matrixStats.benchmarks

---

colVars() and rowVars() benchmarks

This report benchmark the performance of colVars() and rowVars() against alternative methods.

Alternative methods

• apply() + var()
• colVarColMeans() and rowVarColMeans()
• genefilter::rowVars(t(.)) and genefilter::rowVars()

where

> colVarColMeans <- function(x, na.rm = TRUE) {
+     if (na.rm) {
+         n <- colSums(!is.na(x))
+     }     else {
+         n <- nrow(x)
+     }
+     var <- colMeans(x * x, na.rm = na.rm) - (colMeans(x, na.rm = na.rm))^2
+     var * n/(n - 1)
+ }

and

> rowVarRowMeans <- function(x, na.rm = TRUE) {
+     if (na.rm) {
+         n <- rowSums(!is.na(x))
+     }     else {
+         n <- ncol(x)
+     }
+     mu <- rowMeans(x, na.rm = na.rm)
+     var <- rowMeans(x * x, na.rm = na.rm) - mu^2
+     var * (n/(n - 1))
+ }

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"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5323410 284.4    7916910 422.9  7916910 422.9
Vcells 10683950  81.6   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5323236 284.3    7916910 422.9  7916910 422.9
Vcells 10683390  81.6   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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	colVars	0.002790	0.0036640	0.0048791	0.0049495	0.0056715	0.011849
3	colVarColMeans	0.013767	0.0164575	0.0195932	0.0183650	0.0211365	0.073837
5	genefilter::rowVars(t(.))	0.036209	0.0429655	0.0568208	0.0474250	0.0527465	0.908206
2	colVarsCenter	0.036718	0.0448785	0.0572104	0.0485610	0.0545365	0.778291
4	apply+var	0.120617	0.1341465	0.1461783	0.1463775	0.1534665	0.324228

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.00000	1.000000	1.000000	1.000000
3	colVarColMeans	4.934409	4.491676	4.01573	3.710476	3.726792	6.231496
5	genefilter::rowVars(t(.))	12.978136	11.726392	11.64573	9.581776	9.300273	76.648325
2	colVarsCenter	13.160573	12.248499	11.72557	9.811294	9.615886	65.684108
4	apply+var	43.231900	36.612036	29.96003	29.574199	27.059244	27.363322

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.002495	0.0039705	0.0046387	0.0046195	0.0052830	0.011179
3	rowVarRowMeans	0.014904	0.0172350	0.0187539	0.0183880	0.0199580	0.030437
2	rowVarsCenter	0.023214	0.0273785	0.0326496	0.0295920	0.0338655	0.160114
5	genefilter::rowVars	0.031202	0.0358225	0.0396764	0.0377730	0.0414265	0.077452
4	apply+var	0.117099	0.1242575	0.1377096	0.1363450	0.1441630	0.273160

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	rowVarRowMeans	5.973547	4.340763	4.042904	3.980517	3.777778	2.722694
2	rowVarsCenter	9.304208	6.895479	7.038487	6.405888	6.410278	14.322748
5	genefilter::rowVars	12.505812	9.022163	8.553301	8.176859	7.841473	6.928348
4	apply+var	46.933467	31.295177	29.686985	29.515099	27.288094	24.435101

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on integer+10x10 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	rowVars	2.495	3.9705	4.63872	4.6195	5.2830	11.179
1	colVars	2.790	3.6640	4.87911	4.9495	5.6715	11.849

	expr	min	lq	mean	median	uq	max
2	rowVars	1.000000	1.0000000	1.000000	1.000000	1.000000	1.000000
1	colVars	1.118237	0.9228057	1.051823	1.071436	1.073538	1.059934

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

100x100 integer matrix

> X <- data[["100x100"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5321832 284.3    7916910 422.9  7916910 422.9
Vcells 10300053  78.6   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5321826 284.3    7916910 422.9  7916910 422.9
Vcells 10305160  78.7   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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	colVars	0.029493	0.0314210	0.0359694	0.0324045	0.0391645	0.085549
3	colVarColMeans	0.060018	0.0631925	0.0697104	0.0653240	0.0717360	0.109399
5	genefilter::rowVars(t(.))	0.158201	0.1678695	0.1894418	0.1747950	0.2096820	0.310294
2	colVarsCenter	0.201342	0.2099810	0.2387271	0.2169910	0.2373740	0.623733
4	apply+var	0.867449	0.8876425	0.9861745	0.8971375	1.0521530	1.518791

