Community outreach programs address local needs through employee participation and organizational support initiatives.
By Steve Phillips, Vice President of Human Resources, DataBank
At DataBank, we’re more than digital infrastructure. We’re neighbors, community members, and partners in the cities where our data centers operate. From Dallas to Minneapolis, Phoenix to Northern Virginia, our facilities are woven into the fabric of local communities, and we take that responsibility seriously. Being a good neighbor means showing up when it matters most.
That’s why we launched our company-wide food drive from August through September 2025, inviting teams across our 65-plus data center locations to organize local food donation and other efforts. Food insecurity affects millions of Americans, and the need becomes even more urgent during the fall and winter months when families face additional financial pressures. We wanted to mobilize our national footprint to make a tangible difference where our people live and work, serving the communities we’re proud to call home.
While participation was voluntary, several of our data center teams stepped up in a big way. The results remind us that when you combine local knowledge with genuine commitment, even small actions create meaningful impact.
Here are some highlights from DataBank teams who answered the call.
Denver: Feeding the Rockies
The Denver metro team made a massive impact, donating 1,806 pounds of food to the Food Bank of the Rockies. This generous contribution translated into 1,500 meals for families across the Rocky Mountain region facing food insecurity. The drive brought so many employees together around a shared commitment to community service.
Salt Lake City: Supporting Families and Young Patients
Our Salt Lake City team collected 554 pounds of food and raised $300 in donations for Tabitha’s Way, a Utah organization serving families in need. The team also contributed 43 bags of clothes and six additional boxes of items to Tabitha’s Way.
DataBank SLC team members load donated food items into a vehicle during the company’s 2025 food drive.
Beyond food and clothing security, they demonstrated care for young patients by donating approximately 25-30 handmade fleece blankets along with clothing to Primary Children’s Hospital, bringing comfort to children during difficult times.
Houston: Feeding Families with Variety
The Houston team collected an impressive 286 pounds of food for the Houston Food Bank, carefully assembling donations that provide balanced nutrition for families.
Five DataBank Houston team members stand with donated food items collected during the company’s 2025 food drive.
Their haul included over 130 cans of soup and chili, more than 100 cans of vegetables, 70-plus boxes and bags of pasta, rice, and mac and cheese, a full tote of dried fruit and juices, and 12 bottles of prenatal vitamins — essential nutrition for expectant mothers.
Dallas-Fort Worth: Meals That Matter
Teams across the Dallas-Fort Worth metro came together to donate 174 pounds of food to the North Texas Food Bank. These contributions translated into 145 meals for families facing food insecurity in North Texas.
A DataBank team member delivers donated food items to the North Texas Food Bank, supporting families facing food insecurity across the Dallas-Fort Worth region.
The giving didn’t stop there. The Dallas team also kicked off a full month-long Season of Giving initiative that included toy drives benefiting Toys for Tots to help spread holiday joy throughout the community.
Orangeburg, New York: Community Connections
Our Orangeburg team gathered a diverse collection of food items and delivered them to the Church of the Ascension in New Milford, New Jersey.
Two DataBank Orangeburg team members deliver donated food items to the Church of the Ascension in New Milford, New Jersey.
By partnering with local faith organizations, the team provided donations to the families who need them the most. This collaboration demonstrates how data center teams or other organizations can take advantage of community networks to maximize their total impact.
A Commitment that Extends Beyond our Facilities
The success of our food drive reflects what we believe at DataBank: operating critical infrastructure comes with a responsibility to invest in the communities where we’re located. From supporting local food banks and children’s hospitals to partnering with volunteer fire departments and youth sports leagues, our teams look for meaningful ways to give back throughout the year.
This food drive is just one example of how DataBank employees engage with their neighbors. Our commitment to community partnership doesn’t end when the donations are delivered. It’s how we operate year-round. We’re proud of what our teams accomplished this fall, and we’re grateful to the organizations that work tirelessly to serve families in need. Together, we’re building stronger communities, one act of service at a time.
October was Cybersecurity Awareness Month, but if you missed it, don’t worry. The truth is that cybersecurity education doesn’t have an expiration date. Also, what often gets overlooked is the fact that not everyone learns the same way, and not everyone faces the same threats.
For example, a sixth grader trying to figure out if a Roblox link is legitimate has very different needs than his grandmother who may be dealing with sophisticated phone scams.
This is why we created this list of free or low-cost cybersecurity resources, all organized by age group. Whether you’re looking to protect your kids, help your parents, or increase your own cybersecurity awareness and safety, this list has something for everyone.
Cybersecurity threats don’t stop when you leave work. The same phishing tactics and other threats that target you in the workplace can also show up in personal inboxes, text messages, and social media feeds. When your friends and family know how to spot a scam—no matter their age—everyone benefits.
The resources below aren’t boring training videos or dry compliance courses. Many are interactive games, hands-on simulations, and practical tutorials that actually keep people engaged. Dive in and see what works for the people in your life who need it most.
For Older Children (Upper Elementary and Middle School)
These resources turn cybersecurity lessons into adventures and games, making it easier for kids to learn about phishing, passwords, and online safety without it feeling like homework.
KC7 Cyber Detective: This interactive story-based game allows players to investigate cyber incidents using real-world data.
Education Arcade: Short educational games covering phishing, passwords, and online safety.
KnowBe4: A free cybersecurity kit that includes downloadable activities with puzzles, games, videos, and a Roblox game (Hack-a-Cat) that focus on phishing, passwords, and cyberbullying prevention.
Cyber101: Quick video modules (under five minutes each) also covering phishing and password safety as well as malware and multi-factor authentication (MFA).
Trend Micro Cyber Academy: Interactive lessons and animated videos for children on online safety, privacy, and security basics.
For Teens and Young Adults
Teens and young adults benefit from hands-on training that mirrors real-world threats they’ll encounter in college, first jobs, and daily digital life.
Wizer Security Awareness: Bite-sized, animated modules and phishing simulations that make learning engaging and practical. (Requires free sign up.)
CanIPhish: Hands-on phishing and cybersecurity education with gamified quizzes that test your ability to spot threats. (Requires free sign-up.)
DoD Cyber Awareness Challenge: Mission-based training developed by the Department of Defense covering phishing, malware, data protection, and social engineering through interactive scenarios.
SANS Cyber Aces: Free cybersecurity tutorials covering operating systems, networking, and security fundamentals with hands-on exercises for those interested in cybersecurity careers.
