Amazon Web Services suffered a significant outage across the Middle East that knocked at least two availability zones offline and disrupted essential cloud services for financial institutions, investment platforms, and technology companies across the region. The incident began Sunday evening local time when unidentified objects struck one of AWS’s UAE data centers, triggering a fire that forced authorities to shut off power to the facility.
The outage disrupted core AWS services including EC2, RDS databases, DynamoDB, and over 60 additional cloud services. Customers unable to access critical networking APIs experienced cascade failures across their entire infrastructure. AWS warned that full recovery would take several hours and advised customers to back up critical data and shift workloads to unaffected regions.
Abu Dhabi Commercial Bank reported that its mobile applications and online banking platforms became completely unavailable to customers. The bank acknowledged a region-wide IT disruption though did not explicitly attribute it to the AWS incident. Investment platform Stake reported service disruptions that prevented users from accessing their accounts and trading capabilities. Sarwa, an investment platform based in the UAE, also experienced outages affecting customer access. Liv by Emirates NBD, the digital banking service run by one of the region’s largest banks, reported temporary service disruptions that left customers unable to use the platform.
Al Ghurair Investment, one of the UAE’s largest investment firms managing billions in assets, was impacted by the outage. Dubai Islamic Bank, a major player in Islamic finance across the Gulf, experienced service interruptions. These institutions serve millions of customers and handle significant transaction volumes daily, meaning even brief outages result in substantial operational and financial consequences.
According to industry benchmarks, cloud downtime costs between 5,600 and 9,000 dollars per minute for mid to large enterprises. For financial services companies specifically, a three hour outage translates into millions in losses due to suspended transactions, customer service disruptions, and potential regulatory issues around settlement and reporting.
The financial impact extended across multiple business segments. E-commerce platforms saw transaction processing failures. SaaS products serving enterprise customers went offline. Fintech companies that depend entirely on AWS infrastructure for payment processing and account management were completely non-functional during the outage. Customers received error messages when trying to access accounts, place trades, or conduct any financial transactions.
The most significant impact fell on companies running infrastructure in a single availability zone. Those organizations experienced complete service failure with no automatic failover capability. Companies that had deployed across multiple availability zones within the region experienced degraded performance but maintained partial service capability. Organizations running redundancy across different geographic regions were largely unaffected, though this requires significantly higher infrastructure costs that smaller companies and startups cannot justify.
AWS reported that the second availability zone in the UAE region also experienced a localized power issue, expanding the disruption beyond the initial strike zone. Bahrain region experienced additional connectivity and API errors that degraded services for dozens more AWS offerings. The company said engineering teams were actively deploying mitigation measures and expected recovery to proceed gradually as updates were rolled out.
For startups and smaller financial technology companies in the Middle East, the outage exposed significant vulnerability. Many had migrated to AWS to avoid maintaining their own data center infrastructure and reduce capital expenses. The theory was sound: AWS handles redundancy and disaster recovery so smaller companies do not have to. But this outage showed the limits of that approach when physical infrastructure itself becomes unavailable.
Enterprise customers who use AWS in the region immediately began redesigning their architecture to avoid concentration risk. Some announced plans to deploy backup capacity in other regions entirely, even though this substantially increases monthly cloud bills. Others began evaluating alternative cloud providers or hybrid approaches that keep critical workloads on premises. The outage highlighted that geographic concentration in a single region, even across multiple availability zones, carries significant risk that many organizations had not properly quantified.
Recovery took several hours from the time power was restored, longer than typical AWS outages in other regions where such complications do not arise.
