And so now on to your technical building blocks—the six imperatives
outlined on the last page.
1. A scalable network
In this first building block, the very first must-do is the full environment discovery: a review of baseline architecture and common network functions, as already mentioned as a fundamental in developing a technology transformation plan. You then also need to identify potential future operation modes through tabletop exercises to prepare for at least two to three major event scenarios (e.g., pandemic, natural disaster, disruption in supply chain). This will enable you to build an architectural roadmap that will support your business objectives.
You will soon receive an email with a link to confirm your access. When you click to confirm from your email, your document will be available for download.
If you do not receive an email within 2 hours, please check your spam folder.
You may now close this message and continue to your article.
Perhaps the most important considerations for building a scalable network are flexibility and scalability. This is where SDN comes in.
But perhaps the most important considerations in this building block are flexibility and scalability. For organizations to stay relevant and competitive in a global and turbulent market (both during and after COVID-19), organizations of every shape and size require a network that can support the dynamic and on-demand needs of their users and applications—for example, more bandwidth to support seasonal sales requirements, more VPN connections or cloud access to support work-from-home requirements. This is where software-defined networking (SDN) comes in.
Traditionally, companies have purchased different devices to deliver networking functionality—routers, switches, firewalls, load balancers, etc. With SDN, the network is now software based, and these functions are accomplished virtually. SDN enables organizations to gain network flexibility and agility and also makes their network much more elastic, as software can also be used to increase capacity or introduce new functionality. SDN supports innovation in a way that static networks cannot.
While one can have an SDN without scale, you can’t have a truly scalable network without SDN; SDN is a prerequisite for scalability. A scalable, SDN-enabled network has four clear characteristics:
- Application/user awareness built in
Within any organization, inconsistencies exist across importance of applications and users—and so segmentation matters. Not all applications have the same level of importance to an organization (e.g., an enterprise resource planning application versus web browsing), and not all users have the same level of importance to an organization (e.g., doctors in a telemedicine world, traders in finance). Software-defined WAN (SD-WAN), a subset of SDN, has application awareness built in, giving the organization the ability to prioritize applications or users on the network quickly and efficiently. And levels of sophistication increase dramatically when combined with deeper network visibility (which we discuss in depth below).
- Network function virtualization (NFV) for rapid addition of scale and capability
VPNs deployed as NFV can scale much more easily and quickly (we’re talking hours/days instead of weeks) via APIs, when compared to the need-to-deploy hardware. We know of many customers who had to dramatically increase VPN throughput to support their shift to working from home. For example, two European-based enterprises have started leveraging NFV to easily scale VPN connections across multiple network nodes to >1 Gbps per node to account for increased remote access traffic. In another example, an India-based services company rapidly stood up a software- defined VPN to respond to users at home for the first time. While it’s true that to date, adoption of NFV has been slow, organizations need to prepare for when the tide will turn. Gartner states that “39% of enterprises cited technology and vendor risk as the major blocker to wider adoption of NFV-based services. Leading Network Service Providers are striving to offer greater choice of hardware platforms and software functions while attempting to ensure quality and reduce complexity.”1
- API-driven orchestration
With APIs comes the ability to manage bandwidth and traffic routing across both public and private networks with better visibility. Visibility tools, which we will discuss below, will sense any infrastructure anomalies or issues—which could be caused by a spike in customer demand or a rapid shift to working from home—and feed predefined workflows that then invoke API calls to network elements (whether NFVs, traditional MultiProtocol Label Switching (MPLS) and internet networks or 5G). An organization is therefore able to scale up resources to accommodate these anomalies or issues without human intervention. For example, a large U.S.-based global retailer scaled up network connectivity via API calls for action in an hour, compared with the several days required by a traditional order route. In another example, a European-based leasing company leveraged API-driven policy updates to change how it prioritized applications to support business-critical traffic during a period of network congestion.
- Flexible access to cloud
The final aspect to consider when building a scalable network is cloud interconnection, and this means public and private networks into the hyperscale cloud and software as a service (SaaS) providers. Applications and application elements will often continue to reside in different clouds but need to be quickly and easily accessed. Cloud and SaaS providers therefore need to be seen as an extension of the network, with cloud interconnection architectures able to quickly adjust both bandwidth and access, usage-based (platform) models and software-defined access in co- located data centers, and not rely on the need to install new physical connections. In the case of a Latin American holding company, software-defined access to cloud hyperscalers rapidly enabled increased connectivity to Microsoft Azure for users working from home.