HP LIFE: Interactive course where learners take on the role of an entrepreneur, making security decisions and learning to identify threats like phishing and malware.
CISA Training & Exercises: Government-developed cybersecurity training exercises and resources for building practical security skills. (Requires free sign up.)
Terranova Cyber Games: Immersive 3D game scenarios that put players at the center of real-world cyber threats, teaching detection and response skills through gameplay.
Seniors are increasingly targeted by sophisticated scams, making accessible, straightforward cybersecurity education essential for protecting personal information and finances.
Cyber-Seniors Cybersecurity Center: YouTube videos and downloadable guides designed specifically for older adults learning to navigate online safety.
CISA Older American Resources: Government-curated collection of helpful links and PDF downloads covering common fraud schemes, identity protection, and mobile security.
Internet Safety 101: Comprehensive guide covering everything from password security to credit card scams, with practical tips for staying safe online.
Terranova Social Engineering Kit: Free training videos explaining how scammers manipulate people into sharing confidential information and how to recognize these tactics.
These resources work across age groups and skill levels, offering everything from foundational awareness training to advanced competitions for aspiring cybersecurity professionals.
Amazon Security Awareness (via NIST NICE): Free, digestible cybersecurity awareness training developed by Amazon and aligned with NIST standards, covering phishing, social engineering, data privacy, and secure communication in 17 languages including ASL.
Competitions and Games (NICCS): Directory of cybersecurity competitions spanning all skill levels, from middle school students to federal professionals, including capture-the-flag events, defense challenges, and national competitions.
NIST Online Learning Content: Comprehensive collection of government-curated cybersecurity learning resources, training materials, and career development tools from the National Initiative for Cybersecurity Education.
CyberCIEGE Game: Strategic simulation game developed by the Naval Postgraduate School where players build and defend networks while balancing budget, productivity, and security against real-world attack scenarios.
Building a Safer Digital Future, One Person at a Time
Cybersecurity threats don’t discriminate by age, but the way we learn to defend against them should. Whether you’re helping a child spot suspicious links, guiding a college student through their first phishing simulation, or supporting an older family member in recognizing phone scams, these resources offer practical, engaging ways to build digital safety skills. The best defense against cybercrime starts with education, and it’s never too early or too late to get started.
DataBank’s Red Oak campus is a 3.4-million-square-foot, AI-ready data center development spanning 292 acres. This article goes behind the scenes of the construction, with the numbers and engineering behind one of DataBank’s largest builds.
Phase 1 will expand DataBank’s Dallas market presence to 12 data centers across the metro, with room on the campus for four additional buildings as demand grows. Construction is already underway, powered by a 400MW Oncor substation delivering up to 240MW of critical IT power in Phase 1 and scaling to 480MW at full buildout.
Each two-story building offers 425,000 gross square feet, including 200,000 square feet of data center space. Initial Ready for Service is targeted for Q2 2026.
What does it take to build 3.4 million square feet of AI-ready infrastructure? Our specific construction numbers tell the story.
By the Numbers: The Scale of Construction
The infrastructure going into the ground reveals just how massive this project is.
The electrical backbone:
19,008 power whips requiring over a million feet of Sealtite conduit (198 miles, enough to stretch from Austin to Dallas).
The copper wire inside those whips totals 990 miles (enough to stretch from New York City to Jacksonville, Florida).
The fiber optic network:
1,760,000 feet of 864-count fiber optic cable (333 miles of high-density fiber).
If you connected each individual strand end-to-end, the result would wrap around the Earth 11.5 times.
“Projects like Red Oak don’t just happen because you have land and power available,” said Justin Puccio, executive vice president of corporate strategy. “The infrastructure complexity required to deliver 480 megawatts of AI-ready capacity is enormous. We’re talking about miles of copper, hundreds of miles of fiber, and construction coordination that has to be flawless. This is what it takes to meet the infrastructure demands the AI era is creating, and Dallas is exactly where this capacity needs to be.”
Growing the Dallas Footprint
Red Oak has become a major sub-market near Dallas, attracting some of the world’s largest cloud and technology providers. The new campus expands DataBank’s already substantial presence in the market, which includes its headquarters location (DFW1) and eight other data centers across the metro.
The Red Oak development is DataBank’s third major campus announcement in the past year, following acquisitions in Atlanta and Northern Virginia. When all three sites are fully developed, they’ll add more than 450 acres, 5.8 million square feet of data center space, and 792MW of power to DataBank’s portfolio.
The campus will leverage DataBank’s award-winning Universal Data Hall Design (UDHD), built to meet the demanding sustainability and performance requirements of enterprises, hyperscale cloud providers, and AI applications that need maximum power and cooling density.
Red Oak isn’t just big. It’s 990 miles of copper wire, 333 miles of fiber, and 198 miles of conduit big. That’s the kind of infrastructure the AI era demands.
By Tony Qorri, Vice President of Construction, DataBank.
The artificial intelligence revolution has completely changed so many aspects of virtually every industry today. Organizations have already achieved game-changing, permanent improvements and continue to envision more innovations and benefits.
At the same time, AI is so transformative that it also presents new questions as many companies must develop strategies to take advantage of all it has to offer, outpace their competition or both.
The data center industry is no different, and one subset of our business —data center design and construction— could be affected even more than most. Like any other business, we’re accustomed to responding to market demands and other external factors to constantly pivot when building traditional enterprise data centers.
If those past conditions represented once-predictable winds of change, AI has now created a perfect storm of innovation, opportunity and transformation when designing and building the next generation of AI data centers. Once-standard, long-accepted practices and approaches have now become outdated almost overnight, forcing data center operators and their customers to come up with new ways of thinking related to data center construction.
AI Infrastructure Flexibility
This new environment calls for infrastructure flexibility that adapts as customers’ changing requirements. That means offering multiple configurations for power, space and cooling to accommodate their future computing needs without complete redesigns.
While that flexibility to date has served enterprise and hyperscale customers effectively, AI workloads are now presenting an entirely new dimension. Data center builders are now experiencing “hard right and hard left” requests from different customers. For example, one might need complete air-cooling for network infrastructure while another is looking for extensive liquid cooling equipment for GPU-intensive operations.
These aren’t minor variations. They represent fundamentally different infrastructure approaches and completely new construction strategies.
AI data centers require specialized approaches that go far beyond traditional design considerations and project management approaches. In some cases, they present logistical and technical complexities that didn’t exist just a short time ago.
