Using vendor-neutral platforms to host, connect, and secure edge applications and workloads is the best practice for ensuring the scalability and flexibility of financial edge architectures. Moving away from dedicated device stacks and taking a “platformization” approach allows financial institutions to easily deploy, update, and swap out applications and capabilities on demand. Vendor-neutral platforms help reduce hardware overhead costs to deploy new branches and allow banks to explore different edge software capabilities without costly hardware upgrades.

Additionally, using a centralized, cloud-based edge management and orchestration (EMO) platform is the best practice for ensuring remote teams have holistic oversight of the distributed edge computing architecture. This platform should be vendor-agnostic to ensure complete coverage over mixed and legacy architectures, and it should use out-of-band (OOB) management to provide continuous remote access to edge infrastructure even during a major service outage.

How Nodegrid streamlines edge computing for the banking industry

Nodegrid is a vendor-neutral edge networking platform that consolidates an entire edge tech stack into a single, cost-effective device. Nodegrid has a Linux-based OS that supports third-party VMs and Docker containers, allowing banks to run edge computing workloads, data analytics software, automation, security, and more. 

The Nodegrid Gate SR is available with an Nvidia Jetson Nano card that’s optimized for artificial intelligence workloads. This allows banks to run AI surveillance software, ML-powered recommendation engines, and AIOps at the edge alongside networking and infrastructure workloads rather than purchasing expensive, dedicated GPU resources. Plus, Nodegrid’s Gen 3 OOB management ensures continuous remote access and IMI for improved branch resilience.

Get Nodegrid for your edge computing use cases in banking

Nodegrid’s flexible, vendor-neutral platform adapts to any use case and deployment environment. Watch a demo to see Nodegrid’s financial network solutions in action.

Watch a demo

AI data center infrastructure components

Computing

Generative AI and other artificial intelligence technologies require significant processing power. AI workloads typically run on graphics processing units (GPUs), which are made up of many smaller cores that perform simple, repetitive computing tasks in parallel. GPUs can be clustered together to process data for AI much faster than CPUs.

Storage

AI requires vast amounts of data for training and inference. On-premises AI data centers typically use object storage systems with solid-state disks (SSDs) composed of multiple sections of flash memory (a.k.a., flash storage). Storage solutions for AI workloads must be modular so additional capacity can be added as data needs grow, through either physical or logical (networking) connections between devices.

Networking

AI workloads are often distributed across multiple computing and storage nodes within the same data center. To prevent packet loss or delays from affecting the accuracy or performance of AI models, nodes must be connected with high-speed, low-latency networking. Additionally, high-throughput WAN connections are needed to accommodate all the data flowing in from end-users, business sites, cloud apps, IoT devices, and other sources across the enterprise.

Power

AI infrastructure uses significantly more power than traditional data center infrastructure, with a rack of three or four AI servers consuming as much energy as 30 to 40 standard servers. To prevent issues, these power demands must be accounted for in the layout design for new AI data center deployments and, if necessary, discussed with the colocation provider to ensure enough power is available.

Management

Data center infrastructure, especially at the scale required for AI, is typically managed with a jump box, terminal server, or serial console that allows admins to control multiple devices at once. The best practice is to use an out-of-band (OOB) management device that separates the control plane from the data plane using alternative network interfaces. An OOB console server provides several important functions:

1. It provides an alternative path to data center infrastructure that isn’t reliant on the production ISP, WAN, or LAN, ensuring remote administrators have continuous access to troubleshoot and recover systems faster, without an on-site visit.
2. It isolates management interfaces from the production network, preventing malware or compromised accounts from jumping over from an infected system and hijacking critical data center infrastructure.
3. It helps create an isolated recovery environment where teams can clean and rebuild systems during a ransomware attack or other breach without risking reinfection.

An OOB serial console helps minimize disruptions to AI infrastructure. For example, teams can use OOB to remotely control PDU outlets to power cycle a hung server. Or, if a networking device failure brings down the LAN, teams can use a 5G cellular OOB connection to troubleshoot and fix the problem. Out-of-band management reduces the need for costly, time-consuming site visits, which significantly improves the resilience of AI infrastructure.

AI data center challenges

Artificial intelligence workloads, and the data center infrastructure needed to support them, are highly complex. Many IT teams struggle to efficiently provision, maintain, and repair AI data center infrastructure at the scale and speed required, especially when workflows are fragmented across legacy and multi-vendor solutions that may not integrate. The best way to ensure data center teams can keep up with the demands of artificial intelligence is with a unified AI orchestration platform. Such a platform should include:

• Automation for repetitive provisioning and troubleshooting tasks
• Unification of all AI-related workflows with a single, vendor-neutral platform
• Resilience with cellular failover and Gen 3 out-of-band management.

To learn more, read AI Orchestration: Solving Challenges to Improve AI Value

Improving operational efficiency with a vendor-neutral platform

Nodegrid is a Gen 3 out-of-band management solution that provides the perfect unification platform for AI data center orchestration. The vendor-neutral Nodegrid platform can integrate with or directly run third-party software, unifying all your networking, management, automation, security, and recovery workflows. A single, 1RU Nodegrid Serial Console Plus (NSCP) can manage up to 96 data center devices, and even extend automation to legacy and mixed-vendor solutions that wouldn’t otherwise support it. Nodegrid Serial Consoles enable the fast and cost-efficient infrastructure scaling required to support GenAI and other artificial intelligence technologies.

Make Nodegrid your AI data center orchestration platform

Request a demo to learn how Nodegrid can improve the efficiency and resilience of your AI data center infrastructure.
 Contact Us

The goal of AI orchestration is to provide a single, unified platform for teams to oversee and manage AI-related workflows across the entire organization. This post describes the ideal AI orchestration solution and the technologies that make it work, helping companies use artificial intelligence more efficiently.

