We live in a digital world. Communications Service Providers (CSPs) depend on software-based network services to engage customers, automate processes, drive innovation and unlock new business models. CSPs are increasingly responsible for delivering mission-critical services and must ensure that these services are always available and responsive. Increasingly, this means agreeing to formal SLAs. While the benefits are many, the risk in this evolution is significant – including the financial viability and reputation of the CSP.

Yet despite this, many CSPs continue to operate in dangerous silos. They use multiple tools and manual approaches – many of which are offered by network equipment providers themselves – to monitor the health of their networks. The result? A patchwork approach that leads to limited visibility and cumbersome manual interventions that become barriers to continuous, high quality and low cost operations.

As a multi-disciplinary services partner to dozens of CSPs, Persistent Systems has long recognized the need to deliver always on network operations with the latest performance monitoring. Based on our experience, we have identified the following five key lessons that every provider should recognize if they want to deliver flawlessly while reducing costs:

  • 5G has changed the game: It is impossible to overstate the impact of 5G. In a 4G world, much of the network usage is focused on low criticality, supporting applications, oriented around buying, selling, web browsing and entertainment. Here, some latency can be tolerated and accepted. Yet 5G-oriented applications are markedly different – these mission critical applications require ultra-low network latency. Applications like self-driving taxis (which are already a reality in China) must be supported by a robust, well managed, 5G network with impeccable coverage.
  • Data growth is exponential – more than most legacy solutions can handle: The migration from physical to Virtual Network Functions (VNFs) and network expansion is leading to a tidal wave of event data that can be collected for monitoring purposes. With many CSPs continuing to operate in silos using multiple tools to monitor health, networks are increasingly drowning in disconnected, redundant data. The danger? New events become noise as network staff are forced to manually filter and correlate data looking for patterns.
  • There is no performance monitoring without AI: To accommodate this massive explosion in data, and the need for real-time as well as batch processing, CSPs need a next generation approach with a modernized AI and ML stack (e.g. Cassandra, Hadoop, Spark, Kafka). This means a solution that uses AI and ML to provide automated consolidation and grouping of events into fewer (more actionable) tickets, improved visibility into hidden patterns and increased responsiveness with automation.
  • Scalability is key: Historically, CSP operations teams have worked hard to anticipate major network events and manually redeploy their network events accordingly. However, these activities can now be automatically handled with data feeding NFV (network function virtualization) networks with orchestration capabilities. The new reality? Solutions can now scale automatically during peak hours, leveraging both on-premise and public cloud infrastructure from multiple vendors.
  • Heterogeneous networks are on the rise: Different heterogenous networks are becoming more common. The complexity of these networks and topologies makes the resource mapping even more challenging. Many CSPs find a vendor-neutral performance monitoring solution important for its operations team to easily integrate multiple equipment providers and gain visibility across proprietary monitoring tools that exist in various network environments that make it difficult to correlate and link events.

As networks get more powerful, their use becomes more demanding and so the challenges of running them becomes more complex. This demands a new approach, automating old processes and utilizing the latest in AI and ML technologies to better manage environments.

With Persistent’s extensive experience transforming legacy systems for a digital-first world, combined with its extensive partnerships with IBM, cloud providers and automation leaders, we can help CSPs navigate and address the new world of hyper networking.