Liquid Cooling
The shift to liquid cooling represents one of the most dramatic construction changes. Until recently, most traditional data centers used raised floors and perimeter cooling, but more intense AI workloads are now causing facilities to replace computer room air handler units with cooling distribution units. Those systems use secondary chilled water loops made from welded stainless steel components.
This change creates a domino effect that now requires specialized welding capabilities and stronger structural designs to support heavier thermal storage systems. These implications are affecting once-standard construction timelines and processes. It’s more challenging to attempt to fit more specialized work into the same schedules while also engineering new structural designs.
Supply Chain Constraints
Supply chain constraints add another layer of complexity to AI data center construction. Past procurement approaches can be inadequate when attempting to source specialized materials such as high-grade copper components or stainless steel piping and control valves. As a result, we face the risk of dealing with limited availability and extended lead times.
In response, data center builders need to go two to three tiers deeper into the supply chain than in the past. For example, instead of working exclusively with switchgear vendors, form direct relationships with breaker manufacturers and cable suppliers. It’s evidence of more dominos falling: We need to forecast facility requirements years in advance and begin negotiating letters of intent for raw materials before the actual contracting occurs.
This deeper engagement creates a partnership model where suppliers now actively participate in the success of the project as opposed to simply fulfilling purchase orders. This new approach still requires careful attention to economies of scale, but it helps improve the supply chain security that AI infrastructure timelines now demand.
Labor Challenges
Specialized mechanical and electrical trades are in critically short supply today, creating the most challenging aspect of AI data center construction.
Estimates suggest that in North America an additional 439,000 new workers are required to meet pending construction demand for data center projects. Additional research showed the total number of unfilled skilled trade jobs could hit two million by 2033.
If not managed effectively, this growing gap can force construction timelines and costs into unpredictable territory where past relationships with general contractors and current project management processes could come up short.
Builders need to forecast labor requirements up to three years ahead and, if appropriate, bring contractors into pre-construction activities three to six months earlier. Instead of following conventional bidding procedures, owners should engage contractors in design evolution and resource planning before breaking ground.
The scale of this challenge is evident across the industry. Projects that begin with predictable labor costs often face dramatic shifts as construction progresses. It’s not uncommon to see 20% to 30% price increases for mechanical and electrical trades between phases of the same development.
That’s just the new reality of sourcing qualified workers from other markets with increased per diem costs and competitive wages. It’s reached the point where labor escalation is now recognized in multiple active developments and markets.
Adapting to the Speed of AI
One thing is clear: Data center construction needs to quickly adapt to keep pace with the AI revolution. While these new challenges are unprecedented, they may also present new opportunities for operators who can identify new solutions early in the process.
Success now requires treating every aspect of construction—from supply chains to labor partnerships—as strategic investments rather than transactional relationships. As AI workloads continue to evolve, the data center operators who thrive will be those who view adaptation not as an obstacle but as a potential differentiator in an industry being redefined in real time.
DataBank recently published a white paper examining the reality of cloud repatriation, exploring why enterprises are moving workloads back from public cloud, the hidden challenges of repatriation itself, and strategies for avoiding these pitfalls altogether. To learn more, download “The Reality of Cloud Repatriation” now.
Cloud repatriation has moved from exception to expectation. According to recent research from Barclays, 86% of enterprise CIOs plan to move at least some public cloud workloads back to private cloud or on-premises infrastructure. This represents the highest percentage on record as well as a fundamental shift in how organizations think about infrastructure strategy.
Yet what is does mean is that enterprises are refining their strategies based on hard-earned experience, moving from blanket “cloud-only” thinking to strategic workload placement based on actual business requirements.
Our new white paper examines cloud repatriation from three critical perspectives:
Part 1: Examples of the reasons why enterprises are moving workloads back: cost overruns, security complexities, and support limitations driving infrastructure reconsideration.
Part 2: Hidden repatriation challenges such as data egress fees, vendor lock-in, and required architectural changes.
Part 3: Prevention strategies, including flexible infrastructure planning, vendor agnosticism, and strategic workload placement.
Download the full white paper to explore how organizations are rethinking their cloud strategies and building more adaptive infrastructure for the long term.
When Cloud Economics Stop Working
The shift toward cloud repatriation stems from a convergence of operational and financial challenges that emerge as organizations scale their deployments. What may begin as an attractive proposition often evolves into a more complex reality as workloads grow and business requirements change.
For example, cloud providers market their services with straightforward pricing calculators that promise transparency and predictability. However, many organizations quickly discover that actual costs far exceed initial projections. The challenge extends beyond compute and storage to include a growing number of variable charges that multiply as infrastructure scales.
Autoscaling is a prime example. While automatically adding capacity during demand spikes may initially seem like efficiency, it can generate unexpected costs when not carefully configured. Charges for load balancer requests, monitoring and observability tools, data transfer between availability zones, and countless other micro-charges accumulate quickly. What appeared as a $10,000 monthly infrastructure budget can easily balloon to $15,000 or $20,000 once these variables come into play.
The unpredictability becomes an issue for finance teams trying to forecast IT spending. Unlike fixed monthly costs for colocation or private cloud infrastructure, public cloud bills fluctuate based on usage patterns that may prove difficult to anticipate or control. For companies operating at scale, these variable costs can compound dramatically. Many organizations reach a point where the total cost of ownership for public cloud exceeds what they would pay for dedicated infrastructure, often by significant margins.
The Compliance Burden
Achieving compliance in public cloud environments requires substantially more effort, tools, and ongoing management than many organizations initially anticipate. The shared responsibility model sounds straightforward in theory, with cloud providers securing the infrastructure while customers secure their applications and data. In practice, this division creates complexity and gaps that organizations must actively address.
Meeting regulatory requirements like HIPAA, PCI DSS, FedRAMP, or GDPR in public cloud often requires deploying multiple third-party security tools, implementing extensive logging and monitoring, and maintaining detailed documentation of security controls. Unlike private or hosted environments where infrastructure-level security controls are more directly accessible, public cloud customers must layer security solutions on top of the provider’s baseline offerings.
Companies may discover they inherit only limited controls from cloud providers. While providers maintain certifications and secure their underlying infrastructure, customers bear responsibility for configuration, access management, encryption, and data handling. Security teams must become experts in cloud-specific security practices while managing tools and policies across potentially multiple cloud providers.
