colRowRanges

matrixStats: Benchmark report

---

colRanges() and rowRanges() benchmarks

This report benchmark the performance of colRanges() and rowRanges() against alternative methods.

Alternative methods

• apply() + range()

Data type "integer"

Data

> rmatrix <- function(nrow, ncol, mode = c("logical", "double", "integer", "index"), range = c(-100, 
+     +100), naProb = 0) {
+     mode <- match.arg(mode)
+     n <- nrow * ncol
+     if (mode == "logical") {
+         X <- sample(c(FALSE, TRUE), size = n, replace = TRUE)
+     }     else if (mode == "index") {
+         X <- seq_len(n)
+         mode <- "integer"
+     }     else {
+         X <- runif(n, min = range[1], max = range[2])
+     }
+     storage.mode(X) <- mode
+     if (naProb > 0) 
+         X[sample(n, size = naProb * n)] <- NA
+     dim(X) <- c(nrow, ncol)
+     X
+ }
> rmatrices <- function(scale = 10, seed = 1, ...) {
+     set.seed(seed)
+     data <- list()
+     data[[1]] <- rmatrix(nrow = scale * 1, ncol = scale * 1, ...)
+     data[[2]] <- rmatrix(nrow = scale * 10, ncol = scale * 10, ...)
+     data[[3]] <- rmatrix(nrow = scale * 100, ncol = scale * 1, ...)
+     data[[4]] <- t(data[[3]])
+     data[[5]] <- rmatrix(nrow = scale * 10, ncol = scale * 100, ...)
+     data[[6]] <- t(data[[5]])
+     names(data) <- sapply(data, FUN = function(x) paste(dim(x), collapse = "x"))
+     data
+ }
> data <- rmatrices(mode = mode)

Results

10x10 integer matrix

> X <- data[["10x10"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   805033 43.0    1442291  77.1  1442291  77.1
Vcells 12279862 93.7   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   803947 43.0    1442291  77.1  1442291  77.1
Vcells 12276762 93.7   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on integer+10x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.0046	0.0065	0.0154	0.0087	0.0117	0.323
apply+range	0.0997	0.1646	0.2359	0.1778	0.1879	5.933

expr	min	lq	mean	median	uq	max
colRanges	1.00	1.00	1.00	1.00	1	1.00
apply+range	21.58	25.14	15.32	20.53	16	18.37
Table: Benchmarking of rowRanges() and apply+range() on integer+10x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.0050	0.0067	0.0109	0.0090	0.0116	0.1394
apply+range	0.1471	0.1676	0.1784	0.1736	0.1809	0.4450

expr	min	lq	mean	median	uq	max
rowRanges	1.00	1.00	1.00	1.00	1.00	1.000
apply+range	29.38	24.88	16.31	19.19	15.66	3.193
Figure: Benchmarking of colRanges() and apply+range() on integer+10x10 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on integer+10x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	4.620	6.545	15.40	8.662	11.74	323.0
rowRanges	5.005	6.738	10.94	9.047	11.55	139.4

expr	min	lq	mean	median	uq	max
colRanges	1.000	1.00	1.0000	1.000	1.0000	1.0000
rowRanges	1.083	1.03	0.7105	1.044	0.9836	0.4315
Figure: Benchmarking of colRanges() and rowRanges() on integer+10x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x100 integer matrix

> X <- data[["100x100"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804053 43.0    1442291  77.1  1442291  77.1
Vcells 12277863 93.7   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804047 43.0    1442291  77.1  1442291  77.1
Vcells 12282906 93.8   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on integer+100x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.0343	0.0362	0.0496	0.0439	0.0549	0.1902
apply+range	0.7383	0.8309	1.0708	1.1279	1.2446	1.5987

expr	min	lq	mean	median	uq	max
colRanges	1.00	1.00	1.00	1.0	1.00	1.000
apply+range	21.55	22.96	21.58	25.7	22.69	8.407
Table: Benchmarking of rowRanges() and apply+range() on integer+100x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.0462	0.0489	0.2337	0.0662	0.078	16.71
apply+range	1.2561	1.3520	1.8275	1.4878	1.587	36.62

