• For Specialists

    A blog for service providers focused on QoS, QoE, and network performance. Join us for in-depth analysis of industry news, technology, and solutions driving performance in mobile networks, cable MSO business services, cloud and data center connectivity, enterprise WANs, and financial networks.

  • Join us Live

    We host webinars each month covering topics from solution design to performance assurance technology and demos of our latest innovations. Join us engineers online for tailored insight and Q&A with our network engineers.
    Upcoming Webinars:

    Click Here
  • Learn on YouTube

    Accedian is the Performance Assurance Specialist for mobile networks, enterprise to data center connectivity, and service provider SDN. With dozens of videos covering network performance and QoE, our YouTube channel is a unique training resource.

    Watch Now

Monday, September 25, 2017

Accedian at INCOMPAS Show 2017: Booth 5

Join us at the Incompas Show, a networking and education event for communications technology professionals. It's happening October 15-17, 2017, at the Marriott Marquis in San Francisco, California. We're exhibiting our virtualized network performance assurance solution, SkyLIGHT, at booth 5. 

<click image to enlarge>

register.png     meetus.png

How to Become a Digital Service Provider in Four Easy Steps

Disruption in the mobile telecom industry is driving communication service providers (CSPs) to transform themselves into digital service providers (DSP). In embarking on this transformation, CSPs have three main goals in mind:
  1. Maintain and grow the subscriber base by extending lifetime value to customers
  2. Grow revenue through new business opportunities
  3. Drive out excess costs and inefficiencies, through operational excellence
Such a transformation requires agility in several key areas: network (infrastructure), services (including operational support systems/OSS), and customer experience (including business support systems/BSS). We will explore each of these in turn below, but there’s a prerequisite step to consider as well: learning from failure.

1. Learn from Failure

Thus far, digital transformation efforts on the part of CSPs do not have a stellar track record. TMF Forum reported that 60-70% of such programs fail and that 54% of CSPs say previous attempts at digital transformation have not succeeded.

<click image to enlarge>

A key reason for these failures is lack of ubiquitous, real-time insight into performance and experience. Such insight is possible, however. Rather than than be discouraged, CSPs should learn from past mistakes and move forward in a strong position to succeed using what they know now.

2. Transform Infrastructure for Network Agility

Creating a flexible and open infrastructure fosters competition and innovation from a broad vendor ecosystem, and enables the development and deployment of new revenue-generating services. This type of API-driven network, built in a way to enable flexible slicing and interconnect, benefits both the operator and its partners.

3. Transform OSS for Service Agility

Operators support systems (OSS) must evolve to enable rapid, dynamic service creation, provisioning, activation, and retirement of services. A service-oriented, dynamic software defined network (SDN) benefits the CSP through faster time to market for new services, and ability to react more quickly to market and competitive pressures. It also accelerates time to revenue.

4. Transform BSS for Customer Agility

To ensure subscribers get what they want, when they want it, and even before they realize want it, business support systems (BSS) must support an end-to-end, customer-centric approach using predictive analytics.

Bringing It All Together

Steps 2-4 above are inter-related aspects of network and IT transformation, and cannot really be addressed in isolation. They all require responsiveness to partners using the network through APIs; the operator’s need to rapidly scale out existing services and deploy new ones; and changes to user behavior, experience, or demand. All of this must be done without compromising the user experience.

Such a tall, complex order is possible—within a dynamically orchestrated system that has visibility into all functions, relationships, and behaviors of the network, it services, and its users. This vantage point provides the feedback needed for a closed-loop automation process between service creation and its management and optimization.

<click image to enlarge>

Agility is the key to all methods of understanding service, network, and user behaviors. Several requirements are worth noting here:

  • Lightweight, virtualized, monitoring instrumentation with immediate affinity to services as they are created. 
  • Fail-fast development methods--employed from the first beta--to provide critical feedback into the app experience as agile development adjusts the service until it’s ready for wide-scale deployment. 
  • OSS/SDN architecture that adapts to the network--in ongoing operation mode--to maintain optimal performance and circumvent problems, in context. 
  • At the BSS layer, use of data as input to analytics, as a means of improving customers engagement, experience, and service quality. Result: customer-centric insight drives network provisioning changes and new service availability. 
When these requirements are met, DSPs can find new revenue through a variety of opportunities, such as:
  • Mobile edge computing (MEC) as-a-service
  • Infrastructure-as-a-service (IaaS)
  • Distributed edge cloud
  • Slicing
  • Mobile payments
  • API-enabled network and resource consumption
  • IoT across federated mobile/cloud networks
  • Telehealth 
By positioning their network for cloud-type consumption, mobile operators stand to gain—for the first time—from key over-the-top (OTT) applications traversing their network. In facilitating this interaction with ‘subscribers’ (IoT service providers, content and transactional applications, etc.), MNOs become DSPs, turning their unique assets into an enviable “mobile cloud.”

