What is Network Management?
Network management is the set of processes and capabilities used to provision, configure, monitor, operate, and maintain a network. A network management system helps ensure the overall reliability, performance, and availability of the network. The widespread adoption of cloud-managed networking for scale, agility, and efficiency has given rise to a wave of innovations in network management, resulting in modern, microservices-based, cloud-native network management platforms.
Network Management Explained
Network management is a multi-functional discipline that enables IT and network operations to ensure the overall reliability, performance, and availability of the network. With digital transformation, there is also an increasing focus on the role of network management in delivering high-quality end-user experiences.
Network management typically involves the following functions outlined by the FCAPS framework:
- Fault management: Detection, isolation and issue resolution to ensure optimal network connectivity and minimize downtime due to undetected anomalies. This is also often referred to as MRT (monitoring, reporting, and troubleshooting).
- Configuration management: Deployment and provisioning of network devices such as access points, switches, and gateways, including the ongoing monitoring of devices for configuration changes.
- Accounting management: Network utilization tracking to document bandwidth consumption by users and lines of business for cost management purposes.
- Performance management: Ongoing compilation and analysis of metrics related to network performance such as response time, packet loss, and throughput to ensure acceptable service levels are maintained. Alerts are generated when acceptable thresholds are breached.
- Security management: Network authorization to ensure only authorized users and devices can access the network resources to which they have rights. Most solutions incorporate capabilities that aid firewall configuration, vulnerability management, and intrusion detection.
HPE Aruba Networking Central provides a broad set of capabilities outlined by the FCAPS network management framework to provision, configure , monitor, and troubleshoot network devices.
Why is network management important?
The network underpins every connection in the business across remote locations, branch, campus, data center, public, and private clouds. It enables:
- ‘Always-on’ availability for services and critical business applications
- Secure access for devices, users, and clients, including IoT devices
- Seamless flow of data from where it’s generated (users and IoT devices at the edge) to where it is processed
Networking-related problems have been the single biggest cause of all IT service incidents over the past three years, with nearly 40% of organizations having suffered a major outage caused by human error1. Tools and systems for effective network management are, therefore, critical enablers of key business outcomes and high-quality digital experiences.
What are current network management challenges?
- Network Complexity – As hybrid work becomes mainstream and cloud adoption increases, traditional networks are becoming too complex to manage. Adding to this complexity is the often-independent management of WAN, wired, and wireless networks.
- Visibility – The rapid proliferation of IoT devices in support of digital transformation initiatives presents new visibility and security challenges, increasing cybersecurity risks.
- Tool Sprawl – To address a variety of workflows and functions across application, infrastructure, and cloud environments, network teams have anywhere between 4 to 15 tools on average2. With limited interoperability, large toolsets result in a high rate of manual errors, alert fatigue, and increased downtime.
How is network management evolving?
Businesses continue to modernize their network management systems to help navigate the challenges outlined above. Key areas of innovations include:
|AIOps and automation||Artificial Intelligence (AI) and Machine Learning (ML) on massive amounts of data (typically in a data lake) proactively derive insights across a range of use cases that increase network reliability, improve mean time to resolution (MTTR), and increase IT efficiency.|
|Flexible consumption||Cloud-based deployment models with support for flexible, subscription-based licensing models offers improved ROI, better financial control, and lower TCO.|
|Programmability||Microservices-based architectures with open, standards-based constructs, API support, and webhooks, enable a high degree of interoperability with other tools in the ecosystem and third-party integrations for IoT, OT, and IT convergence.|
|BYOD and IoT security||Built-in capabilities, such as AI-powered profiling, help address the inherent visibility and security challenges with BYOD and IoT, eliminating network blind spots and improving overall security posture. Security models such as Zero Trust use role-based access policies and ensure that no person or device is inherently trusted.|
|Intent-based networking||Software-defined, GUI-based workflows that automatically convert to CLI-based configurations reduce the complexity of the underlying physical network and express business intent.|
Cloud-based vs on-prem network management models
|Cloud-based||A network management solution hosted on the cloud and typically delivered as a Software as a service (SaaS) solution with flexible, consumption-based licensing. It is designed to take advantage of cloud computing to support sophisticated AI/ML operations and advanced analytics on data lakes that host a large volume and variety of anonymized data from across the vendor’s installed base.|
|On-premises||A network management solution hosted within the organization, using dedicated physical or virtual appliances that employees can access over the corporate IT networks. It delivers high-level control with all data residing on-premises, but it requires greater upfront investment and ongoing maintenance by the in-house IT team.|
What are the benefits of cloud-based network management?
According to Gartner, SaaS end-user spending is forecasted to reach $208 billion by 2023. Much like with other SaaS solutions, cloud-based network management delivers the following benefits:
- Single point of control: unifies network management across wired, wireless, and WAN networks, eliminating tool-induced management siloes to provide a single point of control and visibility.
- Scale: Scales rapidly across thousands of sites globally, without incurring CapEx for additional management hardware/appliances.
- Flexibility: Uses a modern, microservices-based architecture that provides flexibility and enables agile, automatic delivery of features without the need for planned maintenance downtime for software upgrades.
- AIOps and Analytics: Leverages data across the customer’s site and similar (anonymous) sites for AI/ML and advanced analytics, enabling anomaly detection, rapid troubleshooting, and optimization recommendations to drive operational efficiency.
- Programmability: Leverages RESTful APIs, streaming APIs, and webhooks for integration with the existing solution ecosystem.
When is on-premises network management preferred?
- Data sovereignty: Certain industries are governed by stringent data protection and/or sovereignty requirements that necessitate that all data generated are retained locally by the organization. Select countries also adhere to strict control of cloud solutions.
- Control and compliance: On-premises solutions offer a greater degree of internal control of the infrastructure, configuration, and ongoing maintenance. Organizations under strict regulatory compliance may require this higher degree of control.
- Architectural constraints: Business critical applications that leverage legacy design principles may inhibit the move to the cloud or make the adoption an expensive and time-consuming process
It must be noted, however, that on-premises network management solutions offer limited AIOps-based optimization and troubleshooting due to the limited size and variety of usable data and the need for expensive compute resources.