expr	min	lq	mean	median	uq	max
rowRanges	1.00	1.00	1.000	1.00	1.00	1.000
apply+range	27.19	27.65	7.821	22.47	20.35	2.191
Figure: Benchmarking of colRanges() and apply+range() on integer+100x100 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on integer+100x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	34.26	36.19	49.61	43.89	54.86	190.2
rowRanges	46.20	48.89	233.68	66.21	77.95	16712.8

expr	min	lq	mean	median	uq	max
colRanges	1.000	1.000	1.00	1.000	1.000	1.00
rowRanges	1.348	1.351	4.71	1.509	1.421	87.88
Figure: Benchmarking of colRanges() and rowRanges() on integer+100x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x10 integer matrix

> X <- data[["1000x10"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804088 43.0    1442291  77.1  1442291  77.1
Vcells 12277886 93.7   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804082 43.0    1442291  77.1  1442291  77.1
Vcells 12282929 93.8   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on integer+1000x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.0293	0.0427	0.0492	0.0475	0.0550	0.1001
apply+range	0.5220	0.5514	0.5881	0.5749	0.6163	0.8088

expr	min	lq	mean	median	uq	max
colRanges	1.00	1.00	1.00	1.00	1.0	1.000
apply+range	17.84	12.91	11.96	12.09	11.2	8.081
Table: Benchmarking of rowRanges() and apply+range() on integer+1000x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.0524	0.0577	0.0647	0.0616	0.0672	0.1193
apply+range	0.5078	0.5476	0.8775	0.5595	0.6082	29.6003

expr	min	lq	mean	median	uq	max
rowRanges	1.000	1.000	1.00	1.000	1.000	1
apply+range	9.698	9.483	13.56	9.084	9.054	248
Figure: Benchmarking of colRanges() and apply+range() on integer+1000x10 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on integer+1000x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	29.26	42.73	49.18	47.54	55.05	100.1
rowRanges	52.35	57.74	64.70	61.59	67.18	119.3

expr	min	lq	mean	median	uq	max
colRanges	1.000	1.000	1.000	1.000	1.00	1.000
rowRanges	1.789	1.351	1.315	1.296	1.22	1.192
Figure: Benchmarking of colRanges() and rowRanges() on integer+1000x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

10x1000 integer matrix

> X <- data[["10x1000"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804126 43.0    1442291  77.1  1442291  77.1
Vcells 12278463 93.7   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804120 43.0    1442291  77.1  1442291  77.1
Vcells 12283506 93.8   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on integer+10x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.0793	0.0828	0.0995	0.1055	0.1118	0.1509
apply+range	8.0571	8.5797	9.9144	8.9132	9.7031	41.1724

expr	min	lq	mean	median	uq	max
colRanges	1.0	1.0	1.00	1.0	1.00	1.0
apply+range	101.6	103.7	99.64	84.5	86.77	272.8
Table: Benchmarking of rowRanges() and apply+range() on integer+10x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.0758	0.0781	0.0929	0.0924	0.1051	0.1424
apply+range	8.1868	8.5968	9.5285	8.8701	9.3623	29.3173

expr	min	lq	mean	median	uq	max
rowRanges	1	1	1.0	1.00	1.00	1.0
apply+range	108	110	102.6	96.01	89.09	205.8
Figure: Benchmarking of colRanges() and apply+range() on integer+10x1000 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on integer+10x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowRanges	75.84	78.15	92.89	92.39	105.1	142.4
1	colRanges	79.30	82.77	99.50	105.48	111.8	150.9

	expr	min	lq	mean	median	uq	max
2	rowRanges	1.000	1.000	1.000	1.000	1.000	1.00
1	colRanges	1.046	1.059	1.071	1.142	1.064	1.06
Figure: Benchmarking of colRanges() and rowRanges() on integer+10x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x1000 integer matrix