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	colVarColMeans	2.034991	2.011155	1.938045	2.015893	1.831659	1.278788
5	genefilter::rowVars(t(.))	5.364019	5.342589	5.266746	5.394158	5.353879	3.627091
2	colVarsCenter	6.826772	6.682824	6.636946	6.696323	6.060948	7.290944
4	apply+var	29.412030	28.249976	27.417025	27.685584	26.864967	17.753463

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.029612	0.0323630	0.0366160	0.035609	0.0396005	0.056789
2	rowVarsCenter	0.098741	0.1032345	0.1222379	0.112988	0.1293390	0.385498
3	rowVarRowMeans	0.105727	0.1081810	0.1242227	0.115162	0.1371180	0.197712
5	genefilter::rowVars	0.143625	0.1524535	0.1702272	0.162399	0.1859700	0.252621
4	apply+var	0.869321	0.8938000	1.0341690	0.962347	1.1378060	1.552729

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowVarsCenter	3.334493	3.189893	3.338377	3.173018	3.266095	6.788251
3	rowVarRowMeans	3.570411	3.342737	3.392583	3.234070	3.462532	3.481519
5	genefilter::rowVars	4.850230	4.710735	4.648989	4.560617	4.696153	4.448414
4	apply+var	29.357051	27.617959	28.243668	27.025387	28.732112	27.342073

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on integer+100x100 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	colVars	29.493	31.421	35.96942	32.4045	39.1645	85.549
2	rowVars	29.612	32.363	36.61596	35.6090	39.6005	56.789

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.00000	1.000000	1.000000	1.000000	1.0000000
2	rowVars	1.004035	1.02998	1.017975	1.098891	1.011132	0.6638184

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

1000x10 integer matrix

> X <- data[["1000x10"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5322648 284.3    7916910 422.9  7916910 422.9
Vcells 10304337  78.7   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5322630 284.3    7916910 422.9  7916910 422.9
Vcells 10309424  78.7   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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	colVars	0.029744	0.0326025	0.0379358	0.0373775	0.0413415	0.056173
3	colVarColMeans	0.057804	0.0621950	0.0725662	0.0706915	0.0786780	0.113807
2	colVarsCenter	0.134100	0.1441075	0.1762446	0.1669745	0.1880185	0.515380
4	apply+var	0.185813	0.2007390	0.2334204	0.2226225	0.2503955	0.386809
5	genefilter::rowVars(t(.))	0.354563	0.3672265	0.4287826	0.4043750	0.4703875	0.625783

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	colVarColMeans	1.943383	1.907676	1.912868	1.891285	1.903124	2.026009
2	colVarsCenter	4.508472	4.420136	4.645863	4.467246	4.547936	9.174871
4	apply+var	6.247075	6.157166	6.153035	5.956056	6.056759	6.886031
5	genefilter::rowVars(t(.))	11.920488	11.263753	11.302843	10.818674	11.378095	11.140281

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.030753	0.0324700	0.0372454	0.0335195	0.0405495	0.076945
4	apply+var	0.190341	0.1975405	0.2231841	0.2042645	0.2409250	0.389536
2	rowVarsCenter	0.196543	0.2008960	0.2324903	0.2124490	0.2391820	0.688585
3	rowVarRowMeans	0.300913	0.3038920	0.3421468	0.3065935	0.3728095	0.546615
5	genefilter::rowVars	0.342112	0.3464690	0.3848694	0.3507630	0.4204030	0.594216

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
4	apply+var	6.189347	6.083785	5.992253	6.093901	5.941504	5.062525
2	rowVarsCenter	6.391019	6.187127	6.242115	6.338072	5.898519	8.949055
3	rowVarRowMeans	9.784834	9.359162	9.186274	9.146721	9.193936	7.103970
5	genefilter::rowVars	11.124508	10.670434	10.333328	10.464446	10.367649	7.722607

