Skip to content

M
misc
Commit 63696046 authored by Elvis Pranskevichus's avatar Elvis Pranskevichus
Browse files
Options

Compute lower latency quartile, use numpy style of calculating quantiles

parent b77b5f81
Hide whitespace changes
Inline Side-by-side
Showing with 3 additions and 2 deletions
echo_client
View file @ 63696046
...@@ -96,7 +96,8 @@ if __name__ == '__main__': ...@@ -96,7 +96,8 @@ if __name__ == '__main__':
'quantiles should be in [0, 1]' 'quantiles should be in [0, 1]'
weighted_quantiles = np.cumsum(weights) - 0.5 * weights weighted_quantiles = np.cumsum(weights) - 0.5 * weights
weighted_quantiles /= np.sum(weights) weighted_quantiles -= weighted_quantiles[0]
weighted_quantiles /= weighted_quantiles[-1]
return np.interp(quantiles, weighted_quantiles, values) return np.interp(quantiles, weighted_quantiles, values)
...@@ -142,7 +143,7 @@ if __name__ == '__main__': ...@@ -142,7 +143,7 @@ if __name__ == '__main__':
latency_std = math.sqrt(variance) latency_std = math.sqrt(variance)
latency_cv = latency_std / mean_latency latency_cv = latency_std / mean_latency
percentiles = [50, 75, 90, 99] percentiles = [25, 50, 75, 90, 99]
percentile_data = [] percentile_data = []
quantiles = weighted_quantile(arange, [p / 100 for p in percentiles], quantiles = weighted_quantile(arange, [p / 100 for p in percentiles],
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7