Introduction
Telecom operators often provision standard telecom services such as internet, voice, and L2/L3 VPNs without properly documenting them in a service inventory system. When telecom operators decide to initiate a service assurance project, a populated service inventory becomes a prerequisite. Each service instance must be precisely mapped to its underlying resources. For instance, a standard internet service instance must be linked to resources such as a CPE device, a port on the access device, a public IP address, VLAN, and more.
Given that telecom operators can have millions of customers by the time they implement a service inventory, documenting all existing service instances becomes a nearly impossible task without some form of automated service discovery mechanism.
In this blog, we explore the crucial concept of service discovery in telecommunications networks and its essential role in maintaining efficient and reliable network services. We cover what service discovery is and how it operates within modern networks, various methodologies and technologies employed in service discovery, and how it enhances network performance, reliability, and scalability.
Additionally, we discuss the interplay between service discovery and service assurance, how effective service discovery can support service assurance efforts, and the common obstacles in discovering services along with strategies to overcome them. We also highlight best practices for implementing effective service discovery in telecom networks and explain how UMBOSS optimizes and supports these processes.
By the end of this blog, you'll have a comprehensive understanding of service discovery and its key role in the complex OSS/BSS ecosystems.
What is Service Discovery?
Service discovery is the process of searching for predefined configuration patterns on network devices, EMS systems, or platforms and extracting relevant parameters to populate a service inventory and map service instances to underlying resources.
In other words, service discovery involves identifying specific configurations and settings that indicate the presence of a service on a certain resource (device, system, platform, etc.). Once these configurations are detected, essential details such as IP addresses, device information, ports, VLANs, and provisioned bandwidth are gathered. This information is crucial for accurately documenting and mapping service instances within a service inventory system.
Service discovery is especially crucial for telecom operators as it enables automatic and accurate documentation of service instances, saving significant time and reducing human errors. It ensures that all service instances are correctly mapped to underlying resources, which is essential for all other service assurance functions. Moreover, auto-discovery is the only feasible approach for large-scale networks with millions of customers and services, as manual documentation throughout the network would be impossible.
How does Service Discovery Work?
Service discovery is a complex process that requires customization to fit the unique needs of each telecom operator. While in other industries a universal tool could be used (e.g., detection of micro-services in a software platform), telecom operators have unique service templates reflecting their individual network configurations, service offerings, and operational practices, which leaves no room for universal service discovery tools.
Development of the service discovery mechanism is a process that begins with the identification of these unique templates and the specific business logic behind the service provisioning. The business logic, explaining how services are configured and managed within their network, must be provided by the network engineers within the telecom.
These business rules are necessary for developing tailored service discovery algorithms that will properly find and map instances of services to the resources. These algorithms extract relevant parameters such as IP addresses, device information, ports, VLANs, etc., which are then populated in the service inventory.
The concept of Service Discovery
Algorithms for service discovery need to be carefully designed to accurately capture the unique provisioning templates for a particular telecom. This generally involves iterative testing and refinement, as these algorithms must be true representations of real-world configurations and sometimes-operational nuances of the network.
In real life, service discovery projects are highly collaborative efforts. Network engineers work together with software developers to translate the business rules into functional discovery algorithms, with the goal of making the discovery process both accurate and efficient enough to handle the scale and complexity of modern telecom networks.
Essentially, service discovery aims at the automation of the documentation of service instances, such that each instance is exactly mapped to its underlying resources. This automation is critical for maintaining an up-to-date service inventory, which in turn is essential for effective service assurance and network management.
Types of Service Discovery Mechanisms
Service discovery involves several approaches, each tailored to extract specific parameters and configurations necessary for documenting and managing service instances. Here are the primary methods used:
Parsing of Device Configurations
This method involves searching for predefined configuration patterns within network devices and extracting relevant parameters related to specific service instances. For example, if IP addresses are statically configured on a network device, these can be extracted directly from the device's configuration. Other parameters to identify and document include rate-limit settings, VLAN IDs, VRF names, and route distinguishers. This method works especially well when trying to document static configurations that have been explicitly stated in the device configuration settings.
Current Network State Analysis
In addition to the static configurations, service discovery can analyze the current running state of the network by examining neighboring nodes, MAC address caches, routing protocols, and other dynamic parameters not explicitly stated in the configuration but observable once the device is active in the network. This approach helps in identifying real operational data that is key to understanding current network topology and calculating end-to-end service paths.
