AWS is sending an invoice at the end of each month with the price of the EC2 instance and the number of used hours. Since the price fluctuates, which price is taken to compute the bill? The price at the end of the month, at the beginning of the month or the real price for each used hour?
Thanks
If you are using on-demand EC2 instances, the prices don't fluctuate. When AWS does reduce prices, the new pricing takes effect on a specific date and your bill should reflect the reduced prices as of that date.
There are some complexities in billing when you have credits or Reserved Instances, especially when you use multiple accounts with Organizations or Consolidated billing.
There are some products that have thresholds - for example, S3 standard storage gets cheaper the more you use - the first 50TB are $0.023/GB, the next 450 TB are $0.022/GB, etc.
Related
Does Amazon price for GB storage reset every month? Say in January I used 1000GB of storage and pay it off for $40. Do I get charged $40 in February even if I don’t use any storage or do I pay $0 in February? Basically is AWS storage price static or does it reset every month?
Amazon S3 is charged on two main dimensions:
Storage: A charge for the amount of data kept in Amazon S3, charged per GB per month
Requests: A charge for the number of requests made to S3, charged per 1000 requests
If you keep 1000GB of data in Amazon S3 as Standard storage charged as $0.023 per GB, then you would be charged 1000 x $0.023 = $23 for that month of storage.
If you kept 1000GB of data in Amazon S3 for half a month, then it would be only half that cost.
Think of it like parking your car in a parking garage -- they charge you for every hour that you keep your car parked. However, if your car isn't parked in the garage, then they don't charge you anything.
So I am using multiple resources in AWS. The charges make sense for most services. however, at every first day of the month, I am getting charged more than I should.
For example; charges for services such as, Aws active directory service, EC2, step function, the elastic search is normal throughout the day. as I can drill down daily charges for these services. but every first day of the month, AWS charging more than what it usually charges. For example; if the average charge for EC2 instances for the day is $5 then at every 1st date(let's say 1st January) of the month, the charges end up for EC2 instance somewhere around $15 and again on 2nd January charges would back normal to $5.
also, for the AWS Contact center telecommunication service, my usual charges for the day would be around $10. but again on the first day of the month, I am getting charged around $150. This is way more than what it charges on regular days.
I don't get this behavior. is it done on purpose or am I missing something?
Thanks.
This is Tax, which appears on the very first day of each month.
According to this page, the DynamoDB is always free for 25 RCU and 25 WCU with 25GB of storage.
However, in the capacity tab of a table, it shows me an estimate cost for 10 RCU and 10 WCU to be $5.81 / month.
Will I be charged or not charged for this amount?
The estimation that you see within the DynamoDB page is not directly related to the billing calculation, therefore it will not take free tier into account. It is a simple calculator that calculates the AWS charge based on the configuration that you provided for DynamoDB.
Free tier calculations and deductions are applied at time of billing, as long as you are equal to or less than the usage for a free tier service you will not be billed for it. If you exceed this you will either be fully charged (in the case of EC2) or will pay the difference (as is the case in DynamoDB).
In DynamoDBs case this an accumulative deduction across all regions and tables, and if your account if part of an organization across all billed accounts under the organization.
They say there is always a monthly fee. But does it get charged as the month progresses or right away?
Amazon WorkSpaces is charged on a pro-rata basis.
So, if you run one for one week and turn it off, you'll be charged approximately one quarter of the monthly fee.
The first month's fee and the last month's fee are similarly pro-rated, based upon how long the WorkSpace was running.
There is no charge for creating a WorkSpace.
I would like to create an auto scaling group [ASG] with spot instances as a supplement to my production ASG that has on-demand instances.
After reading this post , I would like to mimic it but maybe without the dynamic bidding and auto-scaling at first.
So, I was wondering , if I just set the bid price for the spot instances equal to the on-demand price of the instance I use [m3.large] , would I be paying the lowest bid price [for that instance, in the region, in that hour] ??
So if the spot price is $0.01 and my max bid in the ASG is $0.067 , I would be paying $0.01 for every spot instance the ASG launches, right??
[That is what I understood from the AWS docs but I would like a confirmation]
And finally, by this kind of thinking [that I suppose I am not the only one to have] do I drive the spot instance prices up ?
Spot instance prices are different among Availability Zones. You pay based on the market price (M) where your instances are in.
You set a bid price (B). The payment rule is:
If the market price is lower than or equal your bid (M <= B) and there are free resources on spot-instances pool, you pay the market price.
If the market price is higher than your bid (M > B, outbid) or there are no free resources on spot-instances pool, your spot-instances will be terminated and no need to pay the current partial hour. You only need to pay the full hours that you used.
If you run spot-instances less than an hour and you terminate it by yourself, you need to pay for that partial hour.
Your bid is only the maximum amount that you want to pay for the spot-instances.
For more detail, you can see the official documentation. The market prices are driven by availability free resources and market demand.
For a bidding strategy, you can consult Spot Bid Advisor. In that page, you can find which instance types has lower outbid frequency based on the bid.
Added as per #Michael's suggestion on comment below:
The current market price is the lowest bid among all currently running instances. As an illustration, let say there were 3 instances were available (maybe more in the real case), and there were 4 bids for $1, $2, $3, and $5, the three highest bidders would have running instances, and the market price would be $2. So, your bid price may drive the spot instance indirectly if you are a winning bidder and there are more open bids than available spot instances.