matrixStats.benchmarks

---

colMins() and rowMins() benchmarks on subsetted computation

This report benchmark the performance of colMins() and rowMins() 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  5255602 280.7    8529671 455.6  8529671 455.6
Vcells 10179002  77.7   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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  5246833 280.3    8529671 455.6  8529671 455.6
Vcells 10150363  77.5   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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	colMins_X_S	0.001824	0.0019145	0.0028551	0.0019550	0.0020350	0.088238
2	colMins(X, rows, cols)	0.002216	0.0023135	0.0024317	0.0023595	0.0024805	0.004936
3	colMins(X[rows, cols])	0.002652	0.0028795	0.0030817	0.0029765	0.0031300	0.007775

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colMins(X, rows, cols)	1.214912	1.208409	0.851675	1.206905	1.218919	0.0559396
3	colMins(X[rows, cols])	1.453947	1.504048	1.079362	1.522506	1.538083	0.0881140

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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	rowMins_X_S	0.001895	0.001980	0.0021116	0.0020505	0.0021580	0.004468
2	rowMins(X, cols, rows)	0.002313	0.002386	0.0033428	0.0024475	0.0025790	0.087173
3	rowMins(X[cols, rows])	0.002761	0.002978	0.0031251	0.0030610	0.0031565	0.005473

	expr	min	lq	mean	median	uq	max
1	rowMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowMins(X, cols, rows)	1.220581	1.205051	1.583119	1.193611	1.195088	19.510519
3	rowMins(X[cols, rows])	1.456992	1.504040	1.479982	1.492807	1.462697	1.224933

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	1.824	1.9145	2.85515	1.9550	2.035	88.238
2	rowMins_X_S	1.895	1.9800	2.11156	2.0505	2.158	4.468

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	rowMins_X_S	1.038925	1.034213	0.7395618	1.048849	1.060442	0.0506358

Figure: Benchmarking of colMins_X_S() and rowMins_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 5245449 280.2    8529671 455.6  8529671 455.6
Vcells 9819235  75.0   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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 5245425 280.2    8529671 455.6  8529671 455.6
Vcells 9824288  75.0   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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
2	colMins(X, rows, cols)	0.016246	0.0173075	0.0176170	0.0175725	0.0178660	0.021259
1	colMins_X_S	0.017303	0.0182985	0.0184890	0.0184670	0.0186440	0.023789
3	colMins(X[rows, cols])	0.028228	0.0295035	0.0304097	0.0298320	0.0301315	0.061707

	expr	min	lq	mean	median	uq	max
2	colMins(X, rows, cols)	1.000000	1.000000	1.00000	1.000000	1.000000	1.000000
1	colMins_X_S	1.065062	1.057258	1.04950	1.050903	1.043546	1.119008
3	colMins(X[rows, cols])	1.737535	1.704666	1.72616	1.697653	1.686527	2.902630

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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
2	rowMins(X, cols, rows)	0.020058	0.0213765	0.0225047	0.0222750	0.0228580	0.045919
1	rowMins_X_S	0.022125	0.0234000	0.0241292	0.0243645	0.0247175	0.030877
3	rowMins(X[cols, rows])	0.032824	0.0344555	0.0357612	0.0357735	0.0362945	0.050519

	expr	min	lq	mean	median	uq	max
2	rowMins(X, cols, rows)	1.000000	1.00000	1.000000	1.000000	1.000000	1.0000000
1	rowMins_X_S	1.103051	1.09466	1.072184	1.093805	1.081350	0.6724232
3	rowMins(X[cols, rows])	1.636454	1.61184	1.589050	1.605993	1.587825	1.1001764

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	17.303	18.2985	18.48900	18.4670	18.6440	23.789
2	rowMins_X_S	22.125	23.4000	24.12922	24.3645	24.7175	30.877

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.00000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowMins_X_S	1.27868	1.278793	1.305058	1.319353	1.325762	1.297953