AI challenges to overcome

The challenges an organization must overcome to use AI more cost-effectively and see faster returns can be broken down into three categories:

1. Overseeing AI-led workflows to ensure models are behaving as expected and providing accurate results, when these workflows are spread across the enterprise in different geographic locations and vendor-specific applications.
   .
2. Efficiently provisioning, maintaining, and scaling the vast infrastructure and computational resources required to run intensive AI workflows at remote data centers and edge computing sites.
   .
3. Maintaining 24/7 availability and performance of remote AI workflows and infrastructure during security breaches, equipment failures, network outages, and natural disasters.

These challenges have a few common causes. One is that artificial intelligence and the underlying infrastructure that supports it are highly complex, making it difficult for human engineers to keep up. Two is that many IT environments are highly fragmented due to closed vendor solutions that integrate poorly and require administrators to manage too many disparate systems, allowing coverage gaps to form. Three is that many AI-related workloads occur off-site at data centers and edge computing sites, so it’s harder for IT teams to repair and recover AI systems that go down due to a networking outage, equipment failure, or other disruptive event.

How AI orchestration streamlines AI/ML in an enterprise environment

The ideal AI orchestration platform solves these problems by automating repetitive and data-heavy tasks, unifying workflows with a vendor-neutral platform, and using out-of-band (OOB) serial console management to provide continuous remote access even during major outages.

Automation

Automation is crucial for teams to keep up with the pace and scale of artificial intelligence. Organizations use automation to provision and install AI data center infrastructure, manage storage for AI training and inference data, monitor inputs and outputs for toxicity, perform root-cause analyses when systems fail, and much more. However, tracking and troubleshooting so many automated workflows can get very complicated, creating more work for administrators rather than making them more productive. An AI orchestration platform should provide a centralized interface for teams to deploy and oversee automated workflows across applications, infrastructure, and business sites.

Unification

The best way to improve AI operational efficiency is to integrate all of the complicated monitoring, management, automation, security, and remediation workflows. This can be accomplished by choosing solutions and vendors that interoperate or, even better, are completely vendor-agnostic (a.k.a., vendor-neutral). For example, using open, common platforms to run AI workloads, manage AI infrastructure, and host AI-related security software can help bring everything together where administrators have easy access. An AI orchestration platform should be vendor-neutral to facilitate workload unification and streamline integrations.

Resilience

AI models, workloads, and infrastructure are highly complex and interconnected, so an issue with one component could compromise interdependencies in ways that are difficult to predict and troubleshoot. AI systems are also attractive targets for cybercriminals due to their vast, valuable data sets and because of how difficult they are to secure, with HiddenLayer’s 2024 AI Threat Landscape Report finding that 77% of businesses have experienced AI-related breaches in the last year. An AI orchestration platform should help improve resilience, or the ability to continue operating during adverse events like tech failures, breaches, and natural disasters.

Gen 3 out-of-band management technology is a crucial component of AI and network resilience. A vendor-neutral OOB solution like the Nodegrid Serial Console Plus (NSCP) uses alternative network connections to provide continuous management access to remote data center, branch, and edge infrastructure even when the ISP, WAN, or LAN connection goes down. This gives administrators a lifeline to troubleshoot and recover AI infrastructure without costly and time-consuming site visits. The NSCP allows teams to remotely monitor power consumption and cooling for AI infrastructure. It also provides 5G/4G LTE cellular failover so organizations can continue delivering critical services while the production network is repaired.

Gen 3 OOB also helps organizations implement isolated management infrastructure (IMI), a.k.a, control plane/data plane separation. This is a cybersecurity best practice recommended by the CISA as well as regulations like PCI DSS 4.0, DORA, NIS2, and the CER Directive. IMI prevents malicious actors from being able to laterally move from a compromised production system to the management interfaces used to control AI systems and other infrastructure. It also provides a safe recovery environment where teams can rebuild and restore systems during a ransomware attack or other breach without risking reinfection.

Getting the most out of your AI investment

An AI orchestration platform should streamline workflows with automation, provide a unified platform to oversee and control AI-related applications and systems for maximum efficiency and coverage, and use Gen 3 OOB to improve resilience and minimize disruptions. Reducing management complexity, risk, and repair costs can help companies see greater productivity and financial returns from their AI investments.

The vendor-neutral Nodegrid platform from ZPE Systems provides highly scalable Gen 3 OOB management for up to 96 devices with a single, 1RU serial console. The open Nodegrid OS also supports VMs and Docker containers for third-party applications, so you can run AI, automation, security, and management workflows all from the same device for ultimate operational efficiency.

Streamline AI orchestration with Nodegrid

Contact ZPE Systems today to learn more about using a Nodegrid serial console as the foundation for your AI orchestration platform. Contact Us

This blog describes four edge computing use cases in telecom before describing the benefits and best practices for edge computing in the telecommunications industry.

4 Edge computing use cases in telecom

1. Enhancing the customer experience with real-time analytics

Each customer interaction, from sales calls to repair requests and service complaints, is a chance to collect and leverage data to improve the experience in the future. Transferring that data from customer sites, regional branches, and customer service centers to a centralized data analysis application takes time, creates network latency, and can make it more difficult to get localized and context-specific insights. Edge computing allows telecom companies to analyze valuable customer experience data, such as network speed, uptime (or downtime) count, and number of support contacts in real-time, providing better opportunities to identify and correct issues before they go on to affect future interactions.

2. Streamlining remote infrastructure management and recovery with AIOps

AIOps helps telecom companies manage complex, distributed network infrastructure more efficiently. AIOps (artificial intelligence for IT operations) uses advanced machine learning algorithms to analyze infrastructure monitoring data and provide maintenance recommendations, automated incident management, and simple issue remediation. Deploying AIOps on edge computing devices at each telecom site enables real-time analysis, detection, and response, helping to reduce the duration of service disruptions. For example, AIOps can perform automated root-cause analysis (RCA) to help identify the source of a regional outage before technicians arrive on-site, allowing them to dive right into the repair. Edge AIOps solutions can also continue functioning even if the site is cut off from the WAN or Internet, potentially self-healing downed networks without the need to deploy repair techs on-site.

