criterion performance measurements
overview
want to understand this report?
Environment: AMD Opteron 6168 on Ubuntu 12.04 with GHC 8.0.2
| create | |
| restrict | |
| project | |
| union | |
| join | |
| group | |
| write |
create/relation 10x100
580 585 590 595 600
mean |
100 150 200 250 300 350 50 iters 100 150 200 250 0 s 50 ms
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 599 μs | 599 μs | 600 μs |
| R² goodness-of-fit | 1.000 | 1.000 | 1.000 |
| Mean execution time | 589 μs | 591 μs | 592 μs |
| Standard deviation | 3.79 μs | 4.50 μs | 5.81 μs |
Outlying measurements have slight (1.2%) effect on estimated standard deviation.
create/relation 10x1000
5.68 5.7 5.73 5.75 5.78 5.80 5.83
mean |
20 30 40 10 iters 100 150 200 250 300 0 s 50 ms
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 5.82 ms | 5.82 ms | 5.83 ms |
| R² goodness-of-fit | 1.000 | 1.000 | 1.000 |
| Mean execution time | 5.76 ms | 5.77 ms | 5.78 ms |
| Standard deviation | 22.9 μs | 28.8 μs | 39.1 μs |
Outlying measurements have slight (2.6%) effect on estimated standard deviation.
create/relation 10x10000
57.5 57.6 57.7 57.8 57.9 58.0 58.1
mean |
4 6 8 10 12 2 iters 200 300 400 500 600 700 800 0 s 100 ms
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 58.0 ms | 58.1 ms | 58.2 ms |
| R² goodness-of-fit | 1.000 | 1.000 | 1.000 |
| Mean execution time | 57.7 ms | 57.8 ms | 57.9 ms |
| Standard deviation | 104 μs | 171 μs | 274 μs |
Outlying measurements have slight (7.6%) effect on estimated standard deviation.
restrict/restrict relation 10x10000 to 10x1
14 15 16 14.5 15.5
mean |
10 15 20 25 5 iters 100 150 200 250 300 350 400 0 s 50 ms
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 14.3 ms | 14.5 ms | 14.7 ms |
| R² goodness-of-fit | 0.998 | 0.999 | 0.999 |
| Mean execution time | 14.3 ms | 14.5 ms | 14.8 ms |
| Standard deviation | 277 μs | 477 μs | 829 μs |
Outlying measurements have moderate (11.6%) effect on estimated standard deviation.
restrict/restrict relation 10x10000 to 10x5000
81 82 83 84 85 86
mean |
4 6 8 10 2 iters 400 600 800 0 s 200 ms 1 s
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 84.1 ms | 85.0 ms | 86.2 ms |
| R² goodness-of-fit | 0.999 | 1.000 | 1.000 |
| Mean execution time | 83.2 ms | 84.5 ms | 85.1 ms |
| Standard deviation | 612 μs | 1.40 ms | 2.30 ms |
Outlying measurements have slight (9.0%) effect on estimated standard deviation.
project/project 10x1000 to 1x1000
12 13 11.5 12.5
mean |
10 15 20 25 5 iters 100 150 200 250 300 350 400 0 s 50 ms
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 12.8 ms | 13.0 ms | 13.1 ms |
| R² goodness-of-fit | 0.999 | 0.999 | 1.000 |
| Mean execution time | 12.2 ms | 12.4 ms | 12.6 ms |
| Standard deviation | 302 μs | 460 μs | 596 μs |
Outlying measurements have moderate (14.3%) effect on estimated standard deviation.
project/project 10x1000 to 5x1000
15 16 14.5 14.8 15.3 15.5 15.8
mean |
10 15 20 5 iters 200 300 400 0 s 100 ms
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 15.8 ms | 16.0 ms | 16.2 ms |
| R² goodness-of-fit | 0.999 | 0.999 | 1.000 |
| Mean execution time | 15.3 ms | 15.5 ms | 15.6 ms |
| Standard deviation | 287 μs | 398 μs | 525 μs |
Outlying measurements have slight (4.2%) effect on estimated standard deviation.