For organizations in highly regulated industries, these challenges can become critical factors in repatriation decisions. The resource investment required to maintain compliant cloud environments can often exceed the total operational costs of managing on-premises or colocation infrastructure where security controls are more directly managed.
Support and Visibility Gaps
Public cloud providers excel at scale, but often at the expense of personalized support, especially for customers who spend less than millions annually. Organizations discover that support tiers matter significantly, and the level of assistance available to mid-market companies often falls short during critical incidents.
When infrastructure problems arise, many customers find themselves dependent on provider status pages, community forums, and support ticket systems with limited visibility into underlying issues. Public cloud support can feel impersonal and difficult to navigate during outages or performance problems. Customers may have little insight into what actually went wrong or when resolution will occur.
This visibility gap extends beyond support interactions. Public cloud abstracts infrastructure details that IT teams may need for troubleshooting, optimization, or capacity planning. While this abstraction simplifies some tasks, it can frustrate teams accustomed to deeper infrastructure control and visibility.
For mission-critical applications, these support limitations represent real business risk. When revenue-generating systems experience problems, waiting hours or days for cloud provider support to escalate and resolve issues becomes unacceptable. Organizations with demanding uptime requirements often find that the support and visibility they need justifies moving certain workloads back to environments where they have direct control and responsive support partnerships.
Prevention Over Reaction
Understanding why organizations want to leave the cloud represents only half the story. The repatriation process itself introduces substantial challenges that can prove just as costly and complex as the problems organizations are trying to solve. Data egress fees, proprietary service dependencies, and vendor lock-in often require significant application modifications.
The most effective strategy focuses on prevention rather than reaction. Organizations that partner with infrastructure providers offering multiple deployment options while maintaining vendor agnosticism from the outset can adapt as business requirements evolve without facing costly repatriation challenges or becoming trapped by a single provider’s ecosystem.
Our white paper explores these repatriation challenges in detail and provides concrete strategies for avoiding them. Download “The Reality of Cloud Repatriation“ to learn how strategic infrastructure planning and proactive decision-making prove far more effective than attempting to unwind poor cloud decisions after the fact.
In our next post, we’ll examine the hidden costs and complications organizations encounter during the repatriation process itself.
Cybersecurity awareness focuses on educating employees to recognize risks, follow security practices, and reduce human-related vulnerabilities.
By Calli Schlientz, Director of Compliance, DataBank
As Cybersecurity Awareness Month wraps up, it’s worth asking: Why is building effective cybersecurity so difficult for so many organizations? The answer usually isn’t technical. It’s cultural.
Too many organizations view their security and compliance teams as the “department of no,” the people who exist to tell everyone what can’t be done and why their preferred approach won’t work. This adversarial mindset creates exactly what it aims to prevent. When teams see security as an obstacle rather than a partner, they work around it, avoid conversations until it’s too late, and treat compliance as a burden rather than a useful framework.
The path forward requires integrating security into conversations from the beginning and moving from prohibition to collaboration. When security teams ask, “How can we make this work securely?” instead of simply saying “no,” organizations build stronger defenses while maintaining operational efficiency. This cultural shift improves security outcomes, business agility, and customer confidence simultaneously.
It All Starts with Leadership
Consider how you build a physical structure. Before construction begins, architects and engineers calculate load requirements, soil conditions, and structural integrity. Nobody questions whether foundation planning belongs in early conversations. The same principle applies to cybersecurity infrastructure.
Leadership makes this happen through visible participation, not mandates:
When executives complete security awareness training in the first week it’s available, completion rates across the organization increase dramatically
When senior leaders report phishing attempts using the same channels employees use, it normalizes the behavior
When project sponsors proactively request security architecture reviews, teams understand these aren’t bureaucratic obstacles but standard practice
People follow actions far more reliably than words. Leadership visibility transforms security from a requirement into an expectation.
From Policy to Practice: Making Documentation Useful
Most organizations have comprehensive information security policies that few employees have actually read. This isn’t a failure of discipline. It’s a failure of approach.
A 200-page security policy serves no practical purpose if people don’t know how to extract relevant information when they need it. The solution is teaching employees to treat security policies like reference tools they consult when needed, not documents they memorize. Show them how to use search functions effectively, understand how the document is organized, and locate specific guidance quickly.
Test whether your policies actually work as reference tools by creating quizzes that require employees to find information rather than recall it from memory. When people struggle, you’ve identified either a training gap or a documentation problem that needs fixing.
The security policy should answer questions, not create them. If employees can’t find guidance when they need it, even the most comprehensive or technically accurate policy has failed its purpose.
Compliance as Catalyst, Not Constraint
Compliance frameworks like SOC 2, FedRAMP, ISO, and PCI-DSS often get treated as burdens when they’re actually roadmaps that tell you exactly what needs protection and how to protect it.
Rather than viewing these as obstacles to overcome, use them as structured guides for building security practices. When frameworks mandate specific controls for patching, monitoring, or access management, they’re identifying security measures that actually matter based on industry experience and established risk patterns, not creating arbitrary work.
The key is integrating these requirements into planning conversations from the beginning. If you know FedRAMP compliance will eventually be required, design systems with those controls built in rather than retrofitting them later. When security and compliance participate in initial architecture discussions, evidence collection happens automatically throughout the year instead of becoming a scramble during audit season. This creates shared responsibility where everyone understands their role because they helped design processes from the start.
Frameworks and processes establish what needs to happen. Making it actually happen requires the human side of security.
Practical Best Practices for Building Security Culture
Building lasting security awareness requires three interconnected practices: helping people understand what they’re protecting, rewarding positive behavior, and creating safe reporting channels.
Start with the “why” behind security requirements. When employees understand that their actions protect customer data, intellectual property, or operational systems, they ask better questions and challenge processes that seem to bypass security controls. This isn’t a one-time training module but an ongoing conversation that evolves as systems change and new workflows emerge.
Reward people who identify security gaps or report potential issues. When someone flags a phishing email, acknowledge it publicly. When a team proactively requests a security review, recognize that initiative. Even when someone reports their own mistake that created a vulnerability, reward the transparency. Positive reinforcement encourages people to report problems rather than hide them.
Create environments where reporting feels safe by removing reprimand culture for honest mistakes. If employees fear consequences for admitting errors, security gaps stay hidden until they become incidents. The person who reports “I think I clicked a phishing link” deserves thanks for early warning, not punishment.