3. Preventing environmental conditions from damaging remote equipment

Telecommunications equipment is often deployed in less-than-ideal operating conditions, such as unventilated closets and remote cell site shelters. Heat, humidity, and air particulates can shorten the lifespan of critical equipment or cause expensive service failures, which is why it’s recommended to use environmental monitoring sensors to detect and alert remote technicians to problems. Edge computing applications can analyze environmental monitoring data in real-time and send alerts to nearby personnel much faster than cloud- or data center-based solutions, ensuring major fluctuations are corrected before they damage critical equipment.

4. Improving operational efficiency with network virtualization and consolidation

Another way to reduce management complexity – as well as overhead and operating expenses – is through virtualization and consolidation. Network functions virtualization (NFV) virtualizes networking equipment like load balancers, firewalls, routers, and WAN gateways, turning them into software that can be deployed anywhere – including edge computing devices. This significantly reduces the physical tech stack at each site, consolidating once-complicated network infrastructure into, in some cases, a single device. For example, the Nodegrid Gate SR provides a vendor-neutral edge computing platform that supports third-party NFVs while also including critical edge networking functionality like out-of-band (OOB) serial console management and 5G/4G cellular failover.

Edge computing in telecom: Benefits and best practices

Edge computing can help telecommunications companies:

• Get actionable insights that can be leveraged in real-time to improve network performance, service reliability, and the support experience.
• Reduce network latency by processing more data at each site instead of transmitting it to the cloud or data center for analysis.
• Lower CAPEX and OPEX at each site by consolidating the tech stack and automating management workflows with AIOps.
• Prevent downtime with real-time analysis of environmental and equipment monitoring data to catch problems before they escalate.
• Accelerate recovery with real-time, AIOps root-cause analysis and simple incident remediation that continues functioning even if the site is cut off from the WAN or Internet.

Management infrastructure isolation, which is recommended by CISA and required by regulations like DORA, is the best practice for improving edge resilience and ensuring a speedy recovery from failures and breaches. Isolated management infrastructure (IMI) prevents compromised accounts, ransomware, and other threats from moving laterally from production resources to the interfaces used to control critical network infrastructure.

Vendor-neutral platforms help reduce hardware overhead costs to deploy new edge sites, make it easy to spin-up new NFVs to meet increased demand, and allow telecom organizations to explore different edge software capabilities without costly hardware upgrades. For example, the Nodegrid Gate SR is available with an Nvidia Jetson Nano card that’s optimized for AI workloads, so companies can run innovative artificial intelligence at the edge alongside networking and infrastructure management workloads rather than purchasing expensive, dedicated GPU resources.

Streamlined, cost-effective edge computing with Nodegrid

Nodegrid’s flexible, vendor-neutral platform adapts to all edge computing use cases in telecom. Watch a demo to see Nodegrid’s telecom solutions in action.

Watch a demo

Edge computing mitigates many of these challenges by decentralizing cloud and data center resources and distributing them at the network’s “edges,” where most retail operations take place. Running applications and processing data at the edge enables real-time analysis and insights and ensures that systems remain operational even if Internet access is disrupted by an ISP outage or natural disaster. This blog describes five potential edge computing use cases in retail and provides more information about the benefits of edge computing for the retail industry.

5 Edge computing use cases in retail

.

1. Security video analysis

Security cameras are crucial to loss prevention, but constantly monitoring video surveillance feeds is tedious and difficult for even the most experienced personnel. AI-powered video surveillance systems use machine learning to analyze video feeds and detect suspicious activity with greater vigilance and accuracy. Edge computing enhances AI surveillance by allowing solutions to analyze video feeds in real-time, potentially catching shoplifters in the act and preventing inventory shrinkage.

2. Localized, real-time insights

Retailers have a brief window to meet a customer’s needs before they get frustrated and look elsewhere, especially in a brick-and-mortar store. A retail store can use an edge computing application to learn about customer behavior and purchasing activity in real-time. For example, they can use this information to rotate the products featured on aisle endcaps to meet changing demand, or staff additional personnel in high-traffic departments at certain times of day. Stores can also place QR codes on shelves that customers scan if a product is out of stock, immediately alerting a nearby representative to provide assistance.

3. Enhanced inventory management

Effective inventory management is challenging even for the most experienced retail managers, but ordering too much or too little product can significantly affect sales. Edge computing applications can improve inventory efficiency by making ordering recommendations based on observed purchasing patterns combined with real-time stocking updates as products are purchased or returned. Retailers can use this information to reduce carrying costs for unsold merchandise while preventing out-of-stocks, improving overall profit margins.
.

4. Building management

Using IoT devices to monitor and control building functions such as HVAC, lighting, doors, power, and security can help retail organizations reduce the need for on-site facilities personnel, and make more efficient use of their time. Data analysis software helps automatically optimize these systems for efficiency while ensuring a comfortable customer experience. Running this software at the edge allows automated processes to respond to changing conditions in real-time, for example, lowering the A/C temperature or routing more power to refrigerated cases during a heatwave.

5. Warehouse automation

The retail industry uses warehouse automation systems to improve the speed and efficiency at which goods are delivered to stores or directly to users. These systems include automated storage and retrieval systems, robotic pickers and transporters, and automated sortation systems. Companies can use edge computing applications to monitor, control, and maintain warehouse automation systems with minimal latency. These applications also remain operational even if the site loses internet access, improving resilience.

The benefits of edge computing for retail

The benefits of edge computing in a retail setting include:
.

For example, the Nodegrid Gate SR integrated branch services router delivers an entire stack of edge networking, infrastructure management, and computing technologies in a single, streamlined device. The open, Linux-based Nodegrid OS supports VMs and Docker containers for third-party edge computing applications, security solutions, and more. The Gate SR is also available with an Nvidia Jetson Nano card that’s optimized for AI workloads to help retail organizations reduce the hardware overhead costs of deploying artificial intelligence at the edge.

Consolidated edge computing with Nodegrid

Nodegrid’s flexible, scalable platform adapts to all edge computing use cases in retail. Watch a demo to see Nodegrid’s retail network solutions in action.

