On Monday, a significant disruption at Amazon Web Services (AWS) caused widespread internet outages, impacting roughly one-third of online services globally. This interruption affected a broad spectrum of platforms, from financial institutions to artificial intelligence tools. The root cause was identified as a regional gateway malfunction on the US East Coast. Although AWS has since restored full functionality, the incident has sparked intense debate about the risks of depending heavily on a single cloud provider and the financial implications of safeguarding against such failures.
Among the affected were two Web3 startups: Azza, a Nigerian social commerce cryptocurrency application, and AZA Finance, a cross-border payment service reportedly in acquisition talks with Latin American fintech giant dLocal. Both companies faced temporary interruptions in transaction processing and user interface availability, highlighting that even blockchain-based enterprises are not immune to centralized infrastructure vulnerabilities.
This event underscores a critical challenge for emerging tech companies reliant on AWS and similar cloud services. While implementing a multi-cloud, redundant server architecture can significantly enhance system reliability, it also strains limited startup budgets. Balancing the need for operational resilience against cost constraints is becoming a pivotal strategic decision for early-stage ventures.
Palremit, a Nigerian platform facilitating crypto-to-fiat conversions, managed to maintain uninterrupted service during the outage by distributing its servers across multiple AWS regions.
“A redundant infrastructure automatically reroutes traffic to alternative servers when one region experiences downtime,” explained CEO Ugochukwu Mbamalu. “This approach is essential for maintaining continuous uptime.”
Mbamalu noted that while the redundant setup adds approximately $10 to monthly expenses, these costs can escalate as the company scales. Currently, Palremit allocates several hundred dollars monthly to cloud services for managing user data and database operations.
“Having experienced a cloud outage firsthand, I decided to diversify our server options to mitigate future risks,” Mbamalu added.
Similarly, Azawire, a Nigerian startup specializing in stablecoin payments, avoided service disruption because its AWS infrastructure is hosted in Frankfurt, Germany.
“The AWS failure was localized to the US,” said Emmanuel Onyo, Azawire’s CEO. “Since our servers are based in Frankfurt, we experienced minimal latency or downtime.”
James Lucky, CTO of GrowSafe, a Nigerian crypto swapping platform, reported that his startup’s operations were largely unaffected due to hosting most services on DigitalOcean.
“My projects are primarily on DigitalOcean and other providers, so the AWS outage didn’t impact us much,” Lucky said. “However, some applications hosted on Vercel went offline, which made me suspect Vercel relies on AWS infrastructure.”
Lucky praised DigitalOcean, a US-based cloud provider holding about 2% of the market share, for its affordability, flexibility, and rapid deployment capabilities, making it ideal for early-stage startups.
“You can run various applications on DigitalOcean starting at just $5 per month,” he explained. “It supports everything from frontend and backend services to scheduled tasks, and I’ve never encountered downtime with them.”
Adaobi Orajiaku, CEO of Atsur, a Nigerian Web3 art-tech company, revealed that her AWS-hosted platform remained operational during the outage because it was not engaged in resource-intensive tasks at the time.
“We weren’t conducting any onsite archiving or verification processes during the downtime,” Orajiaku said. “Had we been, we might have faced challenges, potentially resorting to manual cataloging and delaying artwork verification.”
While Atsur’s users typically contact the team directly if issues arise, no disruptions were reported during the incident. The startup continues to monitor system performance closely and maintains communication with its B2B gallery clients to ensure stability, according to Orajiaku.
It is paradoxical that Web3 startups, which advocate for decentralization, still rely heavily on centralized cloud infrastructures. Although blockchain technology guarantees continuous transaction processing with high availability, the management of user data and platform operations depends on centralized providers like Amazon, Google, and Microsoft. This dependency creates tension whenever foundational internet services experience outages, yet alternatives remain limited for most startups.
“Such outages are rare,” Onyo remarked. “If major global companies like Perplexity and Coinbase haven’t fully diversified their cloud infrastructure, it’s understandable that startups face similar challenges.”