Robert Hilliard Barrow, a US general serving during both world wars, used to say "Amateurs talk about strategy, professionals talk about logistics", encapsulating in a single sentence a somewhat obvious but often overlooked concept: whatever assets are at our disposal, their usefulness depends primarily on our ability to supply them.
This holds true both in the military sphere, to which Barrow was actually referring, and in a purely business sense. Without materials fuelling production, factories simply stop working. Without merchandise delivered by couriers, shops would be empty. However, in a tech-driven world shifting towards digitalization, goods are not the only things moving around and powering the economy.
Information, digital content, and other intangible (but extremely valuable) resources are constantly gathered and shared, allowing companies to keep in touch with their staff, partners, or customers, coordinate complex business processes, and fuel their decision-making with "fresh data". The goal of corporate network management is precisely to handle this flow of virtual resources.
But what happens when such networks overextend their reach to connect a geographically distributed workforce or when their bandwidth is strained by a steady increase of information exchanged? These are some of the key challenges that network management professionals should be ready to face in the age of remote work. Let’s discuss the nature of these issues and explore some of the technologies and managed IT services necessary to keep enterprise network management up with the times.
An introduction to enterprise network management
While logistics aims to harmonize the flow of physical things, enterprise network management encompasses tools and practices for setting up the "supply chain" of digital assets and services across organizations. This includes setting up, monitoring, and maintaining reliable networks to connect multiple corporate devices and the applications running on them.
The leading role in this process is played by network administrators, professionals making sure that both LANs (local area networks) and WANs (wide area networks) work smoothly and without interruptions.
Network administrators can count on powerful network management systems integrating a variety of software applications designed to automate most key procedures, such as device monitoring and performance analytics. Indeed, complementing human supervision with software-powered capabilities represents a cornerstone of modern network management, as the networks these experts deal with are sprawling webs interconnecting huge amounts of different devices.
Delving into the network management routine
To define the primary tasks of network administrators and the essential elements of the network management discipline as a whole, we can rely on the ISO Network Management Model, which identifies five major areas of activity: fault, configuration, accounting, performance, security (summed up in the acronym FCAPS).
- Fault management: This includes performing diagnostic tests and analyzing the device's error logs to identify abnormal behaviors or malfunctions and correct them through targeted interventions.
- Configuration management: The aim, as you may expect, is to configure the network devices (for example, via circuit provisioning) and keep track of any setup change that may occur, especially when scaling up the network.
- Accounting management: This area focuses on gathering network resource utilization data from each department for proper billing. However, that's not the case for all companies, as those not relying on billed networks usually replace accounting with administration to perform alternative tasks, such as software backups and user permission management.
- Performance management: This involves procedures aimed at streamlining network functioning and avoiding performance drops, including efficiency monitoring via SNMP management systems to analyze parameters such as response time or throughput and spot any fluctuations through machine learning-based anomaly detection before a fault actually occurs.
- Security management: The goal is to implement access control mechanisms such as authentication, encryption, network firewalls, and intrusion detection systems, along with proper security audit and event reporting procedures.
What the market tells us
As you can see, corporate network management represents a complex and multifaceted set of functions requiring not only proper expertise, but also a wide range of technologies and software solutions whose role seems set to grow further. In this regard, Statista reported that the global network management systems market was valued at $6.7 billion and may reach $12.2 billion by 2027.
In its Unified network management market 2022-2027 report, Mordor Intelligence provided further insights into the network management market as a whole, estimating that its unified value was $8.01 billion in 2020 and is likely to reach $21.63 billion by 2026 at a CAGR of 18%. The study also pointed out that such rapid growth might be driven mainly by the increasing adoption of technologies like cloud computing and SDN architectures (a concept we'll investigate later), along with a greater focus on cybersecurity consulting and network analytics.
Such optimistic predictions should come as no surprise, at least for a couple of reasons. First, a long list of historical examples has clearly revealed the devastating business effects of network downtime in a hyper-connected economy like ours and the consequent need for a robust network management architecture.
Remember Facebook's network outage in 2021, which resulted in a 5% value drop in the company's shares? Obviously, we are talking about Facebook, a social media giant that literally finds its raison d'être in keeping people and organizations connected. However, similar episodes may cripple virtually any company.
According to Juniper Networks' 2019 Network Brownout Survey, typical organizations had lost more than $420,000 over the previous two years because of network outages or quality drops. Not to mention other collateral effects such as the mitigation costs to retain dissatisfied customers (such as discounts or other benefits), the economic damage connected to fines and lawsuits, the obvious reputational issues, and 30% productivity decrease due to poor communication among remote team members. And this brings us to the second point, namely remote work.
From local to remote: challenges and tech solutions
We've already mentioned the fact that digitalization, along with innovative working models involving professionals from remote locations, has heavily impacted the way we approach the entire corporate workflow, including network management.
This transformation, as most of us have experienced in the last two years, has been further catalyzed by the global pandemic and consequent social distancing measures, which have forced companies to extend their network-related operations and ensure connectivity for an increasingly decentralized workforce.
