Bank Cloud Platforms to Face Uniform Reliability and Security Tests
Cloud providers hosting bank data would have to pass common tests covering data center reliability, management system uptime and recovery, under a draft document.

Cloud platforms used by banks and fintech companies would have to pass a single set of reliability and security tests before hosting financial data, according to a draft document describing the criteria. The checks cover data center infrastructure, the stability of the cloud management system and the ability to restore data after failures.
The criteria are being developed by cloud providers and financial institutions within a working group of the industry test site at the "Finance" industrial competence center.
What the draft requires
Among the mandatory criteria is hosting the platform in a data center rated Tier III or higher under the Uptime Institute classification, which allows planned equipment maintenance without shutting the facility down. The cloud resource management system must be available at least 99.95% of the time. Backup and virtual machine recovery must also be verified.
Testing will also check whether the platform has an attested environment meeting FSTEC Order No. 239, which covers the security of significant critical information infrastructure objects. Public trials are planned for the fourth quarter of 2026.
Having two independent data centers is proposed as an optional criterion, and testing cloud operation under a full outage of one site remains only a recommendation.
Industry reaction
Nikolay Kozak, deputy chairman of the management board at Dom.RF bank, said critical banking systems already have separate requirements. A full shutdown of one of two sites is a catastrophic-failure scenario, he noted, and banks do not plan to move their entire infrastructure to the cloud, only less critical services or capacity.
K2 director Sergey Zinkevich said uniform criteria are being proposed for the first time. Audits in the infrastructure sector usually give providers specific findings and a chance to fix them, he said, though the exact procedure depends on the working group.
Ilya Vasilchenko of Turbo Cloud said power supply could be a serious problem for sites with weak engineering protection, and the tests should show whether a platform stays available when failures such as the loss of one power feed are simulated. A cloud provider source said the requirements are achievable through data center design, with multiple power lines, uninterruptible power supplies and backup diesel generators.
Apps worth a look
Tap the right tiles, add them up, hit the target. Easy to learn, hard to master. Free, for all ages.
Talk to an AI conversation partner; every sentence comes back corrected and explained in your language. 9 languages, A1 to C2.
Open EDIFACT and X12 messages, see every segment explained, convert to JSON/XML, export PDF and Excel. Runs locally, no upload.
More from this day
See all →
Kakao Mobility and Korean Army to Build Defense AI Control System

Hyundai Motor Group Suspends External AI Tools in Core R&D

GMO Internet Admits Risk Analysis Gaps After VPS Breach
Conlio raises €6.4 million for AI finance automation in freight
More in: Business
See all →
Gökçe Law Partnership Earns Raydorf AI-Integrated Certification

Marvell to Acquire Lumentum in $115 Billion All-Stock Deal

SUSE Announces Major Job Cuts as US Newspapers Sue Brave Software