- Application/user awareness built in
Cloud vendors typically refresh their hardware every two years, compared with four to seven years at most large enterprises, leading to a 20% performance improvement.
2. Cloud-ready applications
In this building block, again, the first must-do is creating an inventory of which applications you already have, and which are already mission-ready to be moved to the cloud. And this is an inventory check that should actually be repeated frequently. A BCG report states, “Cloud vendors typically refresh their hardware every two years, compared with every four to seven years at most large enterprises, leading to a 20% performance improvement. This improvement comes via the benefits of Moore’s Law, as well as from faster data retrieval and improved operating systems and virtualization software.”2 This is not so much of an issue for smaller organizations, who tend to have deployed more SaaS applications and so have an easier “transition to cloud” mission.
Once you know what you’ve got, it’s about cloud-enabling those mission-ready applications (and moving the workflows) and then building a security-first, cloud-first strategy for future application enablement. Again, as BCG finds, “Large enterprises that move to the cloud effectively can improve delivery of IT services by 25% to 50%."2
You then need to define your collaboration strategy and choose which platform you want to use for internal communication—as well as potentially communicating with partners, vendors and customers too—again with an eye on which policies and procedures will need to be in place to address any industry-specific compliance requirements. Working out what type of collaboration you want will enable you to choose a unified communications solution that will work for your specific needs. Some organizations will need more video than others; for others, a chat function is key.
Software-based perimeter protection solutions are one key solution for consideration when enabling cloud applications. These can protect evolving infrastructure and the application threat surface. These solutions are beginning to gain traction (refer to figure below).
It’s then all about monitoring those assets across cloud networks and cloud compute, especially with hybrid deployments. This might require controlling user and developer access to cloud apps, as well as monitoring any changes and updates. You then need to ensure that you can continually assess cloud security in line with industry best practice—for example the ability to monitor cloud security through log integration, penetration testing and vulnerability management. Finally, you also need to consider cloud access (including private network access to major cloud assets) and backup and capabilities to offer simple visibility and control across multiple cloud nodes.
When it comes to strong and secure mobile connectivity, the quality of the physical connection cannot be overlooked, so organizations should consider multiple options to get the right performance.
3. Strong and secure mobile connectivity
The must-dos of this third building block are all about delivering reliable and secure connectivity. The quality of the physical connection cannot be overlooked here, so organizations should consider multiple options to get the right performance. As an example, a Verizon customer (global financial services company) recently deployed LTE hotspots to augment areas where consumer broadband connections didn’t perform as expected.
VPNs are the next key item and a foundational piece of this, which can then be augmented with private wireless gateway services. But security must not be an afterthought and, rather, has to be built in as standard. And this is not just about anti- virus protection but rather zero-day protection, leveraging, if possible, AI technologies and threat intelligence to predict potential data breaches. As an example, a U.S.-based investment and insurance Fortune 500 company has deployed an integrated threat prevention solution that combines the power of AI to block malware infections with high effectiveness and little system impact, helping prevent zero-day attacks. The solution works where most attacks occur (at the endpoint) for better efficacy, fast resolution and less disruption.
It goes without saying that any wired and wireless connectivity devices need to be able to be deployed rapidly but also with control, and you also need to be able to monitor the deployment of any unauthorized mobile applications. For example, a U.S.-based global investment firm recently deployed 7,000+ endpoints to remote workers to manage proper traffic prioritization as well as sufficient visibility tools to address regulatory compliance requirements. And finally, you need to carefully manage internet access and policies across public websites through regional internet breakout services.
Dynamic and adaptable remote access, mobile device management and endpoint security are just some of the solutions that you should be considering to make strong and secure mobile connectivity a reality. These need to be augmented with Domain Name System (DNS) protection services for network-based blocking of malware, Internet gateway and breakout solutions for policy control and overall performance improvement, and DDoS protection to enable your growing remote workforce to continue to securely reach corporate assets.
You also need to remember that people are often an organization’s weakest link, so ensure that you continue to train your employees on simple steps to take to counter phishing, malware or other social engineering scams, and track the latest and greatest with threat and intelligence feeds that work directly with email filtering.
The Verizon Data Breach Investigations Report (DBIR) is always a great source of information on security and threat trends. And remember, the best way to keep your network secure is to do the basics well, so make sure you keep up to speed with patch and update programs.