> X <- data[["100x1000"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804161 43.0    1442291  77.1  1442291  77.1
Vcells 12278804 93.7   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804155 43.0    1442291  77.1  1442291  77.1
Vcells 12328847 94.1   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on integer+100x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.3261	0.3911	0.4213	0.4161	0.4379	1.043
apply+range	7.8823	11.5018	13.1821	12.7656	13.7377	27.096

expr	min	lq	mean	median	uq	max
colRanges	1.00	1.00	1.00	1.00	1.00	1.00
apply+range	24.17	29.41	31.29	30.68	31.37	25.98
Table: Benchmarking of rowRanges() and apply+range() on integer+100x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.3503	0.3838	0.4056	0.3903	0.4279	0.5563
apply+range	8.1934	9.2747	11.7391	11.4503	13.1059	22.4698

expr	min	lq	mean	median	uq	max
rowRanges	1.00	1.00	1.00	1.00	1.00	1.00
apply+range	23.39	24.17	28.94	29.33	30.63	40.39
Figure: Benchmarking of colRanges() and apply+range() on integer+100x1000 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on integer+100x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowRanges	350.3	383.8	405.6	390.3	427.9	556.3
1	colRanges	326.1	391.1	421.3	416.1	437.9	1042.8

	expr	min	lq	mean	median	uq	max
2	rowRanges	1.0000	1.000	1.000	1.000	1.000	1.000
1	colRanges	0.9308	1.019	1.039	1.066	1.023	1.875
Figure: Benchmarking of colRanges() and rowRanges() on integer+100x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x100 integer matrix

> X <- data[["1000x100"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804198 43.0    1442291  77.1  1442291  77.1
Vcells 12279211 93.7   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804192 43.0    1442291  77.1  1442291  77.1
Vcells 12329254 94.1   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on integer+1000x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.1963	0.2837	0.3215	0.2968	0.3318	1.917
apply+range	2.8171	4.1400	4.3449	4.2383	4.4934	5.725

expr	min	lq	mean	median	uq	max
colRanges	1.00	1.00	1.00	1.00	1.00	1.000
apply+range	14.35	14.59	13.52	14.28	13.54	2.986
Table: Benchmarking of rowRanges() and apply+range() on integer+1000x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.2633	0.3946	0.4344	0.4631	0.4947	0.5382
apply+range	2.7235	4.9189	5.2305	5.1759	5.5957	8.6934

expr	min	lq	mean	median	uq	max
rowRanges	1.00	1.00	1.00	1.00	1.00	1.00
apply+range	10.34	12.47	12.04	11.18	11.31	16.15
Figure: Benchmarking of colRanges() and apply+range() on integer+1000x100 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on integer+1000x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	196.3	283.7	321.5	296.8	331.8	1917.1
rowRanges	263.3	394.6	434.4	463.1	494.7	538.2

expr	min	lq	mean	median	uq	max
colRanges	1.000	1.000	1.000	1.00	1.000	1.0000
rowRanges	1.341	1.391	1.351	1.56	1.491	0.2807
Figure: Benchmarking of colRanges() and rowRanges() on integer+1000x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

Data type "double"

Data

> rmatrix <- function(nrow, ncol, mode = c("logical", "double", "integer", "index"), range = c(-100, 
+     +100), naProb = 0) {
+     mode <- match.arg(mode)
+     n <- nrow * ncol
+     if (mode == "logical") {
+         X <- sample(c(FALSE, TRUE), size = n, replace = TRUE)
+     }     else if (mode == "index") {
+         X <- seq_len(n)
+         mode <- "integer"
+     }     else {
+         X <- runif(n, min = range[1], max = range[2])
+     }
+     storage.mode(X) <- mode
+     if (naProb > 0) 
+         X[sample(n, size = naProb * n)] <- NA
+     dim(X) <- c(nrow, ncol)
+     X
+ }
> rmatrices <- function(scale = 10, seed = 1, ...) {
+     set.seed(seed)
+     data <- list()
+     data[[1]] <- rmatrix(nrow = scale * 1, ncol = scale * 1, ...)
+     data[[2]] <- rmatrix(nrow = scale * 10, ncol = scale * 10, ...)
+     data[[3]] <- rmatrix(nrow = scale * 100, ncol = scale * 1, ...)
+     data[[4]] <- t(data[[3]])
+     data[[5]] <- rmatrix(nrow = scale * 10, ncol = scale * 100, ...)
+     data[[6]] <- t(data[[5]])
+     names(data) <- sapply(data, FUN = function(x) paste(dim(x), collapse = "x"))
+     data
+ }
> data <- rmatrices(mode = mode)