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on integer+1000x10 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	rowVars	30.753	32.4700	37.24544	33.5195	40.5495	76.945
1	colVars	29.744	32.6025	37.93582	37.3775	41.3415	56.173

	expr	min	lq	mean	median	uq	max
2	rowVars	1.0000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	colVars	0.9671902	1.004081	1.018536	1.115097	1.019532	0.7300409

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

10x1000 integer matrix

> X <- data[["10x1000"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5322877 284.3    7916910 422.9  7916910 422.9
Vcells 10305352  78.7   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5322871 284.3    7916910 422.9  7916910 422.9
Vcells 10310459  78.7   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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	colVars	0.025437	0.0281355	0.0330787	0.0303585	0.0347165	0.088563
3	colVarColMeans	0.067336	0.0727130	0.0821569	0.0792020	0.0878745	0.159916
5	genefilter::rowVars(t(.))	0.152472	0.1628310	0.1871207	0.1771005	0.2003765	0.302403
2	colVarsCenter	0.808813	0.8516725	0.9225544	0.8970735	0.9537135	1.549413
4	apply+var	7.539137	7.9234815	8.8560889	8.1060290	9.1086705	15.417266

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	colVarColMeans	2.647168	2.584386	2.483683	2.608890	2.531203	1.805675
5	genefilter::rowVars(t(.))	5.994103	5.787386	5.656840	5.833638	5.771794	3.414552
2	colVarsCenter	31.796713	30.270388	27.889715	29.549335	27.471476	17.495037
4	apply+var	296.384676	281.618649	267.728163	267.010195	262.372950	174.082472

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.025184	0.0267130	0.0315241	0.0295795	0.0344945	0.048935
3	rowVarRowMeans	0.092200	0.0984980	0.1119897	0.1068300	0.1206400	0.205226
2	rowVarsCenter	0.096437	0.1044115	0.1277320	0.1192480	0.1425225	0.280779
5	genefilter::rowVars	0.133011	0.1428100	0.1632748	0.1588605	0.1756300	0.257010
4	apply+var	7.510383	7.9770260	8.7878154	8.3303785	8.8686645	15.979274

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	rowVarRowMeans	3.661055	3.687268	3.552514	3.611623	3.497369	4.193849
2	rowVarsCenter	3.829296	3.908640	4.051887	4.031441	4.131746	5.737795
5	genefilter::rowVars	5.281568	5.346086	5.179367	5.370628	5.091536	5.252069
4	apply+var	298.220418	298.619623	278.765167	281.626752	257.103727	326.540799

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on integer+10x1000 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	rowVars	25.184	26.7130	31.52408	29.5795	34.4945	48.935
1	colVars	25.437	28.1355	33.07866	30.3585	34.7165	88.563

	expr	min	lq	mean	median	uq	max
2	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colVars	1.010046	1.053251	1.049314	1.026336	1.006436	1.809809

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

100x1000 integer matrix

> X <- data[["100x1000"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5323133 284.3    7916910 422.9  7916910 422.9
Vcells 10306115  78.7   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5323115 284.3    7916910 422.9  7916910 422.9
Vcells 10356202  79.1   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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	colVars	0.276481	0.2844305	0.3006793	0.2949475	0.3056915	0.392107
3	colVarColMeans	0.500416	0.5305050	0.5686602	0.5443880	0.5724725	1.022393
5	genefilter::rowVars(t(.))	1.202326	1.2812000	1.3567900	1.3134570	1.3727510	1.959446
2	colVarsCenter	1.823917	1.9172815	2.0564000	1.9740960	2.1424225	3.386400
4	apply+var	8.590804	9.0592370	10.4623841	9.2505530	10.0627455	27.779531

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	colVarColMeans	1.809947	1.865148	1.891252	1.845711	1.872713	2.607434
5	genefilter::rowVars(t(.))	4.348675	4.504440	4.512416	4.453189	4.490642	4.997223
2	colVarsCenter	6.596898	6.740773	6.839182	6.693042	7.008446	8.636418
4	apply+var	31.071951	31.850441	34.795830	31.363388	32.917976	70.846812