union/union identical 10x1000
49 50 48.8 49.2 49.4 49.6 49.8
mean |
4 6 8 10 12 2 iters 200 300 400 500 600 700 0 s 100 ms
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 48.9 ms | 49.2 ms | 49.6 ms |
| R² goodness-of-fit | 1.000 | 1.000 | 1.000 |
| Mean execution time | 49.1 ms | 49.2 ms | 49.5 ms |
| Standard deviation | 210 μs | 328 μs | 498 μs |
Outlying measurements have slight (7.1%) effect on estimated standard deviation.
union/union identical 10x10000
515 515 516 516 517
mean |
1 2 2 3 3 4 4 0.5 iters 2 0 s 500 ms 1 s 1.5 2.5
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 509 ms | 516 ms | 521 ms |
| R² goodness-of-fit | 1.000 | 1.000 | 1.000 |
| Mean execution time | 514 ms | 516 ms | 516 ms |
| Standard deviation | 0 s | 1.19 ms | 1.29 ms |
Outlying measurements have moderate (18.7%) effect on estimated standard deviation.
join/join identical 10x100
336 336 337 337 338 338 339 339
mean |
1 2 2 3 3 4 4 0.5 iters 500 750 0 s 250 ms 1 s 1.25 1.5
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 341 ms | 343 ms | 346 ms |
| R² goodness-of-fit | 1.000 | 1.000 | 1.000 |
| Mean execution time | 336 ms | 337 ms | 339 ms |
| Standard deviation | 0 s | 1.70 ms | 1.87 ms |
Outlying measurements have moderate (18.8%) effect on estimated standard deviation.
group/group 10x100
46 47 48 45.5 46.5 47.5 48.5
mean |
4 6 8 10 12 2 iters 200 300 400 500 600 700 0 s 100 ms
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 46.4 ms | 47.0 ms | 47.4 ms |
| R² goodness-of-fit | 0.999 | 1.000 | 1.000 |
| Mean execution time | 46.7 ms | 47.0 ms | 47.5 ms |
| Standard deviation | 518 μs | 739 μs | 1.14 ms |
Outlying measurements have slight (7.1%) effect on estimated standard deviation.
write/write 10x1000
1.45 1.50 1.55 1.6 1.65
mean |
50 75 100 125 25 iters 100 150 200 250 0 s 50 ms
|
| lower bound | estimate | upper bound | |
|---|---|---|---|
| OLS regression | 1.51 ms | 1.56 ms | 1.62 ms |
| R² goodness-of-fit | 0.992 | 0.995 | 0.998 |
| Mean execution time | 1.55 ms | 1.57 ms | 1.59 ms |
| Standard deviation | 61.1 μs | 69.7 μs | 79.5 μs |
Outlying measurements have moderate (30.3%) effect on estimated standard deviation.
understanding this report
In this report, each function benchmarked by criterion is assigned a section of its own. The charts in each section are active; if you hover your mouse over data points and annotations, you will see more details.
- The chart on the left is a kernel density estimate (also known as a KDE) of time measurements. This graphs the probability of any given time measurement occurring. A spike indicates that a measurement of a particular time occurred; its height indicates how often that measurement was repeated.
- The chart on the right is the raw data from which the kernel density estimate is built. The x axis indicates the number of loop iterations, while the y axis shows measured execution time for the given number of loop iterations. The line behind the values is the linear regression prediction of execution time for a given number of iterations. Ideally, all measurements will be on (or very near) this line.
Under the charts is a small table. The first two rows are the results of a linear regression run on the measurements displayed in the right-hand chart.
- OLS regression indicates the time estimated for a single loop iteration using an ordinary least-squares regression model. This number is more accurate than the mean estimate below it, as it more effectively eliminates measurement overhead and other constant factors.
- R² goodness-of-fit is a measure of how accurately the linear regression model fits the observed measurements. If the measurements are not too noisy, R² should lie between 0.99 and 1, indicating an excellent fit. If the number is below 0.99, something is confounding the accuracy of the linear model.
- Mean execution time and standard deviation are statistics calculated from execution time divided by number of iterations.
We use a statistical technique called the bootstrap to provide confidence intervals on our estimates. The bootstrap-derived upper and lower bounds on estimates let you see how accurate we believe those estimates to be. (Hover the mouse over the table headers to see the confidence levels.)
A noisy benchmarking environment can cause some or many measurements to fall far from the mean. These outlying measurements can have a significant inflationary effect on the estimate of the standard deviation. We calculate and display an estimate of the extent to which the standard deviation has been inflated by outliers.