Remember that people are often an organization’s weakest link, so ensure that you continue to train your employees on simple steps to take to counter phishing, malware or other social engineering scams.
4. End-to-end monitoring of performance
The first must-do for this fourth block is the discovery process that we’ve already described above. This is important if you are to be able to properly understand your network data flows. The second is to define how your network is working with respect to visibility, insight and execution. We refer to BCG’s interview with industry experts to cover the nine prevalent core use cases across these three areas (refer to figure below).
The overall objective is to curate an accurate data set so you can understand how you might re-instrument different parts of the network, incorporating predictive analysis to help predict and prevent anomalies or outages, and automating actions, whether for users, transactions or applications.
With the shift in traditional corporate data flows away from the office setting and into remote scenarios, CIOs must be able to glean data directly from the network itself in order to quickly identify and address anomalous behavior. For example, Verizon saw a massive increase (greater than 200%) in UDP (user datagram protocol) versus more normal TCP (transmission control protocol) traffic for one large enterprise. Upon further inspection at the port level, the traffic shift was almost exclusively remote workers using VPNs. In this case, it was an expected shift, but a heightened level of visibility is needed to validate the expectation.
And the holy grail is to leverage AI technologies across IT operations (AIOps solutions) to turn big data into meaningful insights that makes network management and orchestration simple and effective. AIOps solutions is a growing market according to BCG research.
The holy grail of performance monitoring is to leverage AI technologies across IT operations (AIOps solutions) to turn big data into meaningful insights that make network management and orchestration simple and effective.
Monitoring is important for the IT side of the organization; for the business side, it is all about staff morale and productivity, and for this you need visibility. Ultimately, the human experience of working with critical business applications is the most important element. The goal is to be able to understand what is causing productivity loss and take action, even when you don’t own the internet/network your applications are being delivered across and have limited ability to instrument the “as a Service” platforms from which your applications are being delivered. Your digital supply chain is a mixture of public, private, personal, corporate and other assets, and it’s vital to understand how the user experience (both internal and external) is being delivered, end to end. BCG has worked with the U.S. National Institute of Standards and Technology (NIST) to develop both Key Practices for Digital Supply Chains and an open source Supply Chain Interdependency software tool that can help you manage the digital supply chain.
Visibility profiles are packaged to represent best-practice measurements, thresholds, timing and alerts to notify when a service is up, down or degraded and the associated impact. Analyzing your business’ digital ecosystem within the context of experience/digital scoring (similar to telephony quality Mean Opinion Score [MOS] scoring) and business outcome delivery complemented with the lens of human experience provides better correlation and context to drive faster average repair time and automation through AIOps.
Additionally, some of the solutions that could be considered include synthetic transaction, role-based end-user experience or business process monitoring—and of course, intelligent workflow automation.
One example on visibility is a tool that Verizon has started to deploy internally and that is also available for external consumption. Verizon technology partner Actual Experience offers a tool that synthetically measures your global digital ecosystem, then uses advanced “human experience” algorithms and correlation techniques to rapidly determine whether the cause of lost productivity is home Wi-Fi, the local ISP, corporate infrastructure or a cloud service provider. The graphic above shows some of the actionable information gleaned from this type of instrumentation for a select division of the Verizon Business Group. As organizations conduct and scale up similar visibility tests, the availability and transparency of the productivity metric will then guide the set of focus areas within the supply chain to drive best use of resources. Again, the answer set may differ by divisions and/or applications, but that is the beauty of data granularity, where slicing and dicing de-averages the answer to drive specificity.
5. Zero-trust security
The U.S. National Institute of Standards and Technology (NIST) states that “Zero trust [security focuses] on protecting resources, not network segments, as the network location is no longer seen as the prime component to the security posture of the resource.” The first must-do in this building block is to identify those business-critical applications and data feeds that are your organization’s most critical assets (“crown jewels”)—without which you could not operate. These could be virtual or data assets or physical assets such as manufacturing or energy production and transmission equipment.
Verizon has visibility into COVID-19-related security impact through its vast, global voice and data networks. These range from social engineering scams in the voice network to changing communications trends and associated threats and vulnerabilities resulting from the rapid and massive transition from working in an office setting to the home. BCG research has shown that nation state actors are targeting family members of corporate executives, knowing that the unwitting and untrained family member and corporate executive are on the same insecure home network. After compromising the family member’s computer, the nation state adversary attempts to pivot into the executive’s laptop and infiltrate the corporate systems.
