matrixStats.benchmarks

colRanks() and rowRanks() benchmarks on subsetted computation

This report benchmark the performance of colRanks() and rowRanks() on subsetted computation.

Data type “integer”

Data

> rmatrix <- function(nrow, ncol, mode = c("logical", "double", "integer", "index"), range = c(-100, 
+     +100), na_prob = 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 (na_prob > 0) 
+         x[sample(n, size = na_prob * 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"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5291033 282.6    8529671 455.6  8529671 455.6
Vcells 10408751  79.5   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5281804 282.1    8529671 455.6  8529671 455.6
Vcells 10378384  79.2   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
1	colRanks_X_S	0.010045	0.0104090	0.0128028	0.0106035	0.0107815	0.222945
2	colRanks(X, rows, cols)	0.010460	0.0108710	0.0110724	0.0109905	0.0111860	0.014716
3	colRanks(X[rows, cols])	0.011304	0.0116975	0.0119653	0.0118630	0.0120775	0.019480

	expr	min	lq	mean	median	uq	max
1	colRanks_X_S	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	colRanks(X, rows, cols)	1.041314	1.044385	0.8648400	1.036497	1.037518	0.0660073
3	colRanks(X[rows, cols])	1.125336	1.123787	0.9345809	1.118782	1.120206	0.0873758

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	0.004894	0.0051295	0.0053376	0.0052715	0.0054265	0.008029
2	rowRanks(X, cols, rows)	0.005237	0.0054905	0.0074400	0.0055780	0.0058300	0.178301
3	rowRanks(X[cols, rows])	0.005874	0.0062575	0.0064235	0.0063795	0.0065030	0.009244

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowRanks(X, cols, rows)	1.070086	1.070377	1.393895	1.058143	1.074357	22.207124
3	rowRanks(X[cols, rows])	1.200245	1.219904	1.203456	1.210187	1.198378	1.151326

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on integer+10x10 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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
2	rowRanks_X_S	4.894	5.1295	5.33757	5.2715	5.4265	8.029
1	colRanks_X_S	10.045	10.4090	12.80283	10.6035	10.7815	222.945

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.00000
1	colRanks_X_S	2.052513	2.029243	2.398625	2.011477	1.986824	27.76747

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on integer+10x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x100 integer matrix

> X <- data[["100x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5280515 282.1    8529671 455.6  8529671 455.6
Vcells 10047426  76.7   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5280491 282.1    8529671 455.6  8529671 455.6
Vcells 10052479  76.7   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
1	colRanks_X_S	0.150295	0.1586490	0.1835278	0.1793895	0.1988485	0.254387
2	colRanks(X, rows, cols)	0.152071	0.1623015	0.1849575	0.1818660	0.2003695	0.272985
3	colRanks(X[rows, cols])	0.158641	0.1688090	0.1939169	0.1884575	0.2098555	0.337548

	expr	min	lq	mean	median	uq	max
1	colRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colRanks(X, rows, cols)	1.011817	1.023022	1.007790	1.013805	1.007649	1.073109
3	colRanks(X[rows, cols])	1.055531	1.064041	1.056608	1.050549	1.055354	1.326907

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	0.141898	0.155831	0.1764304	0.1726100	0.1914505	0.247824
2	rowRanks(X, cols, rows)	0.143292	0.159793	0.1791030	0.1742860	0.1918850	0.289692
3	rowRanks(X[cols, rows])	0.150126	0.168037	0.1869155	0.1858995	0.2006625	0.262457

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowRanks(X, cols, rows)	1.009824	1.025425	1.015149	1.009710	1.002269	1.168942
3	rowRanks(X[cols, rows])	1.057985	1.078328	1.059429	1.076992	1.048117	1.059046

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on integer+100x100 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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
2	rowRanks_X_S	141.898	155.831	176.4304	172.6100	191.4505	247.824
1	colRanks_X_S	150.295	158.649	183.5278	179.3895	198.8485	254.387