Figure: Benchmarking of colMins_X_S() and rowMins_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 5246191 280.2    8529671 455.6  8529671 455.6
Vcells 9823276  75.0   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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 5246167 280.2    8529671 455.6  8529671 455.6
Vcells 9828329  75.0   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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
1	colMins_X_S	0.012624	0.0127975	0.0130295	0.0128785	0.0130165	0.022591
2	colMins(X, rows, cols)	0.013766	0.0141075	0.0145299	0.0143240	0.0146560	0.019234
3	colMins(X[rows, cols])	0.024394	0.0247165	0.0253511	0.0249745	0.0251640	0.050858

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colMins(X, rows, cols)	1.090463	1.102364	1.115148	1.112241	1.125956	0.851401
3	colMins(X[rows, cols])	1.932351	1.931354	1.945663	1.939240	1.933239	2.251250

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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	rowMins_X_S	0.021538	0.0224915	0.0231572	0.0227275	0.0237150	0.036367
2	rowMins(X, cols, rows)	0.020882	0.0220075	0.0226914	0.0228020	0.0232315	0.028191
3	rowMins(X[cols, rows])	0.033943	0.0355290	0.0369136	0.0360005	0.0374970	0.071707

	expr	min	lq	mean	median	uq	max
1	rowMins_X_S	1.0000000	1.0000000	1.0000000	1.000000	1.0000000	1.0000000
2	rowMins(X, cols, rows)	0.9695422	0.9784808	0.9798866	1.003278	0.9796121	0.7751808
3	rowMins(X[cols, rows])	1.5759588	1.5796634	1.5940471	1.584006	1.5811512	1.9717601

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	12.624	12.7975	13.02954	12.8785	13.0165	22.591
2	rowMins_X_S	21.538	22.4915	23.15717	22.7275	23.7150	36.367

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000
2	rowMins_X_S	1.706115	1.757492	1.777282	1.764763	1.821918	1.6098

Figure: Benchmarking of colMins_X_S() and rowMins_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 5246395 280.2    8529671 455.6  8529671 455.6
Vcells 9824092  75.0   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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 5246371 280.2    8529671 455.6  8529671 455.6
Vcells 9829145  75.0   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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	colMins_X_S	0.027145	0.030413	0.0311386	0.0310520	0.0317120	0.049130
2	colMins(X, rows, cols)	0.032023	0.034360	0.0360924	0.0356090	0.0374095	0.054955
3	colMins(X[rows, cols])	0.040437	0.043705	0.0449355	0.0447285	0.0458695	0.058580

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.00000	1.000000	1.000000	1.000000	1.000000
2	colMins(X, rows, cols)	1.179702	1.12978	1.159087	1.146754	1.179664	1.118563
3	colMins(X[rows, cols])	1.489667	1.43705	1.443078	1.440439	1.446440	1.192347

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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
1	rowMins_X_S	0.030415	0.0322585	0.0336089	0.0337605	0.0346190	0.050089
2	rowMins(X, cols, rows)	0.028900	0.0328595	0.0344084	0.0338655	0.0358275	0.063049
3	rowMins(X[cols, rows])	0.040093	0.0430085	0.0450824	0.0450740	0.0467100	0.058829

	expr	min	lq	mean	median	uq	max
1	rowMins_X_S	1.0000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowMins(X, cols, rows)	0.9501891	1.018631	1.023789	1.003110	1.034909	1.258739
3	rowMins(X[cols, rows])	1.3181983	1.333246	1.341383	1.335111	1.349259	1.174489

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	27.145	30.4130	31.13861	31.0520	31.712	49.130
2	rowMins_X_S	30.415	32.2585	33.60889	33.7605	34.619	50.089

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.00000
2	rowMins_X_S	1.120464	1.060681	1.079332	1.087225	1.091669	1.01952