Results

10x10 double matrix

> X <- data[["10x10"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804241 43.0    1442291  77.1  1442291  77.1
Vcells 12395358 94.6   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804226 43.0    1442291  77.1  1442291  77.1
Vcells 12395486 94.6   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on double+10x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.0031	0.0042	0.0062	0.0064	0.0073	0.0231
apply+range	0.1009	0.1047	0.1120	0.1068	0.1086	0.3445

expr	min	lq	mean	median	uq	max
colRanges	1.00	1.00	1.00	1.00	1.00	1.00
apply+range	32.74	24.72	18.13	16.82	14.84	14.92
Table: Benchmarking of rowRanges() and apply+range() on double+10x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.0038	0.0050	0.0068	0.0065	0.0081	0.0254
apply+range	0.1012	0.1055	0.1103	0.1074	0.1097	0.2749

expr	min	lq	mean	median	uq	max
rowRanges	1.0	1.00	1.00	1.00	1.00	1.00
apply+range	26.3	21.07	16.28	16.41	13.57	10.82
Figure: Benchmarking of colRanges() and apply+range() on double+10x10 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on double+10x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	3.081	4.236	6.179	6.353	7.316	23.10
rowRanges	3.850	5.005	6.776	6.545	8.085	25.41

expr	min	lq	mean	median	uq	max
colRanges	1.00	1.000	1.000	1.00	1.000	1.0
rowRanges	1.25	1.181	1.097	1.03	1.105	1.1
Figure: Benchmarking of colRanges() and rowRanges() on double+10x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x100 double matrix

> X <- data[["100x100"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804269 43.0    1442291  77.1  1442291  77.1
Vcells 12395369 94.6   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804263 43.0    1442291  77.1  1442291  77.1
Vcells 12405412 94.7   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on double+100x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.062	0.0839	0.0942	0.0947	0.1066	0.1278
apply+range	0.962	1.3296	1.3875	1.3885	1.4474	1.6245

expr	min	lq	mean	median	uq	max
colRanges	1.00	1.00	1.00	1.00	1.00	1.00
apply+range	15.52	15.84	14.72	14.66	13.57	12.71
Table: Benchmarking of rowRanges() and apply+range() on double+100x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.0554	0.0678	0.0915	0.0918	0.1151	0.229
apply+range	0.7784	0.9637	1.3737	1.4309	1.5354	5.086

expr	min	lq	mean	median	uq	max
rowRanges	1.00	1.00	1.00	1.00	1.00	1.00
apply+range	14.04	14.22	15.01	15.58	13.34	22.21
Figure: Benchmarking of colRanges() and apply+range() on double+100x100 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on double+100x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowRanges	55.43	67.75	91.50	91.81	115.1	229.0
1	colRanges	61.98	83.92	94.24	94.70	106.6	127.8

	expr	min	lq	mean	median	uq	max
2	rowRanges	1.000	1.000	1.00	1.000	1.0000	1.000
1	colRanges	1.118	1.239	1.03	1.031	0.9264	0.558
Figure: Benchmarking of colRanges() and rowRanges() on double+100x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x10 double matrix

> X <- data[["1000x10"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804304 43.0    1442291  77.1  1442291  77.1
Vcells 12396062 94.6   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804298 43.0    1442291  77.1  1442291  77.1
Vcells 12406105 94.7   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on double+1000x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.0531	0.0547	0.0660	0.0601	0.0751	0.1340
apply+range	0.3499	0.3563	0.4615	0.4310	0.5532	0.8573

expr	min	lq	mean	median	uq	max
colRanges	1.000	1.000	1.00	1.000	1.000	1.000
apply+range	6.587	6.518	6.99	7.176	7.369	6.399
Table: Benchmarking of rowRanges() and apply+range() on double+1000x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.0489	0.0500	0.0538	0.0531	0.0550	0.1093
apply+range	0.3465	0.3561	0.3869	0.3607	0.4321	0.6125

expr	min	lq	mean	median	uq	max
rowRanges	1.000	1.000	1.000	1.00	1.00	1.000
apply+range	7.087	7.115	7.191	6.79	7.85	5.602
Figure: Benchmarking of colRanges() and apply+range() on double+1000x10 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on double+1000x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowRanges	48.89	50.05	53.81	53.12	55.05	109.3
1	colRanges	53.12	54.66	66.03	60.05	75.07	134.0

	expr	min	lq	mean	median	uq	max
2	rowRanges	1.000	1.000	1.000	1.00	1.000	1.000
1	colRanges	1.087	1.092	1.227	1.13	1.364	1.225
Figure: Benchmarking of colRanges() and rowRanges() on double+1000x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