In the first months of the pandemic, this massive shift caught many enterprises unprepared and caused severe network monitoring and management gaps, perhaps because decision-makers ignored the issue, believing that the switch to remote work would last just a few weeks. But according to Murphy's law, "anything that can go wrong will go wrong". Well, once again, it turned out to be true.
Fortunately, technology has come to our aid, providing us with a full range of new tools to:
- Address the loss of end-to-end control over dispersed workplaces and personal devices.
- Oversee this expanded operational scenario, especially in terms of network performance and security.
- Support the growing reliance on high-bandwidth applications for business productivity (such as teleconferencing).
Let’s take a look at some of the key network management technology trends implemented to tackle remote work challenges.
1. SD-WAN, a wide area network on steroids
We've previously mentioned the growing adoption of SDN (software-defined networking), a network management architecture separating networking control mechanisms from network hardware and forwarding functions to ensure more dynamic network configuration and easier administration.
This flexible approach, which actually implements some of the key concepts of cloud computing into network management, has been fully embraced by SD-WAN, a type of wide area network providing dynamically scalable performance and on-demand remote access from software-based entry points. In simple words, employees can access their enterprise network without using company-issued devices and rely on a stable connection even while operating very demanding applications.
These peculiar features make software-defined networks the ideal candidate for streamlining remote work. Indeed, MarketsandMarkets reported that the global SD-WAN market may grow from $1.9 billion in 2020 to $8.4 billion by 2025, while Fortinet pointed out that 64% of corporations are planning to adopt the SD-WAN technology.
2. Bringing network management in the cloud with SASE
We may consider SASE technology (which stands for Secure Access Service Edge) as a kind of common ground between WANs and the cloud, since it leverages cloud computing to offer network control and security capabilities directly to the connected device via the cloud without passing through a data center.
This model ensures the typical scalability and adaptability offered by cloud services and allows network administrators to easily implement their enterprise security protocols in the company’s SaaS applications. Regarding security, SASE technology relies on digital identities, which can be assigned to different network elements, be they users, devices or systems.
According to Gartner’s Hype Cycle for Cloud Security, SASE is still in the early stages of adoption, but its impact on enterprise network management is expected to grow in the near future. So much so that by 2024, over 40% of companies may have developed a solid plan to deploy this technology, compared to only 1% in 2018.
Nevertheless, it is essential to note that implementing the SASE architecture typically implies moving workloads to the cloud, for example, to a platform such as Amazon Web Services (AWS). This, in turn, means that developers should be familiar with AWS migration best practices or similar recommendations in case you are opting for another cloud provider.
3. Zero trust architecture, aka better safe than sorry
It might sound like technology out of a spy novel, but the zero trust security model is definitely real, and its adoption has grown over time (actually quadrupled between 2019 and 2021, according to TeleGeography's estimates) to better deal with the increased amount of cyberattacks and data breaches affecting enterprise networks. The shift towards remote work and the consequent over-extension network, as well as the more intensive use of not-so-trustworthy personal devices from home office locations, has further exacerbated this problem.
Zero trust systems try to mitigate such issues by requiring validation for any user or device, even when previously connected to the enterprise network, and implementing a full array of control mechanisms. These may include granular access control to define different access levels, IAM (identity and access management) to implement strict access policies, microsegmentation to keep workloads separate and individually secured, and more.
4. Making network management smarter with AI and IBN
One last trend worth mentioning in our overview encompasses all those tools and mechanisms powered by artificial intelligence and deployed to streamline and automate the most time-consuming enterprise network management operations.
Among the major use cases of AI in this discipline, we can find automated systems for user authentication and device monitoring (the former representing one of the most common uses cases for artificial intelligence in cyber security), machine learning-based analytical systems for predictive maintenance, and intent based networking (IBN). This last methodology taps into machine learning to autonomously develop policies and perform routine tasks based on experience and user feedback. This means, for example, that an IBN system may:
- Process a user request with advanced analytics to understand their intent.
- Build an abstracted model of the network representing its functioning.
- Orchestrate a potential network configuration to meet user needs (such as making a certain server reachable via the network).
- Validate the request if it's actually feasible and suggest the configuration to the network administrator.
- Wait for approval and deploy the configuration.
- Analyze the outcome and fine-tune the configuration for future reuse.
Breaking down borders with network management
Over the past two years, enterprise network management has evolved to face the challenges of remote work, including the introduction of new tools and technologies. Considering the increasing scope of network management due to the growing complexity of enterprise networks, the combination of human and machine will become a cornerstone of this discipline, and we can reasonably say that the aforementioned trends are unlikely to reverse.
Indeed, the effects of the pandemic on our way of approaching work, according to several researches, seem set to persist. Based on PwC's 2021 US Remote Work Survey, for example, 52% of employers believe that remote work increases productivity and most executives are planning new investments to support hybrid working.
This also implies that managers shouldn't make the same mistakes mentioned above, namely relying on inefficient, outdated network management models while waiting for their employees to come back to the office. After all, even in network management, our approach is changing, all things are flowing. Or, as Heraclitus said, "Panta rei".