Extracting Information from Vendor-Specific EMS Systems
Quite often, certain parts of the network are managed by different Element Management Systems (EMS) provided by the same vendors that supply the network equipment. For instance, one EMS system can be used for the management of a mobile network, another for a DWDM network, and a third for a GPON network. In such cases, the ability of the service discovery engine to integrate with all of these EMS systems becomes crucial for extracting service-related parameters which must be populated into the service inventory.
Extracting information from Specialized Platforms
Services are not always provisioned just at the network layer; sometimes they involve specialized platforms like voice platforms, DNS, DHCP, IPTV platforms, etc. In such cases, the service discovery engine must be integrated with these platforms to extract relevant service parameters. This way, all the service parameters, including those not present in device configurations or EMS systems, are documented and correctly mapped.
Hybrid Approach
Normally, in the case of complex services spanning across multiple network segments, EMS systems, and specialized platforms, no single service discovery approach would be sufficient for end-to-end service discovery. In a hybrid approach, different methods are combined to extract various service parameters from diverse sources. Furthermore, these parameters are stitched together to build a precise and comprehensive documentation of every single service instance. This ensures that all aspects of the service are captured, providing a complete view for effective service management and assurance.
What are the Benefits of Using Service Discovery?
Implementing service discovery in telecom networks provides a lot of benefits, especially in the context of having an up-to-date and accurate Service Inventory. The advantages are as follows:
Efficiency and Time Savings
Automation of the service documentation process saves considerable time in populating service instances in the service inventory. Compared to manual documentation—which could be a nightmare given that millions of service instances can be active across different domains like IP/MPLS, microwave, DWDM, GPON, etc.—automated discovery reduces this time by more than 95%.
Additionally, by automating the process of service discovery, the need for human resources for service documentation is drastically reduced. In contrast, pure manual documentation would require significant manpower.
Accuracy and Reliability
Manual documentation may be subject to human errors that can degrade the accuracy of service inventory. Such errors are eliminated in automated service discovery, with accurate documentation of each service instance and its mapping to its underlying resources. Automated discovery can also be periodically run to perform sanity checks, ensuring that any changes in service provisioning, deprovisioning, or modifications are correctly reflected within the service inventory.
Scalability
Telecom operators typically manage networks with millions of end-customers and service instances. With automatic service discovery, scalability is inherent, efficiently handling large-scale network documentation burdens. This makes it practical to maintain an up-to-date Service Inventory as the network evolves and grows.
Improved Service Assurance
Service assurance heavily depends on an accurately updated service inventory. With all service instances precisely mapped to their underlying resources and populated in the service inventory, identifying and resolving service issues becomes straightforward, leading to improved network performance and reliability. Accurate and timely service documentation enables a telecom operator to manage its network proactively, allowing potential issues to be avoided before they affect customers.
Cost Savings
Automating the service discovery process reduces the need for extensive manual labor, leading to significant cost savings in operational expenses. On the other hand, accurate service documentation helps in faster troubleshooting and resolution of network issues, minimizing downtime and associated costs.
To illustrate the cost and time savings, here is a use-case example for one small telecom. Manually documenting 300,000 service instances at an average of 2 minutes per instance in the service inventory would require a total of 600,000 minutes (or 10,000 hours). Assuming the assignment of 5 full-time workers to this task, it would take approximately 250 working days to complete. With an average annual gross salary in the European Union of €37,900 in 2023, the total cost of the project would be approximately €189,500. This example represents the high effort and duration a manual documentation may take, hence reinforcing the benefits of automating service discovery.
With automatic service discovery, it becomes a far more efficient, accurate, and scalable way to manage the network for a telecom operator.
How Service Discovery Enhances Network Service Assurance?
Service discovery guarantees that the service inventory is always accurate and up-to-date. It automates the process of documentation, eliminating human errors and ensuring that every change in the network is documented within the inventory. An accurate service inventory, where each service instance is correctly mapped with its underlying resources, is key to any service assurance system.
Service impact analysis depends on understanding the dependencies between different network components and services. With a populated service inventory, monitoring systems will be capable of accurately assessing how the failure of certain resources impacts customers' services.