Figure: Benchmarking of colMins_X_S() and rowMins_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 5246608 280.2    8529671 455.6  8529671 455.6
Vcells 9846763  75.2   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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 5246584 280.2    8529671 455.6  8529671 455.6
Vcells 9896816  75.6   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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
2	colMins(X, rows, cols)	0.098981	0.1108115	0.1220782	0.1182920	0.1350380	0.211816
1	colMins_X_S	0.104215	0.1201845	0.1315901	0.1284340	0.1429875	0.186393
3	colMins(X[rows, cols])	0.172079	0.1982065	0.2196798	0.2150395	0.2381535	0.284584

	expr	min	lq	mean	median	uq	max
2	colMins(X, rows, cols)	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colMins_X_S	1.052879	1.084585	1.077917	1.085737	1.058869	0.879976
3	colMins(X[rows, cols])	1.738505	1.788682	1.799501	1.817870	1.763604	1.343544

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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
2	rowMins(X, cols, rows)	0.119397	0.1233845	0.1430347	0.1374560	0.1559030	0.243470
1	rowMins_X_S	0.135333	0.1472975	0.1705212	0.1705720	0.1829455	0.239549
3	rowMins(X[cols, rows])	0.197957	0.2138430	0.2458982	0.2423385	0.2661420	0.328259

	expr	min	lq	mean	median	uq	max
2	rowMins(X, cols, rows)	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	rowMins_X_S	1.133471	1.193809	1.192167	1.240921	1.173457	0.9838953
3	rowMins(X[cols, rows])	1.657973	1.733143	1.719151	1.763026	1.707100	1.3482524

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	104.215	120.1845	131.5901	128.434	142.9875	186.393
2	rowMins_X_S	135.333	147.2975	170.5212	170.572	182.9455	239.549

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowMins_X_S	1.298594	1.225595	1.295851	1.328091	1.279451	1.285182

Figure: Benchmarking of colMins_X_S() and rowMins_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 5246818 280.3    8529671 455.6  8529671 455.6
Vcells 9847530  75.2   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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 5246794 280.3    8529671 455.6  8529671 455.6
Vcells 9897583  75.6   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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
2	colMins(X, rows, cols)	0.073723	0.0784890	0.0864956	0.0828675	0.0915205	0.112950
1	colMins_X_S	0.081442	0.0870510	0.0939272	0.0903125	0.1003715	0.133458
3	colMins(X[rows, cols])	0.155940	0.1666595	0.1831948	0.1729565	0.1990415	0.270186

	expr	min	lq	mean	median	uq	max
2	colMins(X, rows, cols)	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colMins_X_S	1.104703	1.109085	1.085918	1.089842	1.096711	1.181567
3	colMins(X[rows, cols])	2.115215	2.123349	2.117966	2.087145	2.174830	2.392085

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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
2	rowMins(X, cols, rows)	0.105034	0.1174560	0.1336602	0.132533	0.1419420	0.248245
1	rowMins_X_S	0.125060	0.1356705	0.1562093	0.153204	0.1693875	0.210504
3	rowMins(X[cols, rows])	0.191453	0.2076660	0.2399987	0.233307	0.2587460	0.332191

	expr	min	lq	mean	median	uq	max
2	rowMins(X, cols, rows)	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	rowMins_X_S	1.190662	1.155075	1.168705	1.155969	1.193357	0.8479687
3	rowMins(X[cols, rows])	1.822772	1.768032	1.795588	1.760369	1.822899	1.3381579

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	81.442	87.0510	93.92716	90.3125	100.3715	133.458
2	rowMins_X_S	125.060	135.6705	156.20930	153.2040	169.3875	210.504

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.00000	1.000000	1.000000	1.000000
2	rowMins_X_S	1.535571	1.558517	1.66309	1.696376	1.687606	1.577305

Figure: Benchmarking of colMins_X_S() and rowMins_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 5247036 280.3    8529671 455.6  8529671 455.6
Vcells 9938626  75.9   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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 5247003 280.3    8529671 455.6  8529671 455.6
Vcells 9938764  75.9   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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	colMins_X_S	0.001838	0.0019275	0.0021597	0.001980	0.0020700	0.016292
2	colMins(X, rows, cols)	0.002220	0.0023050	0.0024166	0.002356	0.0024585	0.004970
3	colMins(X[rows, cols])	0.002677	0.0029595	0.0031801	0.003045	0.0031770	0.008924