Watch a demo

The world as we know it is connected to IT, and IT relies on its underlying infrastructure. Organizations must prioritize maintaining this infrastructure; otherwise, any disruption or breach has a ripple effect that takes services offline for millions of users (take the recent CrowdStrike outage, for example). A big part of this maintenance is ensuring that all hardware components, including console servers, are up-to-date and secure. Most console servers reach end-of-life (EOL) and need to be replaced, but for many reasons, whether budgetary concerns or the “if it isn’t broken” mentality, IT teams often keep their EOL devices. Let’s look at the risks of using EOL console servers, and why replacing them goes hand-in-hand with securing IT.

The Risks of Using End-of-Life Console Servers

End-of-life console servers can undermine the security and functionality of IT systems. These risks include:

1. Lack of Security Features and Updates

Aging console servers lack adequate hardware and management security features, meaning they can’t support a zero trust approach. On top of this, once a console server reaches EOL, the manufacturer stops providing security patches and updates. The device then becomes vulnerable to newly discovered CVEs and complex cyberattacks (like the MOVEit and Ragnar Locker breaches). Cybercriminals often target outdated hardware because they know that these devices are no longer receiving updates, making them easy entry points for launching attacks.

2. Compliance Issues

Many industries have stringent regulatory requirements regarding data security and IT infrastructure. DORA, NIS2 (EU), NIST2 (US), PCI 4.0 (finance), and CER Directive are just a few of the updated regulations that are cracking down on how organizations architect IT, including the management layer. Using EOL hardware can lead to non-compliance, resulting in fines and legal repercussions. Regulatory bodies expect organizations to use up-to-date and secure equipment to protect sensitive information.

3. Prolonged Recovery

EOL console servers are prone to failures and inefficiencies. As these devices age, their performance deteriorates, leading to increased downtime and disruptions. Most console servers are Gen 2, meaning they offer basic remote troubleshooting (to address break/fix scenarios) and limited automation capabilities. When there is a severe disruption, such as a ransomware attack, hackers can easily access and encrypt these devices to lock out admin access. Organizations then must endure prolonged recovery (just look the still ongoing CrowdStrike outage, or last year’s MGM attack) because they need to physically decommission and restore their infrastructure.

The Importance of Replacing EOL Console Servers

Here’s why replacing EOL console servers is essential to securing IT:

1. Modern Security Approach

Zero trust is an approach that uses segmentation across IT assets. This ensures that only authorized users can access resources necessary for their job function. This approach requires SAML, SSO, MFA/2FA, and role-based access controls, which are only supported by modern console servers. Modern devices additionally feature advanced security through encryption, signed OS, and tampering detection. This ensures a complete cyber and physical approach to security.

2. Protection Against New Threats

New CVEs and evolving threats can easily take advantage of EOL devices that no longer receive updates. Modern console servers benefit from ongoing support in the form of firmware upgrades and security patches. Upgrading with a security-focused device vendor can drastically shrink the attack surface, by addressing supply chain security risks, codebase integrity, and CVE patching.

3. Ease of Compliance

EOL devices lack modern security features, but this isn’t the only reason why they make it difficult or impossible to comply with regulations. They also lack the ability to isolate the control plane from the production network (see Diagram 1 below), meaning attackers can easily move between the two in order to launch ransomware and steal sensitive information. Watchdog agencies and new legislation are stipulating that organizations follow the latest best practice of separating the control plane from production, called Isolated Management Infrastructure (IMI). Modern console servers make this best practice simple to achieve by offering drop-in out-of-band that is completely isolated from production assets (see Diagram 2 below). This means that the organization is always in control of its IT assets and sensitive data.

Diagram 1: Though an acceptable approach, Gen 2 out-of-band lacks isolation and leaves management interfaces vulnerable to the internet.

Diagram 2: Gen 3 out-of-band fully isolates the control plane to guarantee organizations retain control of their IT assets and sensitive info.

4. Faster Recovery

New console servers are designed to handle more workloads and functions, which eliminates single-purpose devices and shrinks the attack surface. They can also run VMs and Docker containers to host applications. This enables what Gartner calls the Isolated Recovery Environment (IRE) (see Diagram 3 below), which is becoming essential for faster recovery from ransomware. Since the IMI component prohibits attackers from accessing the control plane, admins retain control during an attack. They can use the IMI to deploy their IRE and the necessary applications — remotely — to decommission, cleanse, and restore their infected infrastructure. This means that they don’t have to roll trucks week after week when there’s an attack; they just need to log into their management infrastructure to begin assessing and responding immediately, which significantly reduces recovery times.

Diagram 3: The Isolated Recovery Environment allows for a comprehensive and rapid response to ransomware attacks.

Watch How To Secure The Network Backbone

I recently presented at Cisco Live Vegas on how to secure the network’s backbone using Isolated Management Infrastructure. I walk you through the evolution of network management, and it becomes obvious that end-of-life console servers are a major security concern, both from the hardware perspective itself and their lack of isolation capabilities. Watch my 10-minute presentation from the show and download some helpful resources, including the blueprint to building IMI.

Edge computing deploys data analytics applications and computing resources around the edges of the network, where much of the most valuable data is created. This significantly reduces latency and mitigates many security and compliance risks. In a healthcare setting, edge computing enables real-time medical insights and interventions while keeping HIPAA-regulated data within the local security perimeter. This blog describes six potential edge computing use cases in healthcare that take advantage of the speed and security of an edge computing architecture.

6 Edge computing use cases in healthcare

Edge computing use cases for EMS

Mobile emergency medical services (EMS) teams need to make split-second decisions regarding patient health without the benefit of a doctorate and, often, with spotty Internet connections preventing access to online drug interaction guides and other tools. Installing edge computing resources on cellular edge routers gives EMS units real-time health analysis capabilities as well as a reliable connection for research and communications. Potential use cases include:
.

Edge computing use cases in hospitals & clinics

Hospitals and clinics use IoT devices to monitor vitals, dispense medications, perform diagnostic tests, and much more. Sending all this data to the cloud or data center takes time, delaying test results or preventing early intervention in a health crisis, especially in rural locations with slow or spotty Internet access. Deploying applications and computing resources on the same local network enables faster analysis and real-time alerts. Potential use cases include:
.