When Security Culture Becomes Customer-Facing
Internal security culture directly impacts customer relationships. When employees understand security principles and processes, they have better conversations with customers about protection, compliance, and risk management.
Sales teams educated in security fundamentals know when to involve specialists in customer discussions, creating comfort for customers to ask difficult questions about data protection, incident response, and compliance certifications. This proactive approach demonstrates that security isn’t an afterthought addressed when problems arise.
Transparency builds trust. Resources like trust centers that document security practices, compliance certifications, and privacy commitments give prospects detailed information without requiring custom responses for each inquiry. Security culture becomes a competitive advantage when customers see it reflected in every interaction, from initial sales conversations through ongoing service delivery.
Where to Begin
If you can only implement one security culture initiative, start with consistent micro-learning. Weekly security tips delivered via Slack, email, or team meetings work better than quarterly training marathons that nobody remembers.
Keep messages short and practical—badge security one week, phishing identification the next, password management the following week. Small, regular touchpoints build awareness over time without overwhelming people or disrupting workflows.
Make leadership participation visible. When executives complete training promptly, report suspicious emails, and request security reviews for their own projects, it signals organizational priority more effectively than any policy mandate.
Security awareness isn’t a destination you reach through annual training. It’s a culture you build through consistent, practical, positive reinforcement that makes security everyone’s responsibility, not just the security team’s job.
Connectivity enables data centers to link servers, storage, cloud platforms, and users, making infrastructure operational rather than isolated hardware.
By Tim Glatz, Head of Interconnection, DataBank
Most infrastructure decisions begin with the same question: What hardware do we need?
It’s a logical starting point. Organizations carefully evaluate server specifications, storage capacity, and processing power. They compare vendors, negotiate pricing, and plan for future upgrades. Yet, hardware selection represents only part of the infrastructure equation.
The other part–equally if not even more important–is connectivity. Without the right connectivity options, even the most powerful hardware would be left isolated, unable to deliver on its full potential. For example:
Application servers can’t serve customers if network latency undermines response times
Backup systems provide no protection if connectivity failures prevent replication
GPU clusters can’t train AI models if they can’t efficiently sync data between nodes
In the era of AI, real-time data processing, and distributed applications, connectivity architecture has evolved from a supporting consideration to a primary driver of business outcomes.
To put it another way, the question isn’t whether your hardware is powerful enough. It’s whether your connectivity design allows that hardware to perform at the level your infrastructure demands.
Beyond Bandwidth: What Modern Connectivity Actually Means
When evaluating connectivity, most organizations default to a single question: How much bandwidth do we get? Yet, focusing on bandwidth alone leaves out other important parts of the story.
Latency, jitter, and packet loss often matter more than raw throughput, especially for applications requiring real-time responsiveness. A 10Gbps connection with inconsistent latency performs worse than a 1Gbps connection with predictable, low-latency characteristics.
Path diversity and redundancy architecture determine what happens when primary connections fail. A single high-capacity link creates a single point of failure. Multiple diverse paths, even with lower individual capacity, provide resilience that keeps applications running during network disruptions.
The quality of peering relationships and direct access to cloud on-ramps can have a real impact on performance and cost. Direct connections to AWS, Microsoft Azure, or Google Cloud bypass the public internet entirely, reducing latency and improving security while often lowering data transfer costs at scale.
For AI inference, financial transactions, and real-time collaboration, these connectivity characteristics directly determine whether applications can meet business requirements.
The Hidden Cost of “Good Enough” Connectivity
“Good enough” connectivity rarely reveals its costs through obvious outages. Instead, it degrades performance in ways that are difficult to measure and easy to overlook.
Application timeouts trigger retry attempts, which consume resources and cascade across dependent services. For example, pages load slightly slower, API calls take longer, interactive features feel less responsive. These incremental degradations often go unmeasured until customers complain or switch to competitors.
The compound effects become more severe in modern architectures. An additional 50ms of latency multiplies across microservices chains, turning acceptable response times into frustrating delays.
Perhaps, most significantly, inadequate connectivity creates opportunity costs.
AI models requiring sub-100ms inference can’t be deployed. New markets remain inaccessible because connectivity quality can’t support customer expectations. Use cases requiring failover redundancy never launch because path diversity doesn’t exist.
Perhaps most significantly, inadequate connectivity creates opportunity costs. For example:
AI models requiring sub-100ms inference can’t be deployed
New markets remain inaccessible because connectivity quality can’t support customer expectations
Use cases requiring failover redundancy never launch because path diversity doesn’t exist
These limitations don’t just affect current operations; they constrain strategic growth and competitive positioning.
Connectivity Architecture for the AI Era
AI workloads have fundamentally changed connectivity requirements in ways that catch many organizations off guard. Training and inference represent opposite challenges: training demands massive bandwidth for moving datasets and syncing model parameters across distributed systems, while inference prioritizes ultra-low latency for real-time decision-making.
As AI models grow larger and more complex, the “data gravity” problem intensifies. Moving massive datasets to centralized training locations becomes increasingly impractical, yet moving trained models to distributed data sources introduces latency challenges for inference workloads.
These requirements force architectural decisions about when to use private direct connections versus internet transit, where to position edge locations for inference workloads, and how connectivity design impacts total cost of ownership.
Organizations planning AI deployments must recognize that connectivity infrastructure becomes the constraint faster than compute capacity. Today’s “sufficient” connectivity will bottleneck tomorrow’s AI capabilities.
The Interconnection Ecosystem: It’s Who You’re Connected To
Connectivity quality depends not just on specifications but on ecosystem. Carrier-neutral data centers with a wide array of interconnection options create exponentially more value than facilities offering limited connectivity choices.
Direct connections to cloud providers deliver measurable advantages over internet transit: lower latency, improved security, and often reduced costs at scale. For organizations running hybrid or multi-cloud architectures, these direct on-ramps become essential rather than optional.
The physical proximity to partners, customers, and data sources matters more than many realize. API integrations, B2B data exchanges, and real-time partner connections all perform better with direct interconnection. Organizations choosing data center locations based solely on cost or convenience often discover later that they’ve positioned themselves far from the ecosystems they need to access.
Strategic infrastructure positioning means locating where connectivity options are abundant.
Future-Proofing Connectivity Decisions
Technology evolution happens faster than infrastructure refresh cycles. Organizations making connectivity decisions today must anticipate requirements three to five years out, not just current needs.