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colMins(X, rows, cols)	1.207835	1.195850	1.118950	1.189899	1.187681	0.3050577
3	colMins(X[rows, cols])	1.456474	1.535409	1.472427	1.537879	1.534783	0.5477535

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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	rowMins_X_S	0.001897	0.0019800	0.0020982	0.0020360	0.0021405	0.004637
2	rowMins(X, cols, rows)	0.002291	0.0023705	0.0026435	0.0024425	0.0025465	0.018619
3	rowMins(X[cols, rows])	0.002779	0.0030395	0.0033091	0.0031125	0.0032620	0.010612

	expr	min	lq	mean	median	uq	max
1	rowMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowMins(X, cols, rows)	1.207696	1.197222	1.259864	1.199656	1.189675	4.015312
3	rowMins(X[cols, rows])	1.464945	1.535101	1.577125	1.528733	1.523943	2.288549

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	1.838	1.9275	2.15974	1.980	2.0700	16.292
2	rowMins_X_S	1.897	1.9800	2.09821	2.036	2.1405	4.637

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.0000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	rowMins_X_S	1.0321	1.027237	0.9715105	1.028283	1.034058	0.2846182

Figure: Benchmarking of colMins_X_S() and rowMins_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 5247234 280.3    8529671 455.6  8529671 455.6
Vcells 9944566  75.9   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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 5247210 280.3    8529671 455.6  8529671 455.6
Vcells 9954619  76.0   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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	colMins(X, rows, cols)	0.017388	0.018709	0.0193960	0.0194430	0.0197540	0.026693
1	colMins_X_S	0.019686	0.020248	0.0209610	0.0209495	0.0212135	0.028075
3	colMins(X[rows, cols])	0.035459	0.036099	0.0376484	0.0373310	0.0376300	0.081881

	expr	min	lq	mean	median	uq	max
2	colMins(X, rows, cols)	1.00000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colMins_X_S	1.13216	1.082260	1.080689	1.077483	1.073884	1.051774
3	colMins(X[rows, cols])	2.03928	1.929499	1.941036	1.920023	1.904931	3.067508

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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	rowMins_X_S	0.019915	0.0211350	0.0216981	0.0215095	0.0221730	0.028498
2	rowMins(X, cols, rows)	0.020550	0.0214005	0.0225884	0.0219840	0.0226715	0.051804
3	rowMins(X[cols, rows])	0.035124	0.0369345	0.0378889	0.0373850	0.0385780	0.052682

	expr	min	lq	mean	median	uq	max
1	rowMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowMins(X, cols, rows)	1.031886	1.012562	1.041032	1.022060	1.022482	1.817812
3	rowMins(X[cols, rows])	1.763696	1.747551	1.746183	1.738069	1.739864	1.848621

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	19.686	20.248	20.96105	20.9495	21.2135	28.075
2	rowMins_X_S	19.915	21.135	21.69812	21.5095	22.1730	28.498

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowMins_X_S	1.011633	1.043807	1.035164	1.026731	1.045231	1.015067

Figure: Benchmarking of colMins_X_S() and rowMins_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 5247433 280.3    8529671 455.6  8529671 455.6
Vcells 9945961  75.9   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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 5247409 280.3    8529671 455.6  8529671 455.6
Vcells 9956014  76.0   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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
1	colMins_X_S	0.016357	0.0169535	0.0174170	0.0170480	0.0172185	0.031485
2	colMins(X, rows, cols)	0.017032	0.0179310	0.0181883	0.0182175	0.0183750	0.022566
3	colMins(X[rows, cols])	0.032399	0.0338045	0.0342687	0.0341165	0.0342590	0.062102