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colRanks_X_S	1.059176	1.018084	1.040228	1.039276	1.038642	1.026482

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on integer+100x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x10 integer matrix

> X <- data[["1000x10"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5281240 282.1    8529671 455.6  8529671 455.6
Vcells 10051441  76.7   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5281234 282.1    8529671 455.6  8529671 455.6
Vcells 10056524  76.8   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
3	colRanks(X[rows, cols])	0.207244	0.2091410	0.2413682	0.2221680	0.2671150	0.407985
2	colRanks(X, rows, cols)	0.200697	0.2033600	0.2343624	0.2240790	0.2557120	0.343630
1	colRanks_X_S	0.198479	0.2046835	0.2382378	0.2296785	0.2592915	0.346835

	expr	min	lq	mean	median	uq	max
3	colRanks(X[rows, cols])	1.0000000	1.0000000	1.0000000	1.000000	1.0000000	1.0000000
2	colRanks(X, rows, cols)	0.9684092	0.9723584	0.9709745	1.008602	0.9573105	0.8422614
1	colRanks_X_S	0.9577069	0.9786866	0.9870303	1.033805	0.9707111	0.8501170

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	0.188780	0.1913855	0.2243023	0.2063585	0.2438725	0.320750
2	rowRanks(X, cols, rows)	0.191312	0.1939085	0.2310270	0.2260505	0.2544410	0.315402
3	rowRanks(X[cols, rows])	0.198135	0.2006595	0.2371962	0.2281725	0.2592030	0.401751

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowRanks(X, cols, rows)	1.013412	1.013183	1.029981	1.095426	1.043336	0.9833266
3	rowRanks(X[cols, rows])	1.049555	1.048457	1.057485	1.105709	1.062863	1.2525362

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on integer+1000x10 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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
2	rowRanks_X_S	188.780	191.3855	224.3023	206.3585	243.8725	320.750
1	colRanks_X_S	198.479	204.6835	238.2378	229.6785	259.2915	346.835

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colRanks_X_S	1.051377	1.069483	1.062128	1.113007	1.063226	1.081325

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on integer+1000x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

10x1000 integer matrix

> X <- data[["10x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5281463 282.1    8529671 455.6  8529671 455.6
Vcells 10052322  76.7   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5281439 282.1    8529671 455.6  8529671 455.6
Vcells 10057375  76.8   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
1	colRanks_X_S	0.095829	0.1036630	0.1162648	0.112670	0.1272660	0.205064
2	colRanks(X, rows, cols)	0.098671	0.1075995	0.1161861	0.113351	0.1206080	0.166406
3	colRanks(X[rows, cols])	0.108334	0.1184410	0.1279909	0.123494	0.1362925	0.168258

	expr	min	lq	mean	median	uq	max
1	colRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.0000000	1.0000000
2	colRanks(X, rows, cols)	1.029657	1.037974	0.999323	1.006044	0.9476844	0.8114832
3	colRanks(X[rows, cols])	1.130493	1.142558	1.100857	1.096068	1.0709262	0.8205146

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
2	rowRanks(X, cols, rows)	0.093307	0.0992185	0.1099324	0.1038515	0.116065	0.182439
1	rowRanks_X_S	0.091753	0.0977660	0.1083293	0.1045810	0.114656	0.154599
3	rowRanks(X[cols, rows])	0.099857	0.1075825	0.1185675	0.1136840	0.127422	0.184145

	expr	min	lq	mean	median	uq	max
2	rowRanks(X, cols, rows)	1.0000000	1.0000000	1.0000000	1.000000	1.0000000	1.000000
1	rowRanks_X_S	0.9833453	0.9853606	0.9854171	1.007024	0.9878603	0.847401
3	rowRanks(X[cols, rows])	1.0701984	1.0842988	1.0785488	1.094679	1.0978503	1.009351

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on integer+10x1000 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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	rowRanks_X_S	91.753	97.766	108.3293	104.581	114.656	154.599
1	colRanks_X_S	95.829	103.663	116.2648	112.670	127.266	205.064