Certain design principles transcend specific technologies: build for path diversity rather than single high-capacity links, prioritize low-latency options even when current applications don’t demand them, and choose locations based on ecosystem richness rather than just immediate connectivity requirements.
Vendor-neutral connectivity architecture preserves flexibility as business needs evolve. The ability to add new connections, switch providers, or pivot strategies without physical relocation provides business value that’s difficult to quantify but easy to recognize when needed.
From Infrastructure to Business Capability
Infrastructure decisions should begin with connectivity architecture, not hardware specifications. The right connectivity environment makes powerful hardware more valuable. Poor connectivity makes even the best hardware underperform.
Before your next deployment, ask: To whom and to what do we need to connect? What latency do our applications require? How will our needs evolve? What happens when primary paths fail?
Organizations that position themselves in connectivity-rich environments gain advantages that competitors struggle to replicate: faster application performance, more deployment options, and the flexibility to adopt emerging technologies as business demands change.
Without connectivity, your equipment is just hardware. With the right connectivity architecture, your infrastructure can become a new–and sustainable–competitive advantage.
The infrastructure choices organizations make for AI workloads have significant implications for performance, cost, security, and compliance. Our survey reveals that while cloud remains important, a major shift toward hybrid and distributed infrastructure is already underway.
Today’s Reality: Cloud Dominates, But Not Exclusively
Currently, nearly two-thirds (64%) of AI workloads run in either public cloud (49%) or private cloud (15%) environments. Another 15% favor third-party SaaS/web-based platforms like Salesforce.
On-premises and colocation data centers represent a much smaller portion today – just 22% of AI workloads according to our survey (15% on-premises/company-controlled and 7% colocation data centers).
This makes sense. Public cloud and SaaS solutions offer easy starting points for AI adoption, with minimal upfront investment and quick deployment times. However, as AI implementations mature, the limitations of a cloud-only approach become apparent.
Tomorrow’s Strategy: The Hybrid Shift
Looking ahead five years, 96% of respondents expect their AI infrastructure distribution to change. Only 4% report “no significant changes planned.”
Over half of respondents are planning substantial expansions in physical infrastructure:
31% will build more AI-dedicated private/on-premises data centers
22% plan to expand colocation data center deployments
Meanwhile, 43% still expect to increase their reliance on cloud for AI workloads, confirming that the future isn’t about choosing between cloud and physical infrastructure. It’s about strategically combining both.
What’s Driving the Hybrid Approach?
When we asked which factors are most critical in choosing between cloud and colocation for AI workloads, three priorities emerged:
Security and compliance requirements (37% rated this most critical): Colocation facilities like those operated by DataBank provide robust physical security features including 24/7 surveillance, restricted access controls with biometric scanners and security guards, and secure equipment cabinets. They’re also certified with various industry and regulatory compliance standards such as ISO 27001, SOC 1/2, and PCI DSS.
Performance considerations (33%): Low-latency requirements for advanced AI applications demand infrastructure closer to where data is generated and consumed.
Scalability and flexibility (27%): Different workloads have different needs; hybrid approaches allow organizations to optimize each application’s infrastructure placement.
As Philips’ Chief Innovation Officer Shez Partovi noted in the research process: “In a de-globalized world, there is an increasing need to ensure that data is housed and processed in compliance with the specific country or jurisdiction, which is leading to a more decentralized approach.”
Geographic Distribution: AI Is Going Local
Perhaps, the most striking finding in our survey is the extent to which AI is driving geographic expansion of infrastructure. Over three-quarters of respondents (76%) expect their infrastructure to expand geographically over the next five years:
44% need expansion to be closer to data sources
32% need expansion to be closer to end-users
Only 11% anticipate consolidation into fewer, but larger, data hubs, while just 13% expect no significant impact on geographic distribution.
This geographic dispersion addresses multiple needs simultaneously. For compliance, it enables organizations to meet data sovereignty requirements by storing and processing data within specific countries or jurisdictions. For performance, it reduces latency for real-time AI applications like autonomous vehicles, smart city sensors, and industrial IoT.
“We operate in every corner of the world and there are obviously sovereign data requirements,” said Chris Bedi, ServiceNow’s Chief Customer Officer. “To make sure we are serving customers in each region, we need to have a geographically dispersed data center strategy.”
Training vs. Inference: Different Workloads, Different Strategies
Interestingly, while inference is becoming more distributed, AI training is moving in the opposite direction. Nearly three-quarters of respondents (73%) said that “training will be centralized while inference will be more distributed.” This split makes technical sense.
Training requires massive computational power and benefits from centralized, GPU-rich environments. Inference, by contrast, needs to happen close to users and data sources to minimize latency and ensure compliance with local regulations.
Challenges in Scaling Geographically
Expanding AI infrastructure across multiple regions isn’t without challenges. When we asked about major obstacles to geographic scaling, respondents cited:
Security risks in distributed AI (27%)
Access to GPU infrastructure for training (25%)
Data sovereignty and compliance (23%)
Network and latency concerns for inference (20%)
Cost of infrastructure in key regions (19%)
Power and energy constraints (18%)
According to DataBank CEO Raul Martynek, infrastructure readiness varies significantly by location. “It’s not just access to GPUs that matters. You also need somewhere to put them. Not all data centers can accommodate the power and cooling requirements of these infrastructures, and new facilities take time to deploy. That’s why we developed a Universal Data Hall Design for all our new data center builds. It allows us to configure each data hall more quickly in a new facility for whatever infrastructure need that market might have – from traditional air-cooled configurations capable of 15-30kW per cabinet, to liquid-cooled systems capable of 100-200kW per cabinet.”
The Bottom Line
The future of AI infrastructure isn’t an either/or proposition between cloud and physical data centers. It’s a strategic hybrid approach that places workloads where they perform best, meet compliance requirements, and deliver optimal cost-efficiency.
Less sensitive workloads leveraging public datasets can reside in public cloud. More sensitive applications requiring stringent security, compliance, or low-latency performance are increasingly deployed in colocation or private data centers—and those deployments are spreading geographically to be closer to the data and users that need them.
In our next article, we’ll explore how AI strategies are improving with approaches such as blending off-the-shelf applications, custom solutions, and tailored deployment models.
This is the third in a five-part series examining key trends in enterprise AI adoption based on our 2025 AI infrastructure survey. You can read our previous posts on AI ROI findings and what’s blocking AI success, or download the full report, “Accelerating AI: Navigating the Future of Enterprise Infrastructure.”
