1 回答
- 最新
- 投票最多
- 评论最多
1
I am not aware of any way to find out the size of the capacity pools.
But I can make an assessment of the likelihood of a specific instance type being unavailable (or being withdrawn at short notice) by looking at the historic SPOT PRICE movements.
Go to the EC2 console, choose "Spot Requests", and click on "Pricing History". This will give you the historic trend of prices for a specific instance type in that Region. I usually look at the "3 month" trend for that instance type, such as p3.16xlarge.
You can also click on "Spot placement score". This helps you to select optimal Regions or Availability Zones to run workloads that can use multiple instance types.
Hope this helps.
已回答 2 年前
相关内容
- AWS 官方已更新 2 年前
- AWS 官方已更新 1 年前
- AWS 官方已更新 2 年前
Thanks a lot! From aws doc, I can find the following statement, "When the maximum price per hour for your request exceeds the Spot price, Amazon EC2 fulfills your request if capacity is available." The pricing history can help me know the spot instance price, this is helpful for price biding preemption. But there is also capacity-based preemption. I wonder how the capacity changes since it will determine how many spot instances I can request.
There's no way for customers to see capacity information (other than "did I get a launch failure"). The best ways to have overall fleet availability are:
Thanks a lot!
Spot Placement Score is the most appropriate way to understand how likely it is that a Spot request will succeed in a Region or Availability Zone. (https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-placement-score.html)