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colRanks_X_S	1.044424	1.060318	1.073254	1.077347	1.109981	1.326425

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on integer+10x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x1000 integer matrix

> X <- data[["100x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5281673 282.1    8529671 455.6  8529671 455.6
Vcells 10074984  76.9   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5281649 282.1    8529671 455.6  8529671 455.6
Vcells 10125037  77.3   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
1	colRanks_X_S	1.460714	1.543330	1.614133	1.582775	1.710147	2.427528
2	colRanks(X, rows, cols)	1.463529	1.546407	1.611092	1.589112	1.666816	2.333369
3	colRanks(X[rows, cols])	1.525266	1.612488	1.695617	1.654181	1.812447	2.213348

	expr	min	lq	mean	median	uq	max
1	colRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colRanks(X, rows, cols)	1.001927	1.001994	0.998116	1.004003	0.974663	0.9612120
3	colRanks(X[rows, cols])	1.044192	1.044811	1.050482	1.045114	1.059820	0.9117703

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	1.418860	1.497006	1.563823	1.538423	1.555997	2.317449
2	rowRanks(X, cols, rows)	1.427341	1.507645	1.552392	1.547082	1.550842	2.095641
3	rowRanks(X[cols, rows])	1.494656	1.583282	1.629205	1.623316	1.631626	2.200605

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.000000	1.0000000	1.000000	1.0000000	1.0000000
2	rowRanks(X, cols, rows)	1.005977	1.007107	0.9926903	1.005629	0.9966873	0.9042879
3	rowRanks(X[cols, rows])	1.053420	1.057632	1.0418091	1.055182	1.0486052	0.9495808

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on integer+100x1000 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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	rowRanks_X_S	1.418860	1.497006	1.563823	1.538423	1.555997	2.317449
1	colRanks_X_S	1.460714	1.543330	1.614133	1.582775	1.710147	2.427528

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.00000	1.000000	1.0000
1	colRanks_X_S	1.029498	1.030944	1.032171	1.02883	1.099068	1.0475

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on integer+100x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x100 integer matrix

> X <- data[["1000x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5281886 282.1    8529671 455.6  8529671 455.6
Vcells 10075782  76.9   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5281862 282.1    8529671 455.6  8529671 455.6
Vcells 10125835  77.3   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
1	colRanks_X_S	1.899728	2.011998	2.070741	2.060049	2.069953	3.085677
2	colRanks(X, rows, cols)	1.910097	2.014720	2.081040	2.063831	2.070780	2.708117
3	colRanks(X[rows, cols])	1.961788	2.069022	2.104722	2.119623	2.130853	2.906828

	expr	min	lq	mean	median	uq	max
1	colRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colRanks(X, rows, cols)	1.005458	1.001353	1.004973	1.001836	1.000400	0.8776411
3	colRanks(X[rows, cols])	1.032668	1.028342	1.016410	1.028918	1.029421	0.9420390

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	1.865247	1.968962	2.009408	2.020777	2.026666	2.441963
2	rowRanks(X, cols, rows)	1.875188	1.977621	2.041434	2.029410	2.035922	3.091483
3	rowRanks(X[cols, rows])	1.938070	2.051200	2.091851	2.102823	2.117586	2.970129

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowRanks(X, cols, rows)	1.005330	1.004398	1.015938	1.004272	1.004567	1.265983
3	rowRanks(X[cols, rows])	1.039042	1.041767	1.041029	1.040601	1.044862	1.216287

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on integer+1000x100 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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
2	rowRanks_X_S	1.865247	1.968962	2.009408	2.020777	2.026666	2.441963
1	colRanks_X_S	1.899728	2.011998	2.070741	2.060049	2.069953	3.085677

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colRanks_X_S	1.018486	1.021857	1.030523	1.019434	1.021359	1.263605