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.284963	0.3013385	0.3331891	0.3121415	0.3636835	0.479973
2	rowVarsCenter	0.733769	0.7770355	0.8441678	0.7918840	0.8349365	2.058860
3	rowVarRowMeans	0.765500	0.8070590	0.8580390	0.8255715	0.8656500	1.262896
5	genefilter::rowVars	1.078433	1.1613040	1.2323091	1.1982455	1.2523120	1.768914
4	apply+var	8.664578	9.0873635	10.0694555	9.2850940	9.9189880	21.841665

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowVarsCenter	2.574962	2.578613	2.533600	2.536939	2.295778	4.289533
3	rowVarRowMeans	2.686314	2.678247	2.575232	2.644863	2.380229	2.631181
5	genefilter::rowVars	3.784467	3.853819	3.698528	3.838789	3.443412	3.685445
4	apply+var	30.405976	30.156663	30.221447	29.746426	27.273682	45.506028

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on integer+100x1000 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	colVars	276.481	284.4305	300.6793	294.9475	305.6915	392.107
2	rowVars	284.963	301.3385	333.1891	312.1415	363.6835	479.973

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowVars	1.030678	1.059445	1.108121	1.058295	1.189708	1.224087

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

1000x100 integer matrix

> X <- data[["1000x100"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5323376 284.3    7916910 422.9  7916910 422.9
Vcells 10306961  78.7   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5323358 284.3    7916910 422.9  7916910 422.9
Vcells 10357048  79.1   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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	colVars	0.269630	0.2836650	0.3066505	0.2912140	0.316396	0.478095
3	colVarColMeans	0.481399	0.5000665	0.5435529	0.5164555	0.581450	0.781772
2	colVarsCenter	1.124666	1.1605340	1.3431052	1.2034195	1.356642	8.585024
5	genefilter::rowVars(t(.))	1.352351	1.4074475	1.5987525	1.4800315	1.611156	8.836442
4	apply+var	1.712556	1.7864945	2.0692958	1.8218650	2.068468	9.698091

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	colVarColMeans	1.785406	1.762877	1.772548	1.773457	1.837729	1.635181
2	colVarsCenter	4.171146	4.091213	4.379921	4.132423	4.287799	17.956732
5	genefilter::rowVars(t(.))	5.015581	4.961654	5.213598	5.082281	5.092214	18.482607
4	apply+var	6.351504	6.297903	6.748059	6.256104	6.537592	20.284862

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.297531	0.3048665	0.3284983	0.317067	0.323917	0.710802
2	rowVarsCenter	0.828426	0.8628325	0.9372421	0.886775	0.965110	2.109736
3	rowVarRowMeans	0.959662	0.9895235	1.0586666	1.020647	1.069908	1.831917
5	genefilter::rowVars	1.253544	1.3330455	1.5628775	1.370334	1.419963	9.320973
4	apply+var	1.716572	1.7808530	1.9932946	1.827252	1.941906	9.358598

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowVarsCenter	2.784335	2.830198	2.853111	2.796806	2.979498	2.968106
3	rowVarRowMeans	3.225418	3.245760	3.222746	3.219026	3.303030	2.577254
5	genefilter::rowVars	4.213154	4.372555	4.757642	4.321907	4.383725	13.113319
4	apply+var	5.769389	5.841419	6.067899	5.762982	5.995071	13.166252

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on integer+1000x100 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	colVars	269.630	283.6650	306.6505	291.214	316.396	478.095
2	rowVars	297.531	304.8665	328.4983	317.067	323.917	710.802

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowVars	1.103479	1.074741	1.071247	1.088777	1.023771	1.486738