Edge computing use cases for wearable medical devices

Wearable medical devices give patients and their caregivers greater control over health outcomes. With edge computing, health data analysis software can run directly on the wearable device, providing real-time results even without an Internet connection. Potential use cases include:
.

The benefits of edge computing for healthcare

Using edge computing in a healthcare setting as described in the use cases above can help organizations:

• Improve patient care in remote settings, where a lack of infrastructure limits the ability to use cloud-based technology solutions.
• Process and analyze patient health data faster and more reliably, leading to earlier interventions.
• Increase efficiency by assisting understaffed medical teams with diagnostics, patient monitoring, and communications.
• Mitigate security and compliance risks by keeping health data within the local security perimeter.

Edge computing can also help healthcare organizations lower their operational costs at the edge by reducing bandwidth utilization and cloud data storage expenses. Another way to reduce costs is by using consolidated, vendor-neutral solutions to host, connect, and secure edge applications and workloads.

For example, the Nodegrid Gate SR is an integrated branch services router that delivers an entire stack of edge networking, infrastructure management, and computing technologies in a single, streamlined device. Nodegrid’s open, Linux-based OS supports VMs and Docker containers for third-party edge applications, security solutions, and more. Plus, an onboard Nvidia Jetson Nano card is optimized for AI workloads at the edge, significantly reducing the hardware overhead costs of using artificial intelligence at remote healthcare sites. Nodegrid’s flexible, scalable platform adapts to all edge computing use cases in healthcare, future-proofing your edge architecture.

Streamline your edge deployment with Nodegrid

The vendor-neutral Nodegrid platform consolidates an entire edge technology stack into a unified, streamlined solution. Watch a demo to see Nodegrid’s healthcare network solutions in action.

Watch a demo

On July 19, 2024, CrowdStrike, a leading cybersecurity firm renowned for its advanced endpoint protection and threat intelligence solutions, experienced a significant outage that disrupted operations for many of its clients. This outage, triggered by a software upgrade, resulted in crashes for Windows PCs, creating a wave of operational challenges for banks, airports, enterprises, and organizations worldwide. This blog post explores what transpired during this incident, what caused the outage, and the broader implications for the cybersecurity industry.

What happened?

The incident began on the morning of July 19, 2024, when numerous CrowdStrike customers started reporting issues with their Windows PCs. Users experienced the BSOD (blue screen of death), which is when Windows crashes and renders devices unusable. As the day went on, it became evident that the problem was widespread and directly linked to a recent software upgrade deployed by CrowdStrike.

Timeline of Events

1. Initial Reports: Early in the day, airports, hospitals, and critical infrastructure operators began experiencing unexplained crashes on their Windows PCs. The issue was quickly reported to CrowdStrike’s support team.
2. Incident Acknowledgement: CrowdStrike acknowledged the issue via their social media channels and direct communications with affected clients, confirming that they were investigating the cause of the crashes.
3. Root Cause Analysis: CrowdStrike’s engineering team worked diligently to identify the root cause of the problem. They soon determined that a software upgrade released the previous night was responsible for the crashes.
4. Mitigation Efforts: Upon isolating the faulty software update, CrowdStrike issued guidance on how to roll back the update and provided patches to fix the issue.

What caused the CrowdStrike outage?

The root cause of the outage was a software upgrade intended to enhance the functionality and security of CrowdStrike’s Falcon sensor endpoint protection platform. However, this upgrade contained a bug that conflicted with certain configurations of Windows PCs, leading to system crashes. Several factors contributed to the incident:

1. Insufficient Testing: The software update did not undergo adequate testing across all possible configurations of Windows PCs. This oversight meant that the bug was not detected before the update was deployed to customers.
2. Complex Interdependencies: The incident highlights the complex interdependencies between software components and operating systems. Even minor changes can have unforeseen impacts on system stability.
3. Rapid Deployment: In the cybersecurity industry, quick responses to emerging threats are crucial. However, the pressure to deploy updates rapidly can sometimes lead to insufficient testing and quality assurance processes.

We need to remember one important fact: whether software is written by humans or AI, there will be mistakes in coding and testing. When an issue slips through the cracks, the customer lab is the last resort to catch it. Usually, this can be done with a controlled rollout, where the IT team first upgrades their lab equipment, performs further testing, puts in place a rollback plan, and pushes the update to a less critical site. But in a cloud-connected SaaS world, the customer is no longer in control. That’s why they sign waivers stating that if such an incident occurs, the company that caused the problem is not liable. Experts are saying the only way to address this challenge is to have an infrastructure that’s designed, deployed, and operated for resilience. We discuss this architecture further down in this article.

How to recover from the CrowdStrike outage

CrowdStrike gives two options for recovering:

• Option 1: Reboot in Safe Mode – Reboot the affected device in Safe Mode, locate and delete the file “C-00000291*.sys”, and then restart the device.
• Option 2: Re-image – Download and configure the recovery utility to create a new Windows image, add this image to a USB drive, and then insert this USB drive into the target device. The utility will automatically find and delete the file that’s causing the crash.

The biggest obstacle that is costing organizations a lot of time and money is that with either of these recovery methods, IT staff need to be physically present to work on each affected device. They need to go one by one manually remediating via Safe Mode or physically inserting the USB drive. What makes this more difficult is that many organizations use physical and software/management security controls to limit access. Locked device cabinets slow down physical access to devices, and things like role-based access policies and disk encryption can make Safe Mode unusable. Because this outage is affecting more than 8.5 million computers, this kind of work won’t scale efficiently. That’s why organizations are turning to Isolated Management Infrastructure (IMI) and the Isolated Recovery Environment (IRE).