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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), na_prob = 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 (na_prob > 0) 
+         x[sample(n, size = na_prob * 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"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5282103 282.1    8529671 455.6  8529671 455.6
Vcells 10166883  77.6   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5282070 282.1    8529671 455.6  8529671 455.6
Vcells 10167021  77.6   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
1	colRanks_X_S	0.009751	0.010362	0.0112964	0.0105985	0.0108070	0.062305
2	colRanks(X, rows, cols)	0.010260	0.010843	0.0111679	0.0111210	0.0113475	0.015962
3	colRanks(X[rows, cols])	0.010818	0.011675	0.0120948	0.0118465	0.0121015	0.027480

	expr	min	lq	mean	median	uq	max
1	colRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colRanks(X, rows, cols)	1.052200	1.046420	0.988630	1.049299	1.050014	0.2561913
3	colRanks(X[rows, cols])	1.109425	1.126713	1.070679	1.117752	1.119784	0.4410561

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	0.005247	0.0055745	0.0058052	0.0056865	0.0059020	0.009326
2	rowRanks(X, cols, rows)	0.005811	0.0059970	0.0065303	0.0061595	0.0063700	0.035639
3	rowRanks(X[cols, rows])	0.006347	0.0067040	0.0068870	0.0068480	0.0070055	0.009154

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowRanks(X, cols, rows)	1.107490	1.075792	1.124909	1.083179	1.079295	3.8214669
3	rowRanks(X[cols, rows])	1.209644	1.202619	1.186347	1.204256	1.186970	0.9815569

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on double+10x10 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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
2	rowRanks_X_S	5.247	5.5745	5.80520	5.6865	5.902	9.326
1	colRanks_X_S	9.751	10.3620	11.29637	10.5985	10.807	62.305

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.0000	1.000000	1.000000
1	colRanks_X_S	1.858395	1.858821	1.945905	1.8638	1.831074	6.680785

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on double+10x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x100 double matrix

> X <- data[["100x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5282301 282.2    8529671 455.6  8529671 455.6
Vcells 10172836  77.7   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5282277 282.2    8529671 455.6  8529671 455.6
Vcells 10182889  77.7   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
2	colRanks(X, rows, cols)	0.157793	0.1706415	0.1952590	0.188473	0.2122810	0.312737
1	colRanks_X_S	0.156409	0.1670390	0.1971071	0.193650	0.2153525	0.303292
3	colRanks(X[rows, cols])	0.167291	0.1815095	0.2084319	0.204399	0.2266885	0.359424

	expr	min	lq	mean	median	uq	max
2	colRanks(X, rows, cols)	1.000000	1.0000000	1.000000	1.000000	1.000000	1.0000000
1	colRanks_X_S	0.991229	0.9788885	1.009465	1.027468	1.014469	0.9697989
3	colRanks(X[rows, cols])	1.060193	1.0636891	1.067464	1.084500	1.067870	1.1492852

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	0.143959	0.1522625	0.1777158	0.1720695	0.1958270	0.259568
2	rowRanks(X, cols, rows)	0.144963	0.1565980	0.1780939	0.1732365	0.1907675	0.298066
3	rowRanks(X[cols, rows])	0.155006	0.1689710	0.1880969	0.1841210	0.2026895	0.260504

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.0000000	1.000000
2	rowRanks(X, cols, rows)	1.006974	1.028474	1.002128	1.006782	0.9741634	1.148316
3	rowRanks(X[cols, rows])	1.076737	1.109735	1.058414	1.070039	1.0350437	1.003606

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on double+100x100 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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	rowRanks_X_S	143.959	152.2625	177.7158	172.0695	195.8270	259.568
1	colRanks_X_S	156.409	167.0390	197.1071	193.6500	215.3525	303.292

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colRanks_X_S	1.086483	1.097046	1.109114	1.125417	1.099708	1.168449

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on double+100x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x10 double matrix

