HomePricingDemo
Sign inTry it free
Unlimited users, services, and hosts
$100
budget
=
1237GB
space
*
$0.08
per gb
BudgetGigabytesPrice per gbNumber of spansTimeseries
>= $550GB$0.080.1+ billions1000
>= $2003000GB$0.076.4+ billions40000
>= $5008000GB$0.0617+ billions100000
Predictable pricing
Uptrace bills for the amount of ingested gigabytes monthly or every $100, whichever happens first. You can cancel at any point.
Monthly budget
To limit your monthly bill, you can specify a budget which works as an upper limit on your bill. Uptrace never bills more than your budget.
Volume discounts
As your monthly usage grows, you automatically get a volume-based discount so you can continue to sample 100% of data.
No limits
Because our pricing is based solely on the amount of ingested data, you can have as many services, hosts, and users as you need.
Generous trial
On the free trial period you can ingest up to 1 terrabyte of data and 10000 timeseries. No credit card required.
Secure payments
Uptrace does NOT have direct access to your payment details and instead relies on Paddle to process payments.

Why does Uptrace bill for data size?

Pricing by data size is simple to understand and works equally well for spans, events (logs and errors), and metrics. An average span is about 500 bytes plus the size of custom attributes.

When compared with billing for the number of spans, such pricing is more fair for services that generate a lot of small spans and events, for example, micro-services that make a lot of RPC calls.

Do you support sampling?

To reduce the bill, you can configure Uptrace to sample a fraction of data and drop the rest.

The cost of sampling is $0.01 per dropped gigabyte, for example, 1000GB of dropped data will cost you $10. If you don't use sampling, you don't pay for it.

How do you bill for metrics?

The cost of metrics is $2 for 1000 timeseries. Uptrace stores 1 datapoint per minute for each timeseries and uses the number of active datapoints to calculate the bill.

Also, for every $5 spent on tracing or logs, you get 1000 timeseries for free. For example, if your tracing bill is $100, you get 20k timeseries for free.

A timeseries is a metric with an unique set of attributes/labels. For example, if you collect 10 metrics on each host and you have 10 hosts, that is 10*10=100 timeseries.

A datapoint is a single measured occurrence of a metric within a timeseries, consisting of a value and timestamp. A complete timeseries contains 43200 datapoints: 1 datapoint * 60 minutes * 24 hours * 30 days.

A histogram timeseries contains 129600 (3 * 43200) datapoints and therefore is more expensive than other timeseries like counters or gauges

For how long data is available?

Uptrace stores spans data (including logs and errors) for 30 days and metrics data for 90 days.

How do you bill?

Uptrace uses Paddle for billing and handling global tax compliance. Your billing details stay secure within Paddle and Uptrace doesn't have direct access to them.

Paddle acts like a reseller and can supply you with a receipt if you need one.

Do I need an account to try Uptrace?

No, you can see Uptrace in action using our public playground.

Do you have a free tier?

If you want to use Uptrace for free, you can install a forever free open source Uptrace version.

That said, you can ingest up to 50GB of data each month for free.

Do you offer an on-premise installation?

If you monthly usage exceeds 10 terabytes, you can install Uptrace on your own servers to reduce the bill or pass SOC2 audit.

Our on-premise offer includes:

  • Volume discounts with annual commitments
  • Custom data retention
  • Dedicated support engineer
  • Integration with Keycloak or similar software

Still have questions?

Send us an email and we will reply as soon as possible.