How IMI and IRE help you recover faster

IMI is a dedicated control plane network that’s meant for administration and recovery of IT systems, including Windows PCs affected by the CrowdStrike outage. It uses the concept of out-of-band management, where you deploy a management device that is connected to dedicated management ports of your IT infrastructure (e.g., serial ports, IPMI ports, and other ethernet management ports). IMI also allows you to deploy recovery services for your digital estate that is immutable and near-line when recovery needs to take place.

IMI does not rely at all on the production assets, as it has its own dedicated remote access via WAN links like 4G/5G, and can contain and encrypt recovery keys and tools with zero trust.

IMI gives teams remote, low-level access to devices so they can recover their systems remotely without the need to visit sites. Organizations that employ IMI are able to revert back to a golden image through automation, or deploy bootable tools to all the computers at the site to rescue them without data loss.

The dedicated out-of-band access to serial/IPMI and management ports gives automation software the same abilities as if a physical crash cart was pulled up to the servers. ZPE Systems’ Nodegrid (now a brand of Legrand) enables this architecture as explained next. Using Nodegrid and ZPE Cloud, teams can use either option to recover from the CrowdStrike outage:

• Option 1: Reboot in Pre-Execution Environment Software – Nodegrid gives low-level network access to connected Windows as if teams were sitting directly in front of the affected device. This means they can remote-in, reboot to a network image, remote into the booted image, delete the faulty file, and restart the system.
• Option 2: Re-image – ZPE Cloud serves as a file repository and orchestration engine. Teams can upload their working Windows image, and then automatically push this across their global fleet of affected devices. This option speeds up recovery times exponentially.
• Option 3 – Run Windows Deployment server on the IMI device at the location and re-image servers and workstations if a good backup of the data has been located. This backup can be made available through the IMI after the initial image has been deployed. The IMI can provide dedicated secure access to the InTune services in your M365 cloud, and the backups do not have to transit the entire internet for all workstations at the time, speeding up recovery many times over.

All of these options can be performed at scale or even automated. Server recovery with large backups, although it may take a couple of hours, can be delivered locally and tracked for performance and consistency.

But what about the risk of making mistakes when you have to repeat these tasks? Won’t this cause more damage and data loss?

Any team can make a mistake repeating these recovery tasks over a large footprint, and cause further damage or loss of data, slowing the recovery further. Automated recovery through the IMI addresses this, and can provide reliable recording and reporting to ensure that the restoration is complete and trusted. 

What does IMI look like?

Here’s a simplified view of Isolated Management Infrastructure. You can see that ZPE’s Nodegrid device is needed, which sits beside production infrastructure and provides the platform for hosting all the tools necessary for fast recovery.

What you need to deploy IMI for recovery:

1. Out-of-band appliance with serial, USB, ethernet interfaces (e.g., ZPE’s Nodegrid Net SR)
2. Switchable PDU: Legrand Server Tech or Raritan PDU
3. Windows PXE Boot image

Here’s the order of operations for a faster CrowdStrike outage recovery:

• Option 1 – Recover

1. 1. IMI deployed with a ZPE Nodegrid device that will start Pre-Execution Environment (PXE) which are Windows boot images that the Nodegrid will push to the computers when they boot up
   2. Send recovery keys from Intune to IMI remote storage over ZPE Cloud’s zero trust platform easily available in cloud or air-gapped through Nodegrid Manager
   3. Enable PXE service (automated across entire enterprise) and define the PXE recovery image
   4. Use serial or IP control of power to the computers, or if possible Intel vPro or IPMI capable machines, to reboot all machines
   5. All machines will boot and check in to a control tower for PXE, or be made available to remote into using stored passwords on the PXE environment, Windows AD, or other Privileged Access Management (PAM)
   6. Delete Files
   7. Reboot

• Option 2 – Lean re-image

1. 1. IMI deployed with a Windows Pre-Execution boot image running PXE service
   2. Enable access to cloud and Azure Intune to the IMI remote storage for the local image for the PC
   3. Enable PXE service (automated across entire enterprise) and define the PXE recovery image
   4. Use serial or IP control of power to the computers, or if possible, Intel vPro or IPMI capable machines, to reboot all machines
   5. Machines will boot and check in to Intune either through the IMI or through normal Internet access and finish imaging
   6. Once the machine completes the InTune tasks, InTune will signal backups to come down to the machines. If these backups are offsite, they can be staged on the IMI through backup software running on a virtual machine located on the IMI appliance to speed up recovery and not impede the Internet connection at the remote site
   7. Pre-stage backups onto local storage, push recovery from the virtual machine on the IMI

• Option 3 – Windows controlled re-image

1. 1. Windows Deployment Server (WDS) installed as a virtual machine running on the IMI appliance (offline to prevent issues or online but under a slowed deployment cycle in case there was an issue) 
   2. Send recovery keys from Intune to IMI remote storage over a zero trust interface in cloud or air-gapped
   3. Use serial or IP control of power to the computers, or if possible, Intel vPro or IPMI capable machines, to reboot all machines
   4. Machines will boot and check in to the WDS for re-imaging
   5. Machines will boot and check in to Intune either through the IMI or through normal Internet access and finish imaging
   6. Once the machine completes the InTune tasks, InTune will signal backups to come down to the machines. If these backups are offsite, they can be staged on the IMI through backup software running on a virtual machine located on the IMI appliance to speed up recovery and not impede the Internet connection at the remote site
   7. Pre-stage backups onto local storage, push recovery from the virtual machine on the IMI

Deploy IMI now to recover and avoid the next outage

Get in touch for help choosing the right size IMI deployment for your organization. Nodegrid and ZPE Cloud are the drop-in solution to recovering from the CrowdStrike outage, with plenty of device options to fit any budget and environment size. Contact ZPE Sales now or download the blueprint to help you begin implementing IMI.

Edge computing delivers data processing and analysis capabilities to the network’s “edge,” at remote sites like branch offices, warehouses, retail stores, and manufacturing plants. It involves deploying computing resources and lightweight applications very near the devices that generate data, reducing the distance and number of network hops between them. In doing so, edge computing reduces latency and bandwidth costs while mitigating risk, enhancing edge resilience, and enabling real-time insights. This blog discusses the five biggest benefits of edge computing, providing examples and additional resources for companies beginning their edge journey.
.