10x1000 double matrix

> X <- data[["10x1000"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804342 43.0    1442291  77.1  1442291  77.1
Vcells 12396088 94.6   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804336 43.0    1442291  77.1  1442291  77.1
Vcells 12406131 94.7   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on double+10x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.1001	0.1272	0.1426	0.1345	0.1538	0.3996
apply+range	5.1757	6.7754	8.0090	8.5032	8.7750	12.9806

expr	min	lq	mean	median	uq	max
colRanges	1.00	1.00	1.00	1.0	1.00	1.00
apply+range	51.71	53.25	56.15	63.2	57.06	32.49
Table: Benchmarking of rowRanges() and apply+range() on double+10x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.0897	0.1172	0.1255	0.1222	0.1444	0.1898
apply+range	5.1876	5.6340	8.0227	8.3831	8.9481	13.6628

expr	min	lq	mean	median	uq	max
rowRanges	1.00	1.00	1.00	1.00	1.00	1.00
apply+range	57.84	48.06	63.91	68.59	61.98	71.99
Figure: Benchmarking of colRanges() and apply+range() on double+10x1000 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on double+10x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowRanges	89.69	117.2	125.5	122.2	144.4	189.8
1	colRanges	100.09	127.2	142.6	134.5	153.8	399.6

	expr	min	lq	mean	median	uq	max
2	rowRanges	1.000	1.000	1.000	1.000	1.000	1.000
1	colRanges	1.116	1.085	1.136	1.101	1.065	2.106
Figure: Benchmarking of colRanges() and rowRanges() on double+10x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x1000 double matrix

> X <- data[["100x1000"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804377 43.0    1442291  77.1  1442291  77.1
Vcells 12397008 94.6   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804371 43.0    1442291  77.1  1442291  77.1
Vcells 12497051 95.4   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on double+100x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.5978	0.6113	0.7082	0.6665	0.8003	0.8993
apply+range	8.2700	8.8984	11.9058	11.4655	13.4822	27.7263

expr	min	lq	mean	median	uq	max
colRanges	1.00	1.00	1.00	1.0	1.00	1.00
apply+range	13.83	14.56	16.81	17.2	16.85	30.83
Table: Benchmarking of rowRanges() and apply+range() on double+100x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.5228	0.5324	0.6375	0.5717	0.7622	0.9431
apply+range	9.1326	9.9736	13.3663	10.5452	13.2749	152.8659

expr	min	lq	mean	median	uq	max
rowRanges	1.00	1.00	1.00	1.00	1.00	1.0
apply+range	17.47	18.73	20.97	18.45	17.42	162.1
Figure: Benchmarking of colRanges() and apply+range() on double+100x1000 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on double+100x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowRanges	522.8	532.4	637.5	571.7	762.2	943.1
1	colRanges	597.8	611.3	708.2	666.5	800.3	899.3

	expr	min	lq	mean	median	uq	max
2	rowRanges	1.000	1.000	1.000	1.000	1.00	1.0000
1	colRanges	1.144	1.148	1.111	1.166	1.05	0.9535
Figure: Benchmarking of colRanges() and rowRanges() on double+100x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x100 double matrix