Wednesday, September 13, 2017

Accedian at Network Virtualization & SDN Asia: Panel Speaker, Booth

Join us at Network Virtualization and SDN Asia (going on October 3-4, 2017, in Singapore); our VP Business Development APAC, Jason Roberts, is speaking on the October 4, 9:40am panel, "How Are Operators Effectively Monetizing SDN & NFV Services?" We're also exhibiting at booth NFV26. Here are the details!

Event: Network Virtualization and SDN Asia

Event Dates: October 3-4, 2017

Booth: NFV26

Panel Date: October 4, 2017

Location: Marina Bay Sands, Singapore

Panel: How Are Operators Effectively Monetizing SDN & NFV Services?

Panel Time: 9:40am

Panelist: Jason Roberts, VP Business Development APAC

register.png     meetus.png

Tuesday, September 12, 2017

Cable MSOs: How to Succeed With SD-WAN Using Virtualized Service Assurance

Cable MSOs hoping to effectively target enterprise customers must find a way to establish a nationwide service area, which means augmenting their own DOCSIS and Carrier Ethernet infrastructure with third-party access to extend their footprint. Increasingly, MSOs are adopting software-defined WAN (SD-WAN) to reach on- and off-net sites, in a uniform way over any access media. But, there’s a pitfall: SD-WAN appliances don’t offer standards-based test, turn up, and monitoring functions required to offer service level agreement (SLA)-grade services.

Instead, SD-WAN solutions use proprietary monitoring and reporting methods, which don’t interoperate with existing network equipment. Problem is, SD-WAN may only be required in certain enterprise customer locations, so any implementation has to interact seamlessly with traditional service delivery methods, which means using standards-based techniques.

How to address this issue?

Virtualized test probes and test reflectors can cost-efficiently replicate network interface device (NID) functionality, bringing the needed turn-up testing, monitoring and operations & maintenance (OAM) functions to SD-WAN endpoints. Virtualized instrumentation uplifts SD-WAN with carrier-grade functionality, making it interoperate with existing network infrastructure, operations procedures, and support systems.

Assuring the SD-WAN Service Lifecycle

The SD-WAN service lifecycle has three main phases, consistent with the MEF’s established model for Carrier Ethernet connectivity:

  1. Deployment: provisioning and service activation testing (SAT)
  2. Performance monitoring and SLA reporting: collecting and presenting key performance metrics
  3. Troubleshooting: techniques to identify, isolate, and troubleshoot service issues

<click image to enlarge>

Metro Ethernet Forum Service Lifecycle 

Approaches to Virtualized Performance Assurance

To effectively assure all of these phases, MSOs may choose to use one of two approaches:
  1. Centralized performance monitoring architecture using virtualized performance assurance controller (vPAC) virtual network functions (VNFs) as probe generators, with a lightweight, stand-alone software agent that instruments the entire network in a software-only implementation. 
  2. Network-embedded architecture that employs small footprint, programmable performance assurance hardware modules (vCPE modules) augmented by virtualized performance assurance functions hosted on a centralized vPAC. 

The first option, because it’s software-only, is less precise and has a smaller feature-set than that offered by vCPE modules. However, it is well-suited for deployments where performance assurance using standard-based protocols is needed, but the added-benefits offered by the NFV-enabled modules are not required.

To facilitate integration with existing operational support systems (OSS), network management systems (NMS), and VNF orchestrators, either approach requires four key elements:
  1. A test session controller
  2. A test packet generator
  3. A test packet reflector or receiver
  4. Precision timestamping
Each approach is discussed in more detail below.

Software-Only Virtualized Performance Assurance

This option provides unprecedented deployment speed and agility, through its ability to remotely and centrally deploy, configure, and run everything needed to instrument an existing network, on-demand, with minimal expense. Standards-based monitoring methods integrate the network itself into a ubiquitous instrumentation layer. With this visibility centralized in data centers shared with SDN control and big data analytics, providers have an integrated foundation to deliver a new level of customer experience.

Here’s how it works:
  • The vPAC assumes all session setup, control, and sequencing functions, as well as results analysis and reporting to file servers. vPAC instances (manifested as VNFs) are deployed and orchestrated seamlessly with the network service descriptors, allowing fully-automated setup and assurance of virtual service chains.
  • The lightweight software agent VNF has two functions: 
    1. Offers reflection capabilities, instrumenting the network with any orchestrator while easily running unprivileged on any Linux based operating system.
    2. Enables bi-directional measurements, unrivaled metrics set, measurement granularity, and third party interoperability—features unavailable when using built-in standard open-source tools (such as ICMP ping) or even proprietary measurement methods offered by SD-WAN vendors.
Enhanced Performance Assurance with NFV-Powered vCPE Modules

This option basically consists of pairing a centralized vPAC with network-embedded vCPE hardware modules, in order to virtualize as many customer-located networking functions as possible while retaining minimum hardware needed for service delivery, consistent with performance, reliability, and quality of experience (QoE) expectations. As noted earlier, this offers more precision and a larger feature set than a software-only implementation. Yet, compared with traditional hardware-based approaches, instrumenting a network in this way is a very affordable, fast to deploy option.

