Spot Instance

A spot instance is a virtual machine that runs on a cloud provider's unused capacity at a lower price than regular on-demand instances. In return, the provider can interrupt or reclaim it when it needs the capacity back, usually with a brief warning. Providers use different names, such as Spot VMs or preemptible VMs, and different rules.

Why it matters

Spot capacity can cut compute cost significantly for work that tolerates interruption: batch jobs, rendering, CI runners, data processing, and stateless workers behind a queue. It is a poor fit for a single database or an application with no replacement plan. Prices and interruption rates vary by instance type and region, so teams commonly diversify across types and zones.

Example

A worker fleet handles a job queue:

10 workers on spot instances
Provider reclaims 2 with a 2-minute warning
Workers stop taking new jobs, return unfinished work to the queue
Auto scaling group launches replacements

Because the jobs are retryable, users see no failure.

Common mistakes

• Running the only copy of a service on spot capacity.
• Ignoring the interruption notice instead of handling it to save state.
• Assuming the discount percentage is fixed; it changes.

Related

• Reserved instance
• Auto scaling
• Message queue
• Amazon EC2

यह पेज सामान्य जानकारी है, जिसे हमारी संपादकीय टीम ने लिखा है और दस्तावेज़ीकरण बदलने पर अपडेट किया जाता है। सर्वर और प्रदाता अलग-अलग होते हैं, इसलिए बदलाव पहले सुरक्षित तरीके से जाँचें। देखें: संपादकीय नीति और अस्वीकरण.