> X <- data[["1000x100"]]
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804414 43.0    1442291  77.1  1442291  77.1
Vcells 12397033 94.6   35610798 271.7 68120027 519.8
> colStats <- microbenchmark(colRanges = colRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 2L, 
+     FUN = range, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
           used (Mb) gc trigger  (Mb) max used  (Mb)
Ncells   804408 43.0    1442291  77.1  1442291  77.1
Vcells 12497076 95.4   35610798 271.7 68120027 519.8
> rowStats <- microbenchmark(rowRanges = rowRanges(X, na.rm = FALSE), `apply+range` = apply(X, MARGIN = 1L, 
+     FUN = range, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanges() and apply+range() on double+1000x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	0.5147	0.5653	0.6685	0.7014	0.732	0.8881
apply+range	2.9969	3.9421	4.9898	5.2325	5.495	16.4171

expr	min	lq	mean	median	uq	max
colRanges	1.000	1.000	1.000	1.00	1.000	1.00
apply+range	5.823	6.973	7.465	7.46	7.507	18.49
Table: Benchmarking of rowRanges() and apply+range() on double+1000x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
rowRanges	0.4531	0.6492	0.7459	0.7982	0.8203	1.424
apply+range	3.0323	4.9026	5.6746	5.9891	6.3460	12.859

expr	min	lq	mean	median	uq	max
rowRanges	1.000	1.000	1.000	1.000	1.000	1.000
apply+range	6.692	7.551	7.608	7.503	7.736	9.028
Figure: Benchmarking of colRanges() and apply+range() on double+1000x100 data as well as rowRanges() and apply+range() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanges() and rowRanges() on double+1000x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr	min	lq	mean	median	uq	max
colRanges	514.7	565.3	668.5	701.4	732.0	888.1
rowRanges	453.1	649.2	745.9	798.2	820.3	1424.3

expr	min	lq	mean	median	uq	max
colRanges	1.0000	1.000	1.000	1.000	1.000	1.000
rowRanges	0.8803	1.149	1.116	1.138	1.121	1.604
Figure: Benchmarking of colRanges() and rowRanges() on double+1000x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

Appendix

Session information

R Under development (unstable) (2015-02-27 r67909)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1

locale:
[1] LC_COLLATE=English_United States.1252 
[2] LC_CTYPE=English_United States.1252   
[3] LC_MONETARY=English_United States.1252
[4] LC_NUMERIC=C                          
[5] LC_TIME=English_United States.1252    

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] markdown_0.7.7          microbenchmark_1.4-2    matrixStats_0.14.0-9000
[4] ggplot2_1.0.0           knitr_1.9.3             R.devices_2.13.0       
[7] R.utils_2.0.0           R.oo_1.19.0             R.methodsS3_1.7.0      

loaded via a namespace (and not attached):
 [1] Rcpp_0.11.4         BiocGenerics_0.13.6 splines_3.2.0      
 [4] MASS_7.3-39         munsell_0.4.2       lattice_0.20-30    
 [7] colorspace_1.2-4    R.cache_0.11.1-9000 multcomp_1.3-9     
[10] stringr_0.6.2       plyr_1.8.1          tools_3.2.0        
[13] parallel_3.2.0      grid_3.2.0          Biobase_2.27.2     
[16] gtable_0.1.2        TH.data_1.0-6       survival_2.38-1    
[19] digest_0.6.8        R.rsp_0.20.0        reshape2_1.4.1     
[22] formatR_1.0.3       base64enc_0.1-3     mime_0.2.1         
[25] evaluate_0.5.7      labeling_0.3        sandwich_2.3-2     
[28] scales_0.2.4        mvtnorm_1.0-2       zoo_1.7-12         
[31] Cairo_1.5-6         proto_0.3-10       

Total processing time was 48.14 secs.

Reproducibility

To reproduce this report, do:

html <- matrixStats:::benchmark('colRanges')

---

Copyright Henrik Bengtsson. Last updated on 2015-03-02 17:19:22 (-0800 UTC). Powered by RSP.

<script> var link = document.createElement('link'); link.rel = 'icon'; link.href = "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAAADFBMVEX9/v0AAP/9/v3//wBEQjoBAAAABHRSTlP//wD//gy7CwAAAGJJREFUOI3N0rESwCAIA9Ag///PXdoiBk0HhmbNO49DMETQCexNCSyFgdlGoO5DYOr9ThLgPosA7osIQP0sHuDOog8UI/ALa988wzdwXJRctf4s+d36YPTJ6aMd8ux3+QO4ABTtB85yDAh9AAAAAElFTkSuQmCC" document.getElementsByTagName('head')[0].appendChild(link); </script>