An example of this vCPE strategy is illustrated below in comparison side-by-side with traditional CPE; here, local networking functionality (e.g. firewall, PBX, routing) is virtualized to software-based VNFs, hosted on low-cost commercial off-the-shelf (COTS) servers or cloud infrastructure

<click image to enlarge>

vCPE: Traditional vs. Virtualized Customer Premises Equipment Example

In the context of SD-WAN, this approach can be used to introduce customer premises-located performance monitoring, turn-up test, service OAM (SOAM) and troubleshooting functionality, which—in the case of fiber business services—is normally provided using a NID. Reducing hardware appliances required at the branch site is a key benefit of SD-WAN; where installing a traditional standard NID along with the SD-WAN appliance may not be a feasible CPE option.

NFV-powered hardware modules can offer the same level of performance monitoring precision, as well as loopback and full line-rate turn-up test capabilities at a fraction of the cost of a NID, making this approach an economically viable fit when deploying SLA-grade business services over SD-WAN.


Whether deployed as software-only or using vCPE modules, all SD-WAN lifecycle phases can benefit from a flexible, NFV-based performance monitoring solution that scales beyond the footprint of the SD-WAN cloud and is capable of sending performance flows from any starting location to any destination in the network infrastructure.

Such a solution can be used to:
  • Cover large scale hub-spoke and full-mesh topologies with active, microsecond accurate, standards-based performance monitoring towards thousands of endpoints continuously.
  • Bring standards-based turn-up testing, monitoring, and OAM functions to all SD-WAN endpoints, by adding NFV-enabled vCPE modules or orchestratable lightweight software agents. Since the solution is standards-based, standard networking devices can also act as responders to performance monitoring flows.
  • Monitor micro-outages, one-way delay & variation, and SLA compliance by delivering precise and granular metrics.
  • Centralize test control and automation, integrated with existing OSS, by pairing vPACs with NMS solutions.
  • Deliver a new level of performance monitoring (PM) workflow automation with results centrally stored for comparison to predefined QoS templates or SLA levels. Tests—conducted one-way or bi-directionally, in an end-to-end or segmented manner—can be scheduled on demand or triggered by service endpoint installation.
  • Provide open access to turn-up data and results—including customer-ready reports reflecting their specific SLAs—using the API. 
All of these applications support MSOs’ goal of delivering SD-WAN managed services to enterprises, over large, diverse geographic areas with the same level of quality as they do with traditional WAN offerings.

Tuesday, September 5, 2017

3 Reasons IoT is a Game-Changer for Mobile Operators

1. IoT network demands are at odds with multimedia services

For a long time now, mobile operators have invested heavily in network bandwidth and speed, in order to deliver voice, video, and other real-time multimedia services. Unfortunately, IoT services instead tend to be low bandwidth (but with potential for burstability) and low data (but high-priority), delivered to low power, low cost devices. 

Can one network serve both types of services? And if the answer is no, where does that leave mobile operators?

Some operators are responding by getting out of the IoT business, while others are looking to partnerships with low-power, wide-area (LPWA) companies as a way to stay in the game without having to build IoT-capable networks or adapt their existing architecture for IoT services.

2. Low power, wide area (LPWA) providers pose a competitive threat
For those operators that do want to compete in the IoT space, especially those intending to adapt their own infrastructure to meet the needs of IoT services, much effort is being put into creating a cost-effective alternative to competitive threats from LPWA companies--as well as cable companies, cloud providers and fixed-line players. Everyone, it seems, wants a piece of the potentially very lucrative IoT pie.

For example, operators are collaborating with the 3GPP, chipset vendors, and network vendors to get NB-IoT standards pushed out as soon as possible.

Building hybrid networks that combine traditional cell and non-traditional technologies is also a promising path forward, but depends on the ability of operators to adopt virtualization.

3. Adopting virtualized instrumentation and orchestration is becoming imperative
In order to efficiently monitor and manage resources for diverse services, operators must quickly adopt network virtualization and roll out a comprehensive instrumentation layer. As with a multi-lane highway previously used by cars only but suddenly opened up for use also by trucks, bicycles, and pedestrians, there has to be some way to intelligently direct traffic according to its specific requirements. Else, chaos would ensue.

Of course, control is only possible with complete visibility into what’s happening in all parts of the network. This must be tied with real-time orchestration to quickly respond when issues crop up.