Effective service quality management (SQM) also relies on an accurate service inventory, as this enables the SQM system to correlate performance data that it collects with specific service instances and the underlying resources.
What-if analysis involves the simulation of various scenarios to predict their impact on network services. This is impossible without a comprehensive and accurate service inventory. Service discovery provides all the necessary data for performing predictive analysis and making data-driven decisions related to network changes and upgrades possible.
Using an accurate inventory, the telecom operator will be able to detect issues well in advance and prevent them before they can affect its customers. Service discovery ensures that all service instances and their dependencies are documented, enabling proactive issue resolution and minimizing downtime.
Clearly, service discovery provides critical inputs that help telecom operators maintain a high-integrity service inventory. The accurate service inventory is a key enabler for a rich set of service assurance functionalities, such as service impact analysis, service quality management, and what-if analysis.
Challenges in Implementing Service Discovery for Network Service Assurance
In general, the implementation of service discovery is a complex task. It typically requires custom development for each telecom operator separately because each telecom has its own specifics that must be addressed.
For example, two telecoms may offer the same service from the end-user perspective, such as 500 Mbit/s Internet access, but the service discovery processes in these two telecom operators are very different because each telecom operator has its own network architecture, uses different network equipment, and different service templates for the activation of such service.
Another common problem is the inconsistency of service provisioning within the telecom. Two instances of the same service type (e.g., 500 Mbit/s internet access) may be provisioned differently for two customers of the same telecom. This happens mainly when services are manually provisioned by different network engineers at different points in time, with slightly different configurations. Such inconsistencies make service discovery cumbersome, as the algorithms must consider these differences in order to provide accurate documentation of the services.
Besides that, telecom operators use different device models over time, where each model has its unique provisioning template that might vary depending on firmware version, software versions, number of ports, etc. Algorithms used for service discovery need to be robust enough to cope with different configuration templates. This adds another layer of complexity because the algorithms have to be continuously updated and refined in order for new devices and configurations that get introduced into the network to be accommodated.
The challenges in implementing service discovery are many, and addressing these challenges requires sophisticated, adaptable algorithms tailored for each telecom operator separately.
Best Practices for Implementing Service Discovery
To develop and deploy effective service discovery mechanisms in telecom operators, it typically requires a joint effort from both the network teams of the telecom and software developers writing the discovery algorithms.
Basically, the development of service discovery algorithms is an iterative process. This process involves much testing and refinement to handle the various exceptions and unique scenarios that could arise. It is this iterative process that will ensure the algorithms for discovery are robust and can correctly map instances of services against their resources in complex and dynamic network environments.
An agile approach will be proper for this type of project. Agility emphasizes the principles of flexibility, iterative progress, and close collaboration between stakeholders. This approach enables the service discovery mechanisms to be constantly adapted and improved in their effectiveness, keeping them current with the evolving network configurations and service templates.
Despite such clear benefits of the agile approach, many telecom operators still expect a plug-and-play solution. However, considering the complexity and variability of telecom networks, one-size-fits-all is far from realistic.
Each of the telecoms has its own unique network architectures, systems, and service provisioning practices that would call for tailored solutions. Therefore, setting realistic expectations for an iterative, collaborative process is crucial for the successful implementation of service discovery mechanisms.
To summarize, the best practices for implementing service discovery will involve close collaboration between network teams and software developers, embracing an iterative development process with continuous testing and refinement, and adopting an agile approach to accommodate the unique and ever-evolving needs of the telecom operator's network.
UMBOSS' Role in Service Discovery
UMBOSS plays a pivotal role in enhancing network service assurance through its comprehensive suite of modules, including the Service Inventory (SI), Service Quality Management (SQM), Fault Management (FM), and Performance Management (PM) modules. These modules collectively ensure that telecom operators can maintain a high level of service quality and reliability.
However, it is important to note that UMBOSS does not include a Service Discovery module out of the box. This is because there is no universal approach to service discovery, as each telecom operator has unique network architectures, systems, and service provisioning practices.
By integrating UMBOSS with a custom-developed service discovery solution, telecom operators can achieve a comprehensive and accurate Service Inventory, which is fundamental for effective network and service assurance.
Have any questions? Want to learn more? Get in touch and let us know how we can help. Send us a message or book a demo today.
