● Covered by 1 source · 1 reportMedium impact1 neutral
AWS Cloud Capacity Largely Sold Out Through 2027, On-Demand Headroom Reserved
Key points
- AWS committed capacity largely sold out through 2027, into 2028.
- Headroom reserved for On Demand, EC2, and Bedrock services.
- Capacity Blocks for ML rates increased by 15% in January.
- Some developers report current constraints on regular EC2 instances.
AWS Capacity Commitments
Amazon CEO Andy Jassy announced that AWS's large committed cloud capacity is largely sold out through 2027, with a significant portion of 2028 also allocated. This includes capacity sold through On-Demand Capacity Reservations and Capacity Blocks for ML, where customers reserve instances in specific Availability Zones for fixed future windows.
Jassy confirmed that capacity for 2026 is already fully allocated. AWS continues to reserve headroom for On Demand, EC2, and its inference service Bedrock, despite the high demand for long-term commitments.
Increased Demand for ML Resources
The high demand for capacity, particularly for GPU instances, led AWS to raise Capacity Blocks for ML rates by approximately 15% across all regions in January. This increase affected P5en, P5e, P5, P4d GPU instances, and Trn2 and Trn1 instances, reflecting the growing need for machine learning infrastructure.
Analyst Perspective on 'Sold Out' Claims
Cloud analyst David Linthicum suggested that the 'sold out' reporting might overstate the situation for typical enterprise users. He distinguished between the massive, pre-reserved hyperscale capacity blocks sought by large model providers and the day-to-day GPU capacity an enterprise might allocate via the AWS console. Linthicum noted that such headlines can contribute to market confusion regarding reserved instances, on-demand services, and committed capacity.
Current Availability Concerns
Despite the long-term commitments, some developers have reported recent issues with obtaining compute capacity for regular EC2 instances. Comments on platforms like Reddit indicate difficulties in securing even on-demand instances, with some experiencing constraints in production Kubernetes clusters and for CI workloads.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Reporting from
Amazon CEO Andy Jassy stated that AWS's large committed cloud capacity is largely sold out through 2027, with portions of 2028 also allocated, while maintaining headroom for On Demand, EC2, and Bedrock services. This indicates high demand for long-term cloud resource commitments, particularly for GPU instances, though some analysts suggest the 'sold out' claim may not apply to typical day-to-day enterprise allocations.