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colMins(X, rows, cols)	1.041267	1.057658	1.044284	1.068600	1.067166	0.7167222
3	colMins(X[rows, cols])	1.980742	1.993954	1.967550	2.001203	1.989662	1.9724313

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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	rowMins_X_S	0.017693	0.0186045	0.0193016	0.0190875	0.0197155	0.033773
2	rowMins(X, cols, rows)	0.020590	0.0218915	0.0228138	0.0227640	0.0234500	0.028774
3	rowMins(X[cols, rows])	0.035052	0.0368290	0.0391780	0.0381565	0.0397365	0.074442

	expr	min	lq	mean	median	uq	max
1	rowMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowMins(X, cols, rows)	1.163737	1.176678	1.181969	1.192613	1.189420	0.8519824
3	rowMins(X[cols, rows])	1.981122	1.979575	2.029785	1.999031	2.015495	2.2041868

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	16.357	16.9535	17.41696	17.0480	17.2185	31.485
2	rowMins_X_S	17.693	18.6045	19.30155	19.0875	19.7155	33.773

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowMins_X_S	1.081678	1.097384	1.108204	1.119633	1.145018	1.072669

Figure: Benchmarking of colMins_X_S() and rowMins_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 5247637 280.3    8529671 455.6  8529671 455.6
Vcells 9946096  75.9   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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 5247613 280.3    8529671 455.6  8529671 455.6
Vcells 9956149  76.0   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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
1	colMins_X_S	0.030396	0.0331165	0.0341850	0.0340105	0.0348855	0.057655
2	colMins(X, rows, cols)	0.029013	0.0336240	0.0360027	0.0352750	0.0379075	0.060231
3	colMins(X[rows, cols])	0.048585	0.0521060	0.0536512	0.0538440	0.0548405	0.066291

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.0000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colMins(X, rows, cols)	0.9545006	1.015325	1.053173	1.037180	1.086626	1.044680
3	colMins(X[rows, cols])	1.5984011	1.573415	1.569437	1.583158	1.572014	1.149787

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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	rowMins_X_S	0.029790	0.0319780	0.0329715	0.0326920	0.0340190	0.041739
2	rowMins(X, cols, rows)	0.026884	0.0339335	0.0361346	0.0358535	0.0377485	0.081865
3	rowMins(X[cols, rows])	0.044573	0.0481225	0.0497464	0.0498195	0.0512985	0.059541

	expr	min	lq	mean	median	uq	max
1	rowMins_X_S	1.0000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowMins(X, cols, rows)	0.9024505	1.061151	1.095932	1.096706	1.109630	1.961355
3	rowMins(X[cols, rows])	1.4962403	1.504863	1.508770	1.523905	1.507937	1.426508

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

Table: Benchmarking of colMins_X_S() and rowMins_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	rowMins_X_S	29.790	31.9780	32.97152	32.6920	34.0190	41.739
1	colMins_X_S	30.396	33.1165	34.18499	34.0105	34.8855	57.655

	expr	min	lq	mean	median	uq	max
2	rowMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colMins_X_S	1.020342	1.035603	1.036804	1.040331	1.025471	1.381322

Figure: Benchmarking of colMins_X_S() and rowMins_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 5247850 280.3    8529671 455.6  8529671 455.6
Vcells 9991541  76.3   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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  5247826 280.3    8529671 455.6  8529671 455.6
Vcells 10091594  77.0   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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	colMins_X_S	0.118586	0.132462	0.1521474	0.1495030	0.1661150	0.227501
2	colMins(X, rows, cols)	0.122183	0.132991	0.1518777	0.1500845	0.1606015	0.278352
3	colMins(X[rows, cols])	0.213240	0.239611	0.2759704	0.2702650	0.2984635	0.395115

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.0000000	1.000000
2	colMins(X, rows, cols)	1.030332	1.003994	0.998227	1.003890	0.9668091	1.223520
3	colMins(X[rows, cols])	1.798189	1.808904	1.813835	1.807756	1.7967282	1.736762