Real-World Use Cases for High-Performance Computing in Edge AI
Today, the buzz surrounding AI may conjure images of futuristic robots and other sci-fi types of scenarios. While these may or may not come to life sometime down the road, there are still many practical applications. This is especially true in edge computing.
High-Performance Computing (HPC) at the edge, when combined with AI, already holds significant potential to revolutionize industries and solve real-world problems. Yet, to harness its true power, it’s essential to shift the focus from AI fantasies best served up in movies to more realistic, practical use cases.
Understanding HPC at the Edge
At its core, HPC at the edge refers to the processing power and computational capabilities deployed closer to where data is generated and/or needed. This approach minimizes latency, enhances efficiency, and enables real-time decision-making. All of this is critical in today’s fast-paced, high-tech world. AI empowers machines to learn from data, recognize patterns, and make intelligent decisions without explicit programming.
The market momentum tells the story. The global edge AI market was valued at $21.2 billion in 2024 and is now projected to reach $143.1 billion by 2034, which represents a 21% CAGR. This explosive growth reflects how organizations across industries are moving beyond experimentation to production-scale deployments.
Benefits and Challenges Related to HPC in Edge AI
Edge AI also empowers devices to perform complex tasks locally, reducing dependence on cloud services and enhancing privacy since sensitive data remains on-premises. With 5G rollouts enabling data rates 10 times faster than previous networks and latency below 10 milliseconds, edge computing can now handle increasingly sophisticated AI workloads in real time.
Despite its potential, deploying AI at the edge presents challenges including limited computing resources, energy constraints, and security concerns. All must be carefully addressed for seamless operation. Additionally, the diversity of edge environments requires adaptable solutions capable of meeting varying needs and conditions, from smart cities to remote industrial sites.
HPC Edge AI Use Cases
While AI has real potential, its implementation and adoption must be driven by real-world considerations. Fortunately, the combination of HPC and AI at the edge present many different possibilities for a wide range of industries. For example, in healthcare, the real-time analysis of patient data at the point of care can help with early diagnosis, personalized treatment plans, and remote monitoring.
The following are additional examples of edge AI use cases.
Healthcare and Research Computing
In healthcare and research, HPC at the edge enables breakthrough capabilities. Organizations like the University of Maryland leverage HPC infrastructure to support groundbreaking research in genomics, medical technology, and materials science. By deploying high-density computing power with optimal cooling solutions, research institutions can process complex datasets in real time, accelerating discoveries that would be impossible with traditional infrastructure.
Predictive Maintenance in Manufacturing
In industrial settings, edge AI can predict equipment failures and minimize downtime. Potential anomalies can be detected by analyzing sensor data in real time. This enables them to be addressed proactively, leading to improved efficiency and significant cost savings.
HPC at the edge can be employed to process sensor data from machinery and equipment within the manufacturing plant in real time. This data may include temperature, vibration, pressure, and other parameters indicative of the equipment’s health and performance.
Artificial intelligence algorithms deployed at the edge can analyze this data to detect patterns, anomalies, and early signs of potential failures. For example, machine learning models can identify deviations from normal operating conditions. Then they can predict when equipment is likely to fail and recommend proactive maintenance actions to prevent downtime.
Industrial Smart Grids
Smart grids are modern electrical grids that integrate advanced communication, control, and monitoring technologies to efficiently manage the generation, distribution, and consumption of electricity. HPC at the edge combined with AI can play a crucial role in optimizing energy usage and grid stability.
For example, HPC resources deployed at the edge can process data from smart meters, sensors, and other grid devices for real-time energy monitoring. AI algorithms can monitor energy consumption patterns, detect fluctuations in demand, and identify areas of inefficiency or potential grid congestion.
AI-powered edge computing can also help with fault detection and response. By analyzing data from grid sensors and devices, AI algorithms can identify anomalies indicative of potential problems and trigger automated responses or alerts for timely maintenance or repair.
Retail: Personalized In-Store Experiences for Shoppers
In retail, edge AI enhances customer experiences through personalized recommendations. Retailers can leverage AI algorithms to analyze vast amounts of customer data in real time. This could include browsing history, purchase behavior, demographic information, and real-time location data within the store.
This enables retailers to create personalized experiences for shoppers that enhance engagement, increase sales, and foster brand loyalty. For example, retailers can use this data to generate real-time product recommendations and discounts. These can then be delivered via mobile apps, digital signage, or in-store displays. This gives retailers the ability to direct shoppers toward products that they’re more likely to buy, increasing sales and fostering customer loyalty.
Autonomous Vehicles
Similarly, in autonomous vehicles, edge computing and AI enables the instant processing of sensor data. It’s an important advantage because autonomous vehicles require real-time processing of vast amounts of sensor data to navigate safely and make split-second decisions. Traditional centralized computing architectures may struggle to meet the stringent latency requirements for processing this data. This includes the need to send data back and forth to a distant data center can introduce delays that are unacceptable in safety-critical applications.
By leveraging edge computing, HPC resources can be deployed closer to the source of data generation, such as within the vehicle itself or at the edge of the network. Modern edge AI chips like NVIDIA’s Drive Thor deliver up to 1,000 tera operations per second (TOPS) of processing power, enabling faster sensor data processing and quicker decision-making without heavy reliance on centralized data centers.
HPC at the Edge: Shaping Future AI Use Cases
The convergence of HPC and AI at the edge holds tremendous promise for companies in virtually every industry. By focusing on practical use cases that can successfully address real-world challenges, companies can drive innovation, enhance efficiency, and improve outcomes.
Multi-cloud compliance involves managing security controls, governance policies, and regulatory requirements consistently across multiple cloud and infrastructure environments.
By Calli Schlientz, Director of Compliance, DataBank
Compliance rarely drives the initial decision to move to the cloud. Organizations typically migrate for cost reduction, scalability, or operational efficiency. Compliance tends to be viewed as necessary but secondary.
Yet once the strategic decision to adopt cloud infrastructure is made, compliance becomes a critical differentiator. Organizations must choose which provider and deployment model best meets internal compliance processes and ultimately, regulatory requirements.
This decision creates immediate challenges for organizations transitioning from traditional on-premises infrastructure to the cloud. For example, in cloud partnerships, responsibility for compliance controls is divided between the customer and the provider through what’s known as the shared responsibility model. Security controls that the company once implemented and maintained now exist in hybrid frameworks where some responsibilities transfer to providers while others remain with customers.