5 benefits of edge computing​

1. Reduces latency

Edge computing leverages data on the same local network as the devices that generate it, cutting down on edge data transmissions over the WAN or Internet. Reducing the number of network hops between devices and applications significantly decreases latency, improving the speed and performance of business intelligence apps, AIOps, equipment health analytics, and other solutions that use edge data.

Some edge applications run on the devices themselves, completely eliminating network hops and facilitating real-time, lag-free analysis. For example, an AI-powered surveillance application installed on an IoT security camera at a walk-up ATM can analyze video feeds in real-time and alert security personnel to suspicious activity as it occurs.​

2. Mitigates risk

Edge computing mitigates security and compliance risks by distributing an organization’s sensitive data and reducing off-site transmission. Large, centralized data stores in the cloud or data center are prime targets for cybercriminals because the sheer volume of data involved increases the chances of finding something valuable. Decentralizing data in much smaller edge storage solutions makes it harder for hackers to find the most sensitive information and also limits how much data they can access at one time.

Keeping data at the edge also reduces the chances of interception in transit to cloud or data center storage. Plus, unlike in the cloud, an organization maintains complete control over who and what has access to sensitive data, aiding in compliance with regulations like the GDPR and PCI DSS 4.0.
.

3. Lowers bandwidth costs

Many organizations use MPLS (multi-protocol label switching) links to securely connect edge sites to the enterprise network. MPLS bandwidth is much more expensive than regular Internet lines, which makes transmitting edge data to centralized data processing applications extremely costly. Plus, it can take months to provision MPLS at a new site, delaying launches and driving up overhead expenses.

Edge computing significantly reduces MPLS bandwidth utilization by running data-hungry applications on the local network, reserving the WAN for other essential traffic. Combining edge computing with SD-WAN (software-defined wide area networking) and SASE (secure access service edge) technologies can markedly decrease the reliance on MPLS links, allowing organizations to accelerate branch openings and see faster edge ROIs.
.

4. Enhances edge resilience

Since edge computing applications run on the same LAN as the devices generating data, they can continue to function even if the site loses Internet access due to an ISP outage, natural disaster, or other adverse event. This also allows uninterrupted edge operations in locations with inconsistent (or no) Internet coverage, like offshore oil rigs, agricultural sites, and health clinics in isolated rural communities. Edge computing ensures that organizations don’t miss any vital health or safety alerts and facilitates technological innovation using AI and other data analytics tools in challenging environments..
.

5. Enables real-time insights

Sending data from the edge to a cloud or on-premises data lake for processing, transformation, and ingestion by analytics or AI/ML tools takes time, preventing companies from acting on insights at the moment when they’re most useful. Edge computing applications start using data as soon as it’s generated, so organizations can extract value from it right away. For example, a retail store can use edge computing to gain actionable insights on purchasing activity and customer behavior in real-time, so they can move in-demand products to aisle endcaps or staff extra cashiers as needed.
.

Simplify your edge computing deployment with Nodegrid

The best way to achieve the benefits of edge computing described above without increasing management complexity or hardware overhead is to use consolidated, vendor-neutral solutions to host, connect, and secure edge workloads. For example, the Nodegrid Gate SR from ZPE Systems delivers an entire stack of edge networking and infrastructure management technologies in a single, streamlined device. The open, Linux-based Nodegrid OS supports VMs and containers for third-party applications, with an Nvidia Jetson Nano card capable of running AI workloads alongside non-AI data analytics for ultimate efficiency.

Improve your edge computing deployment with Nodegrid

Nodegrid consolidates edge computing deployments to improve operational efficiency without sacrificing performance or functionality. Schedule a free demo to see Nodegrid in action.

Schedule a Demo

The current cyber threat landscape is daunting, with attacks occurring so frequently that security experts recommend operating under the assumption that your network is already breached. Major cyber attacks – and the disruptions they cause – frequently make news headlines. For example, the recent LendingTree breach exposed consumer data, which could affect the company’s reputation and compliance status. An attack on auto dealership software company CDK Global took down the platform and disrupted business for approximately 15,000 car sellers – an outage that’s still ongoing as of this article’s writing.

The zero trust security methodology outlines the best practices for limiting the blast radius of a successful breach by preventing malicious actors from moving laterally through the network and accessing the most valuable or sensitive resources. Many organizations have already begun their zero trust journey by implementing role-based access controls (RBAC), multi-factor authentication (MFA), and other security solutions, but still struggle with coverage gaps that result in ransomware attacks and other disruptive breaches. This blog provides advice for improving your zero trust security posture with a multi-layered strategy that mitigates weaknesses for complete coverage.

How to improve your zero trust security posture

.

Gain a full understanding of your protect surface

Many security strategies focus on defending the network’s “attack surface,” or all the potential vulnerabilities an attacker could exploit to breach the network. However, zero trust is all about defending the “protect surface,” or all the data, assets, applications, and services that an attacker could potentially try to access. The key difference is that zero trust doesn’t ask you to try to cover any possible weakness in a network, which is essentially impossible. Instead, it wants you to look at the resources themselves to determine what has the most value to an attacker, and then implement security controls that are tailored accordingly.

Gaining a full understanding of all the resources on your network can be extraordinarily challenging, especially with the proliferation of SaaS apps, mobile devices, and remote workforces. There are automated tools that can help IT teams discover all the data, apps, and devices on the network. Application discovery and dependency mapping (ADDM) tools help identify all on-premises software and third-party dependencies; cloud application discovery tools do the same for cloud-hosted apps by monitoring network traffic to cloud domains. Sensitive data discovery tools scan all known on-premises or cloud-based resources for personally identifiable information (PII) and other confidential data, and there are various device management solutions to detect network-connected hardware, including IoT devices.
,

Micro-segment your network with micro-perimeters

Micro-segmentation is a cornerstone of zero-trust networks. It involves logically separating all the data, applications, assets, and services according to attack value, access needs, and interdependencies. Then, teams implement granular security policies and controls tailored to the needs of each segment, establishing what are known as micro-perimeters. Rather than trying to account for every potential vulnerability with one large security perimeter, teams can just focus on the tools and policies needed to cover the specific vulnerabilities of a particular micro-segment.

