News Summary:
On April 24, 2026, Datadog announced its GPU Monitoring service is now helping service providers and enterprises tackle "GPU sprawl" by providing unified visibility across the AI stack, aiming to reduce AI costs through a single view of GPU fleet health, cost, and performance. Earlier that day, the company highlighted that the new product addresses rising AI infrastructure costs stemming from a lack of clarity regarding GPU usage by offering enhanced visibility. The GPU Monitoring service became available to customers everywhere on April 24, designed to help teams plan capacity, troubleshoot issues quickly, prevent costly failures, and avoid wasted spend as businesses scale AI projects. Previously, on April 23, Datadog globally launched its GPU Monitoring service, offering developers, machine learning engineers, and platform teams a comprehensive view of GPU health, workload performance, and spending to control the cost and use of graphics processing units in AI workloads. Separately, on April 23, MegazoneCloud signed a strategic collaboration agreement (SCA) with Datadog, marking the first such partnership in the Asia-Pacific region, which will combine MegazoneCloud's AI and cloud deployment capabilities with Datadog's observability technology to support enterprise customers running generative AI services more reliably.
Subscribe for full access to Datadog's profile
Subscribe for full access to Datadog's products in full detail 