Companies must reframe their understanding of compliance ownership, particularly when evaluating multiple cloud providers. Compliance processes and requirements may be handled differently by each vendor, leading to complex comparison challenges and potential issues down the road.
Critical Questions Organizations Should Ask
When evaluating cloud and infrastructure providers, organizations often struggle to know exactly what compliance questions to ask. Many companies understand they need “compliance” but aren’t sure how to evaluate whether a provider can deliver what they need. This uncertainty can lead to gaps in due diligence that create problems later.
The most important questions organizations should consider include:
What are the provider’s current audit exceptions? Look beyond certifications to understand other areas of non-compliance that could impact your data security.
How does the provider handle compliance gaps when they occur? The question isn’t whether issues will arise, but how transparently and effectively they’re managed.
What specific compliance responsibilities will transfer to the provider versus remain with you? Get detailed documentation showing exactly who handles what under different regulatory frameworks.
Can the provider participate directly in your compliance assessments? Beyond providing documentation, can they join auditor calls and provide real-time explanations of security controls?
How quickly does the provider resolve compliance issues? Understanding remediation timelines helps you assess potential risks to your own compliance posture.
These questions help organizations move beyond surface-level compliance discussions to understand how partnerships will function during actual regulatory assessments and ongoing compliance management.
What Leading Partners Bring to Compliance
The most valuable infrastructure providers go beyond providing secure facilities. They offer comprehensive compliance frameworks, structured sets of security controls and procedures that meet regulatory standards. Organizations can inherit or build upon these frameworks, transforming compliance from a wholly internal challenge into collaborative effort.
Transparency forms the foundation of effective partnerships. Leading providers should offer valuable resources such as trust centers (centralized portals containing compliance documentation and audit reports) and responsibility matrices that clearly delineate which compliance requirements they handle versus which remain with customers.
These documents should be framework specific. For example, they should recognize that FedRAMP requirements differ significantly from PCI DSS or HIPAA obligations. Service descriptions should reflect this tailored approach, with providers offering different recommendations based on specific compliance objectives rather than one-size-fits-all solutions.
Public Cloud Compliance Challenges
While public cloud providers offer extensive compliance certifications, they operate on fundamentally different models that can create challenges for organizations with specific regulatory requirements.
The challenge stems from how public cloud providers approach compliance. They build their frameworks to work for as many customers as possible, using a one-size-fits-all model.
This approach means organizations often get more compliance controls than they actually need while potentially missing specific requirements for their industry or regulatory environment. The standardized model can leave gaps for companies in heavily regulated sectors.
When Complexity Multiplies
While choosing the right provider helps address many compliance concerns, the complexity multiplies when organizations operate across multiple cloud environments. Many companies find themselves in multi-cloud scenarios due to acquisitions, specialized application requirements, or deliberate risk mitigation strategies that distribute workloads across different providers.
Each additional cloud environment introduces new compliance variables that compound the challenge. What appears straightforward with a single provider becomes exponentially more difficult when coordinated across multiple providers, each with different frameworks, documentation standards, and support models.
Different providers may handle identical compliance requirements through entirely different mechanisms, creating potential coverage gaps or unnecessary overlaps that waste resources. Organizations often discover these inconsistencies only during compliance assessments or when responding to security incidents that span multiple environments.
The Colocation Alternative
Instead of accepting these limitations, organizations can turn to colocation providers with comprehensive managed services that offer significant compliance advantages over both traditional hands-off models and public cloud approaches.
In these cases, colocation partners can assume substantial compliance responsibilities that would otherwise require dedicated internal expertise, including deploying pre-hardened operating systems, maintaining security patches, and managing ongoing compliance monitoring.
More importantly, colocation operators can customize solutions to address specific regulatory requirements rather than offering generic frameworks. This might mean tailored vulnerability scanning for PCI compliance or specialized background check requirements for criminal justice systems.
During compliance assessments, these providers can participate directly in auditor calls and provide real-time explanations of security controls, rather than simply directing customers to documentation or self-service portals.
For organizations already operating in multi-cloud environments, colocation providers can serve as a unifying platform that simplifies compliance management. Rather than maintaining separate compliance documentation and audit processes across multiple public cloud providers, organizations can consolidate workloads or use colocation as a compliance-consistent hub for hybrid architectures.
Partnership in Practice: A Federal Agency Case Study
The value of choosing the right compliance partner becomes clear through long-term relationships that evolve alongside changing regulatory requirements.
Consider the example of a federal agency that engaged with us before achieving their FedRAMP authority to operate (ATO), the formal approval required for government systems to process federal data.
This partnership began years before the agency secured compliance authorization, with us actively participating in audit calls, adjusting configuration settings based on auditor feedback, and redeploying test and production environments as requirements evolved. When the agency needed to migrate to our updated FedRAMP Rev. 5 environment, the transition was handled as a joint project rather than a vendor handoff.
The ultimate validation came when the customer’s agency granted authority to operate using the implemented and inherited controls, creating a shared compliance framework where both parties validated each other’s security controls. This demonstrates how compliance relationships can evolve from vendor-customer transactions into strategic collaborations that benefit both parties.
The Evolution of Compliance: What’s Coming Next
The compliance landscape is changing in two key ways. Organizations can expect new regulations targeting emerging technologies, particularly artificial intelligence, as regulators work to keep up with rapid technological changes.
More importantly, the industry is moving away from traditional annual audit cycles toward continuous compliance monitoring. Forward-thinking providers are already conducting audits throughout the year rather than concentrating everything into single periods. This approach makes more sense since compliance issues can arise anytime, not just during scheduled audit windows.
Privacy regulations are also expanding. While GDPR transformed data protection in Europe, comprehensive federal privacy legislation in the United States remains stalled. Meanwhile, individual states continue implementing their own privacy laws, creating a complex patchwork that will likely push federal lawmakers to act.
Building Resilient Compliance Strategies
Successfully managing compliance across multiple cloud environments comes down to choosing strategic partners who understand how modern infrastructure connects. Organizations that tackle shared responsibility challenges proactively and design systems that streamline audit processes position themselves better for both current requirements and future changes.
The key is thoughtful planning and selecting partners who have deep compliance expertise. Multi-cloud environments don’t have to compromise compliance, but they do require careful coordination and providers who know how to manage complexity effectively.