Network micro-perimeters help improve your zero trust security posture with:

• Granular access policies granting the least amount of privileges needed for any given workflow. Limiting the number of accounts with access to any given resource, and limiting the number of privileges granted to any given account, significantly reduces the amount of damage a compromised account (or malicious actor) is capable of inflicting.

• Targeted security controls addressing the specific risks and vulnerabilities of the resources in a micro-segment. For example, financial systems need stronger encryption, strict data governance monitoring, and multiple methods of trust verification, whereas an IoT lighting system requires simple monitoring and patch management, so the security controls for these micro-segments should be different.

• Trust verification using context-aware policies to catch accounts exhibiting suspicious behavior and prevent them from accessing sensitive resources. If a malicious outsider compromises an authorized user account and MFA device – or a disgruntled employee uses their network privileges to harm the company – it can be nearly impossible to prevent data exposure. Context-aware policies can stop a user from accessing confidential resources outside of typical operating hours, or from unfamiliar IP addresses, for example. Additionally, user entity and behavior analytics (UEBA) solutions use machine learning to detect other abnormal and risky behaviors that could indicate malicious intent.

Isolate and defend your management infrastructure

For zero trust to be effective, organizations must apply consistently strict security policies and controls to every component of their network architecture, including the management interfaces used to control infrastructure. Otherwise, a malicious actor could use a compromised sysadmin account to hijack the control plane and bring down the entire network.

According to a recent CISA directive, the best practice is to isolate the network’s control plane so that management interfaces are inaccessible from the production network. Many new cybersecurity regulations, including PCI DSS 4.0, DORA, NIS2, and the CER Directive, also either strongly recommend or require management infrastructure isolation.

Isolated management infrastructure (IMI) prevents compromised accounts, ransomware, and other threats from moving laterally to or from the production LAN. It gives teams a safe environment to recover from ransomware or other cyberattacks without risking reinfection, which is known as an isolated recovery environment (IRE). Management interfaces and the IRE should also be protected by granular, role-based access policies, multi-factor authentication, and strong hardware roots of trust to further mitigate risk.

The easiest and most secure way to implement IMI is with Gen 3 out-of-band (OOB) serial console servers, like the Nodegrid solution from ZPE Systems. These devices use alternative network interfaces like 5G/4G LTE cellular to ensure complete isolation and 24/7 management access even during outages. They’re protected by hardware security features like TPM 2.0 and GPS geofencing, and they integrate with zero trust solutions like identity and access management (IAM) and UEBA to enable consistent policy enforcement.

Defend your cloud resources

The vast majority of companies host some or all of their workflows in the cloud, which significantly expands and complicates the attack surface while making it more challenging to identify and defend the protect surface. Some organizations also lack a complete understanding of the shared responsibility model for varying cloud services, increasing the chances of coverage gaps. Additionally, many orgs struggle with “shadow IT,” which occurs when individual business units implement cloud applications without going through onboarding, preventing security teams from applying zero trust controls.

The first step toward improving your zero trust security posture in the cloud is to ensure you understand where your cloud service provider’s responsibilities end and yours begin. For instance, most SaaS providers handle all aspects of security except IAM and data protection, whereas IaaS (Infrastructure-as-a-Service) providers are only responsible for protecting their physical and virtual infrastructure.

It’s also vital that security teams have a complete picture of all the cloud services in use by the organization and a way to deploy and enforce zero trust policies in the cloud. For example, a cloud access security broker (CASB) is a solution that discovers all the cloud services in use by an organization and allows teams to monitor and manage security for the entire cloud architecture. A CASB provides capabilities like data governance, malware detection, and adaptive access controls, so organizations can protect their cloud resources with the same techniques used in the on-premises environment.
.

Extend zero trust to the edge

Modern enterprise networks are highly decentralized, with many business operations taking place at remote branches, Internet of Things (IoT) deployment sites, and end-users’ homes. Extending security controls to the edge with on-premises zero trust solutions is very difficult without backhauling all remote traffic through a centralized firewall, which creates bottlenecks that affect performance and reliability. Luckily, the market for edge security solutions is rapidly growing and evolving to help organizations overcome these challenges. 

Security Access Service Edge (SASE) is a type of security platform that delivers core capabilities as a managed, typically cloud-based service for the edge. SASE uses software-defined wide area networking (SD-WAN) to intelligently and securely route edge traffic through the SASE tech stack, allowing the application and enforcement of zero trust controls. In addition to CASB and next-generation firewall (NGFW) features, SASE usually includes zero trust network access (ZTNA), which offers VPN-like functionality to connect remote users to enterprise resources from outside the network. ZTNA is more secure than a VPN because it only grants access to one app at a time, requiring separate authorization requests and trust verification attempts to move to different resources. 

Accelerating the zero trust journey

Zero trust is not a single security solution that you can implement once and forget about – it requires constant analysis of your security posture to identify and defend weaknesses as they arise. The best way to ensure adaptability is by using vendor-agnostic platforms to host and orchestrate zero trust security. This will allow you to add and change security services as needed without worrying about interoperability issues.

For example, the Nodegrid platform from ZPE Systems includes vendor-neutral serial consoles and integrated branch services routers that can host third-party software such as SASE and NGFWs. These devices also provide Gen 3 out-of-band management for infrastructure isolation and network resilience. Nodegrid protects management interfaces with strong hardware roots-of-trust, embedded firewalls, SAML 2.0 integrations, and other zero trust security features. Plus, with Nodegrid’s cloud-based or on-premises management platform, teams can orchestrate networking, infrastructure, and security workflows across the entire enterprise architecture.

Improve your zero trust security posture with Nodegrid

Using Nodegrid as the foundation for your zero trust network infrastructure ensures maximum agility while reducing management complexity. Watch a Nodegrid demo to learn more.

Schedule a Demo