A drone attack shut down Yandex’s data center in Russia’s Vladimir region on Sunday, the third major strike on the technology company’s computing infrastructure in four days. Yandex said no employees were injured, but the facility stopped operating and some services became unavailable, according to Reuters.
The sequence marks a significant shift in the technological dimension of the Russia-Ukraine war. Data centers are not merely office buildings: they concentrate servers, networking equipment, storage systems and specialized processors that support search, cloud computing, transportation, payments and artificial intelligence. Repeated physical attacks can therefore affect both consumer services and the computing capacity needed to develop advanced software.
Three facilities hit in four days
The Vladimir shutdown followed attacks on Yandex facilities in Sasovo and the Kaluga region. In an October 9 company statement, Yandex said the Sasovo center had been seriously damaged and completely shut down after an overnight drone attack. The company said part of the Kaluga center was also out of service after a separate strike the next morning.
Sunday’s attack extended that pattern. AP reported that users in Russian cities and in Kazakhstan, Belarus and Armenia encountered problems with Yandex services, including ride-hailing and banking access. AFP said users experienced failures in ride-sharing and payment applications and that Yandex advised cloud customers to activate disaster-recovery plans.
Yandex’s service alert said the remaining resource configuration was unstable while engineers assessed the damage. That language does not establish permanent data loss or the destruction of particular computing systems. It does indicate that redundancy was under pressure after multiple facilities were damaged in quick succession.
A consumer platform and an AI operator
Yandex operates Russia’s leading search engine and a broad digital ecosystem that includes maps, taxis, food delivery, e-commerce, payments and cloud services. Those connections make outages immediately visible to households and businesses. A failed taxi request may appear minor next to a military target, but widespread platform disruption can interrupt commerce, logistics and access to services across multiple countries.
The company also runs major machine-learning infrastructure. Its technical page describes three supercomputers—Chervonenkis, Galushkin and Lyapunov—built around Nvidia A100 accelerators and used for machine-learning tasks. Yandex says its distributed computing platform can coordinate thousands of graphics processors and more than one million CPU cores.
Reuters reported that the Sasovo site housed two of those three systems. The independent TOP500 registry lists Chervonenkis at 21.53 petaflops of measured performance and 29.42 petaflops at theoretical peak. Yandex has not publicly confirmed whether either supercomputer was damaged, so the attacks’ precise effect on AI training capacity remains unknown.
Redundancy faces a physical test
Large cloud systems are designed to survive individual hardware failures. Providers replicate data, distribute workloads and maintain backup capacity across separate locations. Those protections work best when incidents are isolated. Three damaged sites within days create a different problem: workloads must be moved while engineers are still assessing earlier failures, and the remaining facilities absorb more demand with less reserve capacity.
The situation illustrates the distinction between service continuity and infrastructure resilience. A platform can keep many services online while individual regions or products fail. That does not mean the underlying network is unaffected. Yandex acknowledged that restoration would be difficult after the first two attacks and said it could not yet estimate when damaged equipment would return to service.
Disaster recovery also depends on more than software. Replacement servers, networking switches, power equipment and high-end accelerators may be difficult to source quickly during wartime. Sanctions and export controls can further complicate procurement, although no public evidence yet establishes which components were destroyed or whether Yandex lacks replacements.
Digital infrastructure becomes a wartime target
Ukraine previously focused many long-range attacks on Russian energy and military-industrial facilities. The Yandex strikes suggest that large commercial computing sites may now be treated as strategic infrastructure as well. Russia has also attacked Ukrainian telecommunications and data infrastructure, creating a cycle in which civilian-facing digital systems are increasingly exposed to military operations.
Attribution should remain precise. Yandex’s latest public alert described a drone attack but did not identify the operator. Reuters and AP placed the strike within Ukraine’s expanding campaign inside Russia, while Ukrainian officials had confirmed involvement in at least one of the earlier Yandex attacks. No independently verified damage assessment for the Vladimir site was available Sunday.
What remains uncertain
The most important unanswered questions concern duration, data integrity and AI capacity. Yandex has not disclosed how much traffic was rerouted, which cloud regions or services lost capacity, whether customer data was affected or how much specialized hardware remains operational. Reports of outages establish real disruption but do not by themselves show systemic failure.
Even with those limits, the series is consequential. Three rapidly successive attacks forced Yandex to manage simultaneous physical damage, service instability and recovery operations across a nationally important digital platform. The outcome will offer a practical measure of how well hyperscale infrastructure withstands coordinated physical threats—and how quickly a technology company can restore resilience when multiple backup assumptions fail at once.