Figure: Benchmarking of colVars() and rowVars() 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"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5323622 284.4    7916910 422.9  7916910 422.9
Vcells 10422988  79.6   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5323607 284.4    7916910 422.9  7916910 422.9
Vcells 10423180  79.6   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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	colVars	0.002645	0.0035485	0.0048936	0.0049525	0.0057125	0.011078
3	colVarColMeans	0.013680	0.0158700	0.0190106	0.0172255	0.0203090	0.090102
5	genefilter::rowVars(t(.))	0.035305	0.0408875	0.0462702	0.0448195	0.0502385	0.115590
2	colVarsCenter	0.037691	0.0435315	0.0506277	0.0471360	0.0529150	0.183315
4	apply+var	0.121929	0.1308385	0.1435156	0.1403530	0.1514275	0.208543

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	colVarColMeans	5.172023	4.472312	3.884788	3.478142	3.555186	8.133418
5	genefilter::rowVars(t(.))	13.347826	11.522474	9.455244	9.049874	8.794486	10.434194
2	colVarsCenter	14.249906	12.267578	10.345690	9.517617	9.263020	16.547662
4	apply+var	46.097921	36.871495	29.327207	28.339828	26.508096	18.824968

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.002540	0.0038570	0.0045251	0.0045430	0.0051150	0.010656
3	rowVarRowMeans	0.011867	0.0138375	0.0155078	0.0152215	0.0166525	0.026797
2	rowVarsCenter	0.022382	0.0262275	0.0306844	0.0284040	0.0310435	0.152408
5	genefilter::rowVars	0.029586	0.0348960	0.0379834	0.0371150	0.0395490	0.081871
4	apply+var	0.119218	0.1287145	0.1404636	0.1397160	0.1478400	0.239180

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	rowVarRowMeans	4.672047	3.587633	3.427049	3.350539	3.255621	2.514734
2	rowVarsCenter	8.811811	6.799974	6.780893	6.252256	6.069110	14.302553
5	genefilter::rowVars	11.648032	9.047446	8.393874	8.169712	7.731965	7.683089
4	apply+var	46.936220	33.371662	31.040788	30.754127	28.903226	22.445571

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on double+10x10 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	rowVars	2.540	3.8570	4.52513	4.5430	5.1150	10.656
1	colVars	2.645	3.5485	4.89360	4.9525	5.7125	11.078

	expr	min	lq	mean	median	uq	max
2	rowVars	1.000000	1.0000000	1.000000	1.000000	1.000000	1.000000
1	colVars	1.041339	0.9200156	1.081427	1.090139	1.116813	1.039602

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

100x100 double matrix

> X <- data[["100x100"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5323847 284.4    7916910 422.9  7916910 422.9
Vcells 10424109  79.6   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5323841 284.4    7916910 422.9  7916910 422.9
Vcells 10434216  79.7   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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
1	colVars	0.029262	0.0314835	0.0352890	0.0326715	0.0379055	0.080101
3	colVarColMeans	0.030771	0.0350795	0.0392033	0.0377445	0.0409125	0.072354
5	genefilter::rowVars(t(.))	0.150565	0.1594915	0.1787966	0.1676460	0.1931260	0.294593
2	colVarsCenter	0.169785	0.1835170	0.2073874	0.1919255	0.2075525	0.529378
4	apply+var	0.862029	0.8787865	0.9715484	0.9086425	1.0296300	1.514788

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
3	colVarColMeans	1.051569	1.114219	1.110924	1.155273	1.079329	0.9032846
5	genefilter::rowVars(t(.))	5.145410	5.065876	5.066643	5.131261	5.094933	3.6777693
2	colVarsCenter	5.802235	5.828990	5.876837	5.874401	5.475525	6.6088813
4	apply+var	29.458991	27.912605	27.531229	27.811472	27.163077	18.9109749

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.029708	0.031666	0.0360687	0.0338565	0.0385060	0.087151
3	rowVarRowMeans	0.063648	0.065847	0.0770080	0.0688575	0.0793815	0.231908
2	rowVarsCenter	0.088075	0.090904	0.1105511	0.0953995	0.1116555	0.646862
5	genefilter::rowVars	0.135403	0.141491	0.1549294	0.1449895	0.1631460	0.231369
4	apply+var	0.860343	0.882569	1.0549984	0.8982775	1.0883370	3.406908

