to run on commercial off-the-shelf
(COTS) servers and standard cloud infrastructure, i.e. Linux/KVM/OpenStack. These include routers,
firewalls, SD-WAN agents, DPI, IMS, SBC, EPC, encryption, and many others.

Q: Can we replace appliances with COTS hardware plus


proprietary software?
A: Yes. Leading operators like Masergy, Verizon, AT&T, and CenturyLink are already doing deploying
virtualized services built on software VNFs running on COTS servers. By doing so, they are able to
leverage the benefits of cloud technologies, as described above.

27
Ensemble use cases

Q: What are the key use cases for Ensemble?


A: Here are some of the leading use cases:

Universal CPE (uCPE). With uCPE, service providers can realize the true value of NVF. They can replace
a stack of network appliances with software running on a standard COTS server. Doing so accelerates
service innovation and enables dynamic services for end users.

SD-WAN. Managed SD-WAN is a tremendous revenue opportunity for service providers. With
Ensemble SmartWAN, service providers can deploy software SD-Wan VNFs on COTS servers rather
than old-fashioned closed appliances. See “Q: Does Ensemble have an SD-WAN solution?” below for
more information.

Secure cloud connectivity. The virtualization of network functions has many benefits, but the biggest
is enabling applications that were not possible before. With secure cloud connectivity, the
virtualization of encryption functions enables end-to-end security, all the way from the customer site
to a public cloud provider. Doing so makes hybrid cloud applications secure and suitable for mission-
critical applications.

More Info:
 Verizon Adds Ensemble to Its Virtual Network Services uCPE Solution
 Verizon uCPE Case Study
 Universal Customer Premises Equipment
 Realizing the Value of NFV with Universal Customer Premises Equipment (uCPE) (webinar)
 Ensemble SmartWAN: Transforming SD-WAN
 Don’t Use Hardware Appliances for SD-WAN
 Virtualised encryption: How it could be the killer app for NFV – and help with GDPR too

Q: Does Ensemble have an SD-WAN solution?


A: Yes. It is called “SmartWAN.” Ensemble SmartWAN is an open virtualized networking platform for
deploying and managing virtual SD-WAN and other NFV services at scale. The Ensemble SmartWAN
platform enables service providers to deploy SD-WAN as a software VNF or VNFs, and do so in an
automated and virtualized fashion, consistent with their forward-looking architectures. With this

28
approach, service providers can deliver new and more flexible VPN services to their customers at a
lower cost and with more features than today’s service offerings. What’s more, Ensemble SmartWAN
comes with a rich feature set for connecting off-net customers through the internet or wireless
networks.

More Info:
 Ensemble SmartWAN: Transforming SD-WAN
 Don’t Use Hardware Appliances for SD-WAN

Q: Does Ensemble have live deployments?


A: Yes. We have announced wins at Verizon, Masergy, DartPoints and Indosat.
More Info:

Verizon
 No Shortcuts to NFV Success
 Verizon Adds Ensemble to Its Virtual Network Services uCPE Solution
 Verizon uCPE Case Study
 The Real Innovators of Networking: Chad Thompson, Verizon
 Realizing the Value of NFV with Universal Customer Premises Equipment (uCPE) (webinar)

Masergy
 The Real CTOs of NFV: Tim Naramore, Masergy
 Real-World NFV, Real Lessons Learned

DartPoints
 The Real CTOs of NFV: Satya Baddipudi, DartPoints
 Bringing the Cloud to a Business Near You

Q: What are the benefits of NFV to the end user?


A: End users may not care about NFV, but NFV-based services to bring them benefits. Here are some
examples:
 NFV can provide consolidation of a stack of appliances down into one server. This saves space
and power at the end user location.

29
 Sofware-centric services can be deployed in real-time under user control. This gives end users
live control over their telecom services.
 NFV can deployed on uCPE and combined with zero touch. Doing so enables accelerates
service turnup with local sourcing/sparing of servers and/or bring-your-own-device models.
 NFV enables moving from appliances to software, enabling new applications that were not
previously possible. For example, software-based encryption and service-assurance can be
deployed in public cloud infrastructure, enabling secure and assured hybrid cloud applications.

More info:
 Why Do Customers Care About NFV? (video)
 Software and Support Optimize SD-WAN for Enterprises

Revision 2 – December 2017

30

You might also like