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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
2	rowMins(X, cols, rows)	0.132620	0.1370405	0.1587292	0.1517905	0.1722215	0.308469
1	rowMins_X_S	0.126679	0.1375235	0.1588114	0.1587335	0.1722160	0.245443
3	rowMins(X[cols, rows])	0.218846	0.2326025	0.2707474	0.2678340	0.2956165	0.385106

	expr	min	lq	mean	median	uq	max
2	rowMins(X, cols, rows)	1.0000000	1.000000	1.000000	1.000000	1.0000000	1.0000000
1	rowMins_X_S	0.9552028	1.003524	1.000517	1.045741	0.9999681	0.7956813
3	rowMins(X[cols, rows])	1.6501734	1.697327	1.705718	1.764498	1.7164901	1.2484431

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	118.586	132.4620	152.1474	149.5030	166.115	227.501
2	rowMins_X_S	126.679	137.5235	158.8114	158.7335	172.216	245.443

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowMins_X_S	1.068246	1.038211	1.043799	1.061741	1.036728	1.078866

Figure: Benchmarking of colMins_X_S() and rowMins_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 5248060 280.3    8529671 455.6  8529671 455.6
Vcells 9991682  76.3   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colMins_X_S = colMins(X_S, na.rm = FALSE), `colMins(X, rows, cols)` = colMins(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colMins(X[rows, cols])` = colMins(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  5248036 280.3    8529671 455.6  8529671 455.6
Vcells 10091735  77.0   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowMins_X_S = rowMins(X_S, na.rm = FALSE), `rowMins(X, cols, rows)` = rowMins(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowMins(X[cols, rows])` = rowMins(X[cols, rows], na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colMins_X_S(), colMins(X, rows, cols)() and colMins(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	colMins(X, rows, cols)	0.091964	0.1034160	0.1141226	0.1105280	0.1181925	0.170529
1	colMins_X_S	0.099554	0.1102085	0.1213548	0.1190625	0.1273965	0.195952
3	colMins(X[rows, cols])	0.197339	0.2213435	0.2450470	0.2387210	0.2648765	0.403472

	expr	min	lq	mean	median	uq	max
2	colMins(X, rows, cols)	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colMins_X_S	1.082532	1.065681	1.063372	1.077216	1.077873	1.149083
3	colMins(X[rows, cols])	2.145829	2.140322	2.147227	2.159824	2.241060	2.366002

Table: Benchmarking of rowMins_X_S(), rowMins(X, cols, rows)() and rowMins(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
2	rowMins(X, cols, rows)	0.110872	0.124601	0.1416099	0.140200	0.1502035	0.306694
1	rowMins_X_S	0.111578	0.120331	0.1399997	0.141158	0.1514435	0.194540
3	rowMins(X[cols, rows])	0.204501	0.225596	0.2596275	0.254827	0.2890105	0.357487

	expr	min	lq	mean	median	uq	max
2	rowMins(X, cols, rows)	1.000000	1.0000000	1.0000000	1.000000	1.000000	1.000000
1	rowMins_X_S	1.006368	0.9657306	0.9886291	1.006833	1.008255	0.634313
3	rowMins(X[cols, rows])	1.844478	1.8105473	1.8333996	1.817596	1.924126	1.165615

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

Table: Benchmarking of colMins_X_S() and rowMins_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
1	colMins_X_S	99.554	110.2085	121.3548	119.0625	127.3965	195.952
2	rowMins_X_S	111.578	120.3310	139.9997	141.1580	151.4435	194.540

	expr	min	lq	mean	median	uq	max
1	colMins_X_S	1.000000	1.000000	1.00000	1.000000	1.000000	1.0000000
2	rowMins_X_S	1.120779	1.091849	1.15364	1.185579	1.188757	0.9927942

Figure: Benchmarking of colMins_X_S() and rowMins_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 23.4 secs.

Reproducibility

To reproduce this report, do:

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

---

Copyright Dongcan Jiang. Last updated on 2021-08-25 19:00:50 (+0200 UTC). Powered by RSP.