Your Fabric Capacity Isn’t Slow — It’s Throttled
Last Updated on August 25, 2026 by Editorial Team
Author(s): Priyanka Shah
Originally published on Towards AI.
Bursting, smoothing, carryforward and burndown — plus the Capacity Metrics chart that tells you whether you can wait it out or have to intervene.
The reports didn’t break. That was the confusing part.
The article explains that Fabric throttling isn’t a simple “speed limit” but a shared compute budget managed through bursting (running above your SKU briefly), smoothing (spreading the borrowed cost across future timepoints), and throttling (blocking new work once future capacity is overrun, with carryforward as debt and burndown as repayment). It walks through how the Capacity Metrics app should be read—especially using System events and throttling stages rather than assuming that utilization spikes mean throttling—then provides a real case where smoothing and carryforward ultimately lead to deep rejection that can even prevent fixing the workload. The author also outlines practical diagnostics (including a 60-second checklist), common traps (refreshes can continue while users fail, “failed” jobs may still consume), and what actions stop throttling fastest (wait, temporarily scale up, pause/resume, or—preview—capacity overage). Finally, it covers prevention practices for small capacities, the key trade between scheduling vs sizing, and how billing models (pay-as-you-go vs reserved) affect the decision of how to manage workload patterns.
Read the full blog for free on Medium.
Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.
Published via Towards AI
Towards AI Academy
We Build Enterprise-Grade AI. We'll Teach You to Master It Too.
15 engineers. 100,000+ students. Towards AI Academy teaches what actually survives production.
Start free — no commitment:
→ 6-Day Agentic AI Engineering Email Guide — one practical lesson per day
→ Agents Architecture Cheatsheet — 3 years of architecture decisions in 6 pages
Our courses:
→ AI Engineering Certification — 90+ lessons from project selection to deployed product. The most comprehensive practical LLM course out there.
→ Agent Engineering Course — Hands on with production agent architectures, memory, routing, and eval frameworks — built from real enterprise engagements.
→ AI for Work — Understand, evaluate, and apply AI for complex work tasks.
Note: Article content contains the views of the contributing authors and not Towards AI.
Related posts
Recent Posts