> X <- data[["1000x10"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5282501 282.2    8529671 455.6  8529671 455.6
Vcells 10174248  77.7   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5282477 282.2    8529671 455.6  8529671 455.6
Vcells 10184301  77.8   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
2	colRanks(X, rows, cols)	0.219703	0.222095	0.2588177	0.2431585	0.2774785	0.383786
3	colRanks(X[rows, cols])	0.228932	0.231082	0.2674533	0.2467775	0.2945105	0.453573
1	colRanks_X_S	0.216946	0.220667	0.2593508	0.2539770	0.2825490	0.389968

	expr	min	lq	mean	median	uq	max
2	colRanks(X, rows, cols)	1.0000000	1.0000000	1.000000	1.000000	1.000000	1.000000
3	colRanks(X[rows, cols])	1.0420067	1.0404647	1.033365	1.014883	1.061381	1.181838
1	colRanks_X_S	0.9874512	0.9935703	1.002060	1.044492	1.018274	1.016108

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	0.205128	0.2073270	0.2367801	0.2186765	0.2590685	0.354804
2	rowRanks(X, cols, rows)	0.208523	0.2103955	0.2431503	0.2312000	0.2647270	0.334157
3	rowRanks(X[cols, rows])	0.218719	0.2208370	0.2523473	0.2386930	0.2779100	0.424637

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowRanks(X, cols, rows)	1.016551	1.014800	1.026903	1.057270	1.021842	0.9418073
3	rowRanks(X[cols, rows])	1.066256	1.065163	1.065745	1.091535	1.072728	1.1968213

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on double+1000x10 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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	rowRanks_X_S	205.128	207.327	236.7801	218.6765	259.0685	354.804
1	colRanks_X_S	216.946	220.667	259.3508	253.9770	282.5490	389.968

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colRanks_X_S	1.057613	1.064343	1.095323	1.161428	1.090634	1.099108

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on double+1000x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

10x1000 double matrix

> X <- data[["10x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5282706 282.2    8529671 455.6  8529671 455.6
Vcells 10174384  77.7   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5282682 282.2    8529671 455.6  8529671 455.6
Vcells 10184437  77.8   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
2	colRanks(X, rows, cols)	0.107805	0.1171945	0.1271478	0.1237850	0.1329600	0.210252
1	colRanks_X_S	0.105226	0.1138685	0.1299248	0.1253850	0.1376795	0.216747
3	colRanks(X[rows, cols])	0.121109	0.1329515	0.1454440	0.1387485	0.1543040	0.242364

	expr	min	lq	mean	median	uq	max
2	colRanks(X, rows, cols)	1.0000000	1.0000000	1.000000	1.000000	1.000000	1.000000
1	colRanks_X_S	0.9760772	0.9716198	1.021841	1.012926	1.035496	1.030892
3	colRanks(X[rows, cols])	1.1234080	1.1344517	1.143897	1.120883	1.160529	1.152731

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	0.089888	0.1005880	0.1092712	0.1064815	0.1171740	0.146230
2	rowRanks(X, cols, rows)	0.094083	0.1013450	0.1107924	0.1070895	0.1142800	0.192828
3	rowRanks(X[cols, rows])	0.102016	0.1128115	0.1231601	0.1193580	0.1343425	0.160808

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.0000000	1.000000
2	rowRanks(X, cols, rows)	1.046669	1.007526	1.013921	1.005710	0.9753017	1.318662
3	rowRanks(X[cols, rows])	1.134923	1.121520	1.127104	1.120927	1.1465214	1.099692

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on double+10x1000 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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	rowRanks_X_S	89.888	100.5880	109.2712	106.4815	117.1740	146.230
1	colRanks_X_S	105.226	113.8685	129.9248	125.3850	137.6795	216.747

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000	1.000000
1	colRanks_X_S	1.170635	1.132029	1.189012	1.177528	1.175	1.482233

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on double+10x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x1000 double matrix

