Add performance tests to AWS SDK Extension

[NathanielRN]

Description

Follow up to the Core repo PR which adds a way to add performance tests: open-telemetry/opentelemetry-python#1443

Type of change

Please delete options that are not relevant.

• New feature (non-breaking change which adds functionality)

How Has This Been Tested?

Running tox -e test-sdkextension-aws locally I can produce the following results:

test-sdkextension-aws run-test: commands[0] | pytest
================================================================================================================================================= test session starts =================================================================================================================================================
platform darwin -- Python 3.8.2, pytest-6.1.2, py-1.9.0, pluggy-0.13.1 -- /Users/enowell/git/opentelemetry-python-contrib/.tox/test-sdkextension-aws/bin/python
cachedir: .tox/test-sdkextension-aws/.pytest_cache
benchmark: 3.2.3 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: /Users/enowell/git/opentelemetry-python-contrib, configfile: pytest.ini
plugins: benchmark-3.2.3
collected 24 items

performance/benchmarks/trace/test_benchmark_aws_xray_ids_generator.py::test_generate_xray_trace_id PASSED                                                                                                                                                                                                       [  4%]
performance/benchmarks/trace/test_benchmark_aws_xray_ids_generator.py::test_generate_xray_span_id PASSED                                                                                                                                                                                                        [  8%]
performance/benchmarks/trace/propagation/test_benchmark_aws_xray_format.py::test_extract_single_header PASSED                                                                                                                                                                                                   [ 12%]
performance/benchmarks/trace/propagation/test_benchmark_aws_xray_format.py::test_inject_empty_context PASSED

...

-------------------------------------------------------------------------------------------------- benchmark: 4 tests -------------------------------------------------------------------------------------------------
Name (time in ns)                      Min                    Max                  Mean                StdDev                Median                 IQR            Outliers  OPS (Kops/s)            Rounds  Iterations
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
test_generate_xray_span_id        300.2500 (1.0)       4,729.1500 (1.0)        358.2244 (1.0)        117.4173 (1.0)        329.7750 (1.0)       66.6000 (1.42)    2505;2451    2,791.5467 (1.0)      105608          20
test_generate_xray_trace_id       576.0000 (1.92)     58,391.0000 (12.35)      765.5374 (2.14)       609.1449 (5.19)       739.0000 (2.24)      47.0000 (1.0)    1241;13007    1,306.2719 (0.47)     102533           1
test_inject_empty_context       1,379.0000 (4.59)     59,260.0000 (12.53)    1,899.8021 (5.30)     1,099.5063 (9.36)     1,872.0000 (5.68)     305.0000 (6.49)    1598;2308      526.3706 (0.19)      65433           1
test_extract_single_header      8,246.0000 (27.46)    29,191.0000 (6.17)     9,066.2906 (25.31)    2,448.8693 (20.86)    8,669.0000 (26.29)    301.5000 (6.41)          3;8      110.2987 (0.04)        117           1
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Legend:
  Outliers: 1 Standard Deviation from Mean; 1.5 IQR (InterQuartile Range) from 1st Quartile and 3rd Quartile.
  OPS: Operations Per Second, computed as 1 / Mean

Does This PR Require a Core Repo Change?

• No.

Checklist:

See contributing.md for styleguide, changelog guidelines, and more.

• Followed the style guidelines of this project
  - [ ] Changelogs have been updated
  - [ ] Unit tests have been added
• Documentation has been updated

[NathanielRN]

@codeboten @lzchen Would it be possible for us to get a gh-pages branch on the OpenTelemetry-Contrib repo? :)

[toumorokoshi]

Cool! interested to see this in action.

[toumorokoshi]

Suggested change

	Performance progression of benchmarks for packages distributed by OpenTelemetry Python can be viewed as a [graph of throughput vs commit history](https://open-telemetry.github.io/opentelemetry-python-contrib/benchmarks/index.html). From this page, you can download a JSON file with the performance results.
	Performance benchmarks for packages distributed by OpenTelemetry Python can be viewed as a [graph of throughput vs commit history](https://open-telemetry.github.io/opentelemetry-python-contrib/benchmarks/index.html). From this page, you can download a JSON file with the performance results.

Throughput I'm a bit confused about. What does that mean? number of iterations over some time period?

[toumorokoshi]

Suggesting against "progression" as I think that implies that we are trying to trend benchmarks one direction or another. I don't think we're consciously trying to improve performance, but we are trying to monitor it so it doesn't get out of control.

[NathanielRN]

From what I understand, pytest-benchmark runs the test as many times as it can in a second (this is configurable). So the graph this action outputs shows how many times it was able to run this function (in the mean case) per second as a measure of performance and shows how this metric changes between commits.

[NathanielRN]

@toumorokoshi That makes sense! What if we called it "history" though to confirm that this is with respects to commits (and not a load test or something that would have "time" on the x access)?

Suggested change

	Performance progression of benchmarks for packages distributed by OpenTelemetry Python can be viewed as a [graph of throughput vs commit history](https://open-telemetry.github.io/opentelemetry-python-contrib/benchmarks/index.html). From this page, you can download a JSON file with the performance results.
	Performance history of benchmarks for packages distributed by OpenTelemetry Python can be viewed as a [graph of throughput vs commit history](https://open-telemetry.github.io/opentelemetry-python-contrib/benchmarks/index.html). From this page, you can download a JSON file with the performance results.

[aabmass]

Same comment as core PR. Otherwise LGTM! 🚢

[codeboten]

Thanks for adding these, i've enabled the GH pages for this repo

[NathanielRN]

@codeboten Thanks for enabling gh-pages! Unfortunately I don’t think it will show up because we decided in open-telemetry/opentelemetry-python#1443 to just push to master because that’s where the read the docs build is... so I’ll work on adding a read the docs build that uses root on Contrib today so we can see the graphs.