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	rowVarRowMeans	2.142453	2.079423	2.135034	2.033804	2.061536	2.660991
2	rowVarsCenter	2.964690	2.870713	3.065011	2.817760	2.899691	7.422313
5	genefilter::rowVars	4.557796	4.468231	4.295392	4.282472	4.236898	2.654806
4	apply+var	28.959977	27.871187	29.249661	26.531907	28.264089	39.092013

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on double+100x100 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	colVars	29.262	31.4835	35.28896	32.6715	37.9055	80.101
2	rowVars	29.708	31.6660	36.06874	33.8565	38.5060	87.151

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.00000	1.000000	1.000000
2	rowVars	1.015242	1.005797	1.022097	1.03627	1.015842	1.088014

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

1000x10 double matrix

> X <- data[["1000x10"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5324110 284.4    7916910 422.9  7916910 422.9
Vcells 10425483  79.6   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5324092 284.4    7916910 422.9  7916910 422.9
Vcells 10435570  79.7   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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	colVars	0.030654	0.0352160	0.0402149	0.0393340	0.0445555	0.056132
3	colVarColMeans	0.036769	0.0409280	0.0467151	0.0451770	0.0512440	0.071497
2	colVarsCenter	0.115286	0.1257460	0.1513958	0.1421240	0.1663635	0.441184
4	apply+var	0.181837	0.2017375	0.2336601	0.2264865	0.2485645	0.381642
5	genefilter::rowVars(t(.))	0.253346	0.2715770	0.3169921	0.3047620	0.3566390	0.457169

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	colVarColMeans	1.199485	1.162199	1.161635	1.148548	1.150116	1.273730
2	colVarsCenter	3.760880	3.570707	3.764666	3.613261	3.733849	7.859759
4	apply+var	5.931917	5.728575	5.810284	5.758034	5.578761	6.799010
5	genefilter::rowVars(t(.))	8.264696	7.711750	7.882450	7.748055	8.004377	8.144534

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.031942	0.0352095	0.0424953	0.0404315	0.0463405	0.093285
3	rowVarRowMeans	0.063618	0.0673155	0.0801192	0.0778555	0.0881445	0.149555
2	rowVarsCenter	0.087956	0.0998060	0.1226849	0.1146525	0.1419555	0.301986
4	apply+var	0.180436	0.1949160	0.2333630	0.2286680	0.2531570	0.379833
5	genefilter::rowVars	0.234324	0.2501810	0.2888207	0.2762030	0.3149190	0.434421

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	rowVarRowMeans	1.991672	1.911856	1.885367	1.925615	1.902105	1.603205
2	rowVarsCenter	2.753616	2.834633	2.887025	2.835722	3.063314	3.237241
4	apply+var	5.648864	5.535892	5.491507	5.655689	5.462975	4.071748
5	genefilter::rowVars	7.335921	7.105497	6.796539	6.831381	6.795762	4.656922

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on double+1000x10 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	colVars	30.654	35.2160	40.21492	39.3340	44.5555	56.132
2	rowVars	31.942	35.2095	42.49526	40.4315	46.3405	93.285

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.0000000	1.000000	1.000000	1.000000	1.000000
2	rowVars	1.042017	0.9998154	1.056704	1.027902	1.040062	1.661886

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

10x1000 double matrix

> X <- data[["10x1000"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5324338 284.4    7916910 422.9  7916910 422.9
Vcells 10425622  79.6   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5324332 284.4    7916910 422.9  7916910 422.9
Vcells 10435729  79.7   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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	colVars	0.023794	0.0268210	0.0302548	0.0284730	0.0322055	0.061424
3	colVarColMeans	0.040555	0.0453620	0.0521165	0.0489505	0.0545395	0.125120
5	genefilter::rowVars(t(.))	0.155559	0.1661205	0.1876453	0.1857365	0.1993210	0.289969
2	colVarsCenter	0.781747	0.8249380	0.8807925	0.8574385	0.9077215	1.151497
4	apply+var	7.532308	7.9709260	8.7646401	8.1539695	8.7643480	15.208067

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.00000	1.000000	1.000000
3	colVarColMeans	1.704421	1.691287	1.722