> X <- data[["100x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5282916 282.2    8529671 455.6  8529671 455.6
Vcells 10219850  78.0   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5282892 282.2    8529671 455.6  8529671 455.6
Vcells 10319903  78.8   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
1	colRanks_X_S	1.560653	1.686157	1.735688	1.698892	1.741967	2.577927
2	colRanks(X, rows, cols)	1.560967	1.690041	1.729855	1.705023	1.740768	2.494924
3	colRanks(X[rows, cols])	1.661799	1.793497	1.858614	1.804922	1.848437	2.300170

	expr	min	lq	mean	median	uq	max
1	colRanks_X_S	1.000000	1.000000	1.0000000	1.000000	1.0000000	1.0000000
2	colRanks(X, rows, cols)	1.000201	1.002303	0.9966395	1.003608	0.9993111	0.9678024
3	colRanks(X[rows, cols])	1.064810	1.063660	1.0708229	1.062412	1.0611205	0.8922557

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	1.495770	1.592583	1.652753	1.618496	1.649717	2.525829
2	rowRanks(X, cols, rows)	1.497414	1.619851	1.637196	1.627238	1.660529	2.308154
3	rowRanks(X[cols, rows])	1.621047	1.743577	1.780409	1.762746	1.787400	2.322907

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	rowRanks(X, cols, rows)	1.001099	1.017122	0.9905868	1.005402	1.006553	0.9138204
3	rowRanks(X[cols, rows])	1.083754	1.094811	1.0772382	1.089127	1.083458	0.9196612

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on double+100x1000 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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	rowRanks_X_S	1.495770	1.592583	1.652753	1.618496	1.649717	2.525829
1	colRanks_X_S	1.560653	1.686157	1.735688	1.698892	1.741967	2.577927

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colRanks_X_S	1.043378	1.058756	1.050179	1.049674	1.055919	1.020626

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on double+100x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x100 double matrix

> X <- data[["1000x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5283129 282.2    8529671 455.6  8529671 455.6
Vcells 10219994  78.0   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colRanks_X_S = colRanks(X_S, na.rm = FALSE), `colRanks(X, rows, cols)` = colRanks(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colRanks(X[rows, cols])` = colRanks(X[rows, cols], 
+     na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5283105 282.2    8529671 455.6  8529671 455.6
Vcells 10320047  78.8   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowRanks_X_S = rowRanks(X_S, na.rm = FALSE), `rowRanks(X, cols, rows)` = rowRanks(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowRanks(X[cols, rows])` = rowRanks(X[cols, rows], 
+     na.rm = FALSE), unit = "ms")

Table: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() 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
2	colRanks(X, rows, cols)	2.141629	2.321084	2.403352	2.369630	2.514019	2.984235
1	colRanks_X_S	2.140986	2.323867	2.414664	2.396863	2.509381	3.435969
3	colRanks(X[rows, cols])	2.223409	2.413083	2.533625	2.439485	2.738247	3.295286

	expr	min	lq	mean	median	uq	max
2	colRanks(X, rows, cols)	1.0000000	1.000000	1.000000	1.000000	1.0000000	1.000000
1	colRanks_X_S	0.9996998	1.001199	1.004707	1.011492	0.9981553	1.151374
3	colRanks(X[rows, cols])	1.0381859	1.039636	1.054205	1.029479	1.0891913	1.104231

Table: Benchmarking of rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() 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
1	rowRanks_X_S	2.076805	2.248443	2.266023	2.260276	2.309429	2.748058
2	rowRanks(X, cols, rows)	2.087850	2.229965	2.291291	2.261569	2.321666	3.471563
3	rowRanks(X[cols, rows])	2.191521	2.355176	2.388814	2.379115	2.432278	3.290447

	expr	min	lq	mean	median	uq	max
1	rowRanks_X_S	1.000000	1.0000000	1.000000	1.000000	1.000000	1.000000
2	rowRanks(X, cols, rows)	1.005318	0.9917819	1.011151	1.000572	1.005299	1.263279
3	rowRanks(X[cols, rows])	1.055237	1.0474695	1.054188	1.052577	1.053195	1.197372