The purpose of social engineering scams is usually financial gain (to obtain credit card or bank account information) or to collect valuable identity credentials to commit fraud or compromise the security of critical corporate systems. Using our own intelligence collection tools throughout our network, Verizon captures hundreds to thousands of spam and bot calls per day and then identifies and logs them. This work has identified plots whereby hackers are offering free COVID-19 testing kits, stimulus funds assistance or other related services. Verizon works with federal agencies and partners to identify the sources of these scams to mitigate and eliminate them as they emerge and change.
Similar attacks could breach corporate security perimeters that now extend into the home, and this threat vector is actually growing as hackers move their focus toward smart devices, which are usually viewed by users as trusted. In the future, CIOs and CISOs must incorporate an evolved protection strategy with appropriate tooling or partnerships to account for and mitigate such attacks.
So, firstly, you should look to build ultra-, cross- or micro- segmented networks using the very latest firewall technology, allowing access to applications rather than just to networks themselves. These should follow the NIST 800-207 zero-trust strategy, i.e., servers should be isolated and end-to-end encryption and multi-factor authentication be deployed to defeat credential theft. And you need to ensure that you are looking at applications and data in both corporate data centers and the cloud.
You then need to consider upgrading the security of application and authentication data flows by separating the path of data from the path of its control, and think about enabling privileged access to applications as well. Jump hosts can support this by establishing a separate security zone to access and manage devices. And then you must also consider the true sharing and application access requirements of partners and develop models to enforce the chain of trust across assets such as supply chain, IoT devices and transactions of record. And then, perhaps most importantly, you need to build a comprehensive breach response program, implementing dynamic and up-to- date threat detection and analysis capabilities. BCG has worked closely with Massachusetts Institute of Technology (MIT) and the World Economic Forum to develop unique tabletop exercises that are able to find and fix weaknesses in breach response and business continuity programs.
Perhaps key to zero-trust security is the deployment of software-defined perimeter (SDP) protection solutions. These effectively put a “virtual wall” around your applications and devices to help better protect against potential threats. They also improve the user experience by enabling a shorter path to application, whether SaaS applications or corporate applications in hosted cloud environments, and offer both internet breakout and direct cloud access. Secure infrastructure as a service (IaaS) enclaves should also be considered, especially when you consider IaaS’ rapid growth.
We have seen multiple organizations across different industries deploying SDP solutions. Here are just three examples:
Complementary solutions to zero-trust security include the deployment of network detection and response solutions for advanced inspection services, and managed detection and response solutions to help reduce the time between breach and resolution. And don’t forget to test the solutions in tabletop exercises to determine response readiness.
6. A resilient end-user support model
A resilient end-user support model might not be something many organizations consider as a technical building block, but it’s vitally important when it comes to the transformation agenda. No longer will users necessarily have tech support onsite or similar organizational services. The must-dos here include validating that employees, partners and customers have access to the right technologies in an appropriate way across a business. New technology rollout strategies must be created to reduce adoption friction. And then making sure that you have effective communication in place to support your end users, whether for new technology rollouts or simply with day-to-day business processes. For example, Verizon moved 700 employees from stores to inside sales and customer service representatives. Thus having protocols, trainings and tools available was critical, and such should be baked into part of the overall enterprise planning.
Best practice here includes specialist on-demand chat capabilities, how-to white papers and videos, regular communication updates, and digital customer self-service (to focus support team time on the most important issues and empowering users to solve more minor issues themselves). And ongoing communication also needs to be considered, whether via a web page, video collaboration, audio calls, streamed guide sessions or even a good old-fashioned newsletter. Most important, though, is to build a flexible support system that can adapt as user needs change.
Lastly, organizations must consider the critical technology in their digital supply chain and build business continuity plans around them. If an entire business model is based on leveraging tablets and all manufacturing is shut down due to a natural disaster, pandemic, etc., what happens to the business? Tabletop exercises, war gaming scenarios and identifying gaps are critical activities every organization should undertake.
Tabletop exercises, war gaming scenarios and identifying gaps are critical activities every organization should undertake.
1Market Trends: SD-WAN and NFV for Enterprise Network Services,” January 30, 2020