4 MIN READ | Service Assurance

Performance Monitoring Critical as Carrier Ethernet Adoption Grows

Roopa Honnachari
Oct. 24 2013

By Roopa Honnachari, Program Manager, Business Communication Services — ICT, Stratecast, Frost & Sullivan

While the widespread availability and the ability to support the convergence of applications–voice, video and data–is driving market demand for MPLS VPNs, customer control of network routing has led to the adoption of Carrier Ethernet services for critical and secure applications (e.g., financial services). Both MPLS and Ethernet networks allow communications service providers (CSPs) to prioritize traffic based on class of service, thus allowing real-time applications to consume more bandwidth as required. Customers are now asking for similar types of control and application visibility for the MPLS and Carrier Ethernet services that they purchase.

The need for performance monitoring is more pronounced with increasing adoption of Ethernet services to access cloud-based applications. Enterprises implementing private networks for cloud connectivity demand a high level of performance from their network provider, and CSPs are investing in solutions that enable them to proactively monitor the network and deliver secure multi-tenant reporting of real-time performance and assurance data.

Value-added services, including dynamic capacity allocation, M2M, secure cloud access, tiered services, bundled connectivity and a host of other offerings, require that CSPs ensure the quality of services at the infrastructure, connectivity and customer levels. CSP assurance transformation efforts are underway, but will take time to execute. In the meantime, CSPs are looking for flexible solutions that will support today's services while being agile enough to also support whatever comes next.

All network operators use Element and Network Management Systems (EMS/NMS) that include service assurance and performance management software provided by the Network Equipment Providers (NEPs). Likewise, all network operators have instrumented their networks with probes that read and collect status data for every transaction that occurs across the network. To understand quality and performance at the service level, CSPs have built software to correlate data from individual EMS/NMS and probe management systems. Unfortunately, the majority of network elements implement Simple Network Management Protocol (SNMP) for reporting; however, EMS/NMS, probe management, external data bases and data correlation engines do not, which requires CSPs to implement complex integrations. Most existing performance management correlation solutions were purpose-built for specific service offerings and, in some cases, specific customers, which makes them costly and time-consuming to operate and maintain.

With the home-grown solutions, CSPs have full access and control to manage and improve the service when they need it. The downside, as with any home-grown solution, is that they need to have the right staff with the right skill-sets to support the solution over its full lifespan. The challenge for CSPs is to implement end-to-end performance management and service quality correlation that can be used by multiple internal operations groups, and also deliver status and reporting to customers using a secure, multi-tenant platform.

Market trends vary globally, with the North American market consisting of a huge base of legacy home-grown performance management solutions. In emerging markets, there are fewer in-house and purpose-built tools for service assurance and performance management. Off-the-shelf solutions for Carrier Ethernet assurance and performance management that were not available just a few years ago are now being heavily promoted in emerging markets. However, until a CSP reaches a size that forces the implementation of automated solutions, it usually makes do with manual correlation of EMS/NMS and probe management data. Due to rapid expansion, numerous CSPs in emerging markets are investigating outsourcing of IT–meaning that the management of infrastructure, OSS/BSS and data centers is being done by a third party. Although CSPs that outsource IT retain control of strategy and product procurement, there are delays in implementing service- and customer-specific quality and performance monitoring due to contractual limitations.

Frost & Sullivan research indicates that enterprise customers that purchase Carrier Ethernet services routinely include requirements for reporting. SLA metrics most commonly requested include: latency, jitter, throughput, device, connection and interface availability, frame loss, frame delay and memory utilization. While CSPs confirm that enterprise customers' interest in customer reporting for Carrier Ethernet is rising, and most RFPs ask for it, only 20-25 percent of Carrier Ethernet customers are using such tools. For now, the requirement is important because the 20-25 percent of Carrier Ethernet customers using the reporting capability are among CSPs' largest and most lucrative. Going forward, that percentage is expected to increase as adoption of Carrier Ethernet for access to cloud-based applications increases. However, will a majority of enterprise customers be willing to pay an extra fee to get these reporting tools is something that is yet to be determined. CSPs have had mixed experiences, with a few of them successfully selling the reporting tools at a price, while others have been asked by customers to include this in the service pricing.

CSPs recognize that the data collected from and about the infrastructure is critical to nearly every aspect of the business. To that end, many are evaluating more efficient and accurate methods of collecting and distributing that data to the appropriate systems in the appropriate formats at the appropriate times. Measurement of service quality goes beyond SLAs to include individual locations, applications and users. A CSP's ability to understand which customers are affected by various infrastructure, service and product failures is critical to maintaining quality connections and ensuring performance. As CSPs deploy IP access networks, the reality is that all traffic is data, all products are applications, and every user is unique. Going forward, all CSPs, regardless of access network, will compete on product selection and performance quality.

Ms. Roopashree (Roopa) Honnachari is the program manager of the business communication services group at Frost & Sullivan. She monitors market developments in the wholesale and business network services (MPLS, Ethernet, Wavelength Services, SONET) markets and analyzes technology, competitive, pricing and regulatory trends. She has an extensive experience in building market revenue forecasts, pricing forecasts, market share analysis and competitive analysis. Roopa is also a contributor to the cloud computing practice and specializes in communications service provider cloud business models research. Roopa has more than 12 years of experience in technology marketing, executing and managing research/growth consulting projects, and providing strategic advice to companies on competitive strategy, new market entry and market share expansion.

 

Written By