Figure: Benchmarking of colRanks_X_S(), colRanks(X, rows, cols)() and colRanks(X[rows, cols])() on double+1000x100 data as well as rowRanks_X_S(), rowRanks(X, cols, rows)() and rowRanks(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colRanks_X_S() and rowRanks_X_S() 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
2	rowRanks_X_S	2.076805	2.248443	2.266023	2.260276	2.309429	2.748058
1	colRanks_X_S	2.140986	2.323867	2.414664	2.396863	2.509381	3.435969

	expr	min	lq	mean	median	uq	max
2	rowRanks_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colRanks_X_S	1.030904	1.033545	1.065596	1.060429	1.086581	1.250326

Figure: Benchmarking of colRanks_X_S() and rowRanks_X_S() on double+1000x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

Appendix

Session information

R version 4.1.1 Patched (2021-08-10 r80727)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.5 LTS

Matrix products: default
BLAS:   /home/hb/software/R-devel/R-4-1-branch/lib/R/lib/libRblas.so
LAPACK: /home/hb/software/R-devel/R-4-1-branch/lib/R/lib/libRlapack.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

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

other attached packages:
[1] microbenchmark_1.4-7   matrixStats_0.60.1     ggplot2_3.3.5         
[4] knitr_1.33             R.devices_2.17.0       R.utils_2.10.1        
[7] R.oo_1.24.0            R.methodsS3_1.8.1-9001 history_0.0.1-9000    

loaded via a namespace (and not attached):
 [1] Biobase_2.52.0          httr_1.4.2              splines_4.1.1          
 [4] bit64_4.0.5             network_1.17.1          assertthat_0.2.1       
 [7] highr_0.9               stats4_4.1.1            blob_1.2.2             
[10] GenomeInfoDbData_1.2.6  robustbase_0.93-8       pillar_1.6.2           
[13] RSQLite_2.2.8           lattice_0.20-44         glue_1.4.2             
[16] digest_0.6.27           XVector_0.32.0          colorspace_2.0-2       
[19] Matrix_1.3-4            XML_3.99-0.7            pkgconfig_2.0.3        
[22] zlibbioc_1.38.0         genefilter_1.74.0       purrr_0.3.4            
[25] ergm_4.1.2              xtable_1.8-4            scales_1.1.1           
[28] tibble_3.1.4            annotate_1.70.0         KEGGREST_1.32.0        
[31] farver_2.1.0            generics_0.1.0          IRanges_2.26.0         
[34] ellipsis_0.3.2          cachem_1.0.6            withr_2.4.2            
[37] BiocGenerics_0.38.0     mime_0.11               survival_3.2-13        
[40] magrittr_2.0.1          crayon_1.4.1            statnet.common_4.5.0   
[43] memoise_2.0.0           laeken_0.5.1            fansi_0.5.0            
[46] R.cache_0.15.0          MASS_7.3-54             R.rsp_0.44.0           
[49] progressr_0.8.0         tools_4.1.1             lifecycle_1.0.0        
[52] S4Vectors_0.30.0        trust_0.1-8             munsell_0.5.0          
[55] tabby_0.0.1-9001        AnnotationDbi_1.54.1    Biostrings_2.60.2      
[58] compiler_4.1.1          GenomeInfoDb_1.28.1     rlang_0.4.11           
[61] grid_4.1.1              RCurl_1.98-1.4          cwhmisc_6.6            
[64] rappdirs_0.3.3          startup_0.15.0          labeling_0.4.2         
[67] bitops_1.0-7            base64enc_0.1-3         boot_1.3-28            
[70] gtable_0.3.0            DBI_1.1.1               markdown_1.1           
[73] R6_2.5.1                lpSolveAPI_5.5.2.0-17.7 rle_0.9.2              
[76] dplyr_1.0.7             fastmap_1.1.0           bit_4.0.4              
[79] utf8_1.2.2              parallel_4.1.1          Rcpp_1.0.7             
[82] vctrs_0.3.8             png_0.1-7               DEoptimR_1.0-9         
[85] tidyselect_1.1.1        xfun_0.25               coda_0.19-4            

Total processing time was 27.94 secs.

Reproducibility

To reproduce this report, do:

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