0b47699705
The current design of the packages is for ergonomics, and it works well
at this. However, it is not the most performant. When a user passes a
slice of attributes it will use
[`metric.WithAttributes`](cf787f3e3a/metric/instrument.go (L372-L376)):
```go
func WithAttributes(attributes ...attribute.KeyValue) MeasurementOption {
cp := make([]attribute.KeyValue, len(attributes))
copy(cp, attributes)
return attrOpt{set: attribute.NewSet(cp...)}
}
```
This will copy the passed attributes and then pass that copy to
`attribute.NewSet` which adds its own allocation.
If a user is performance conscious and already have a set, allow them to
pass it directly and reduce the required allocations for the measurement
call.
## Alternatives Considered
### Why not allow users to just use `Inst()` method?
With the current design a user can still just call:
```go
counter.Inst().Add(ctx, val, metric.WithAttributeSet(set))
```
However, the option slice (in this case `[]metric.AddOption`) is
allocated on the call. Meaning a user will also need to recreate the
pooling that our semconv generated packages already handle.
Providing the `*Set` methods is convenient, but it also helps the
application performance as one larger pool will be less strain on the GC
than many smaller ones.
### Why not just call `WithAttributeSet` in the existing methods?
Instead of appending a `WithAttributes` option internal to all the
measurement methods, we could also just create a `Set` and pass that
with `WithAttributeSet`. This will avoid the additional slice copy that
`WithAttributes` does on top of calling `attribute.NewSet`.
However, this copy that `WithAttributes` does is important as calling
`attribute.NewSet` will sort the original slice. Bypassing that will
mean all the passed attribute slices will be modified when used.
This is not great as it is likely going to be unexpected behavior by the
user, and is the reason we copy the slice in `WithAttributes` in the
first place.
### Why not just copy user attributes to an amortized slice pool to
create a new `attribute.Set`?
The additional creation of an `[]attribute.KeyValue` that user
attributes are copied to in order to prevent modification can be
amortized with a pool. For example, #7249. There shouldn't be any
difference than between calling a `*Set` method introduced here vs the
non-`Set` method.
This is true, and motivates this type of change. However, also providing
these additional methods allows the user additional performance gains if
the set is used for more than one measurement. For example:
```go
set := attributes.NewSet(attrs)
counter1.AddSet(ctx, 1, set)
counter2.AddSet(ctx, 2, set) // No additional set is created here.
```
Without these methods:
```go
counter1.Add(ctx, 1, attrs...) // A set is created for attrs
counter2.Add(ctx, 2, attrs...) // Another set is created for attrs
```
Each `attribute.Set` created will require an allocation (i.e. the
internal `any` field needs to be allocated to the heap). Meaning without
these methods a user will still not be able to write the most performant
code.
The attribute pool approach has merit, and should be pursued. However,
it does not invalidate the value of these changes.
OpenTelemetry-Go
OpenTelemetry-Go is the Go implementation of OpenTelemetry. It provides a set of APIs to directly measure performance and behavior of your software and send this data to observability platforms.
Project Status
| Signal | Status |
|---|---|
| Traces | Stable |
| Metrics | Stable |
| Logs | Beta1 |
Progress and status specific to this repository is tracked in our project boards and milestones.
Project versioning information and stability guarantees can be found in the versioning documentation.
Compatibility
OpenTelemetry-Go ensures compatibility with the current supported versions of the Go language:
Each major Go release is supported until there are two newer major releases. For example, Go 1.5 was supported until the Go 1.7 release, and Go 1.6 was supported until the Go 1.8 release.
For versions of Go that are no longer supported upstream, opentelemetry-go will stop ensuring compatibility with these versions in the following manner:
- A minor release of opentelemetry-go will be made to add support for the new supported release of Go.
- The following minor release of opentelemetry-go will remove compatibility testing for the oldest (now archived upstream) version of Go. This, and future, releases of opentelemetry-go may include features only supported by the currently supported versions of Go.
Currently, this project supports the following environments.
| OS | Go Version | Architecture |
|---|---|---|
| Ubuntu | 1.25 | amd64 |
| Ubuntu | 1.24 | amd64 |
| Ubuntu | 1.23 | amd64 |
| Ubuntu | 1.25 | 386 |
| Ubuntu | 1.24 | 386 |
| Ubuntu | 1.23 | 386 |
| Ubuntu | 1.25 | arm64 |
| Ubuntu | 1.24 | arm64 |
| Ubuntu | 1.23 | arm64 |
| macOS 13 | 1.25 | amd64 |
| macOS 13 | 1.24 | amd64 |
| macOS 13 | 1.23 | amd64 |
| macOS | 1.25 | arm64 |
| macOS | 1.24 | arm64 |
| macOS | 1.23 | arm64 |
| Windows | 1.25 | amd64 |
| Windows | 1.24 | amd64 |
| Windows | 1.23 | amd64 |
| Windows | 1.25 | 386 |
| Windows | 1.24 | 386 |
| Windows | 1.23 | 386 |
While this project should work for other systems, no compatibility guarantees are made for those systems currently.
Getting Started
You can find a getting started guide on opentelemetry.io.
OpenTelemetry's goal is to provide a single set of APIs to capture distributed traces and metrics from your application and send them to an observability platform. This project allows you to do just that for applications written in Go. There are two steps to this process: instrument your application, and configure an exporter.
Instrumentation
To start capturing distributed traces and metric events from your application it first needs to be instrumented. The easiest way to do this is by using an instrumentation library for your code. Be sure to check out the officially supported instrumentation libraries.
If you need to extend the telemetry an instrumentation library provides or want to build your own instrumentation for your application directly you will need to use the Go otel package. The examples are a good way to see some practical uses of this process.
Export
Now that your application is instrumented to collect telemetry, it needs an export pipeline to send that telemetry to an observability platform.
All officially supported exporters for the OpenTelemetry project are contained in the exporters directory.
| Exporter | Logs | Metrics | Traces |
|---|---|---|---|
| OTLP | ✓ | ✓ | ✓ |
| Prometheus | ✓ | ||
| stdout | ✓ | ✓ | ✓ |
| Zipkin | ✓ |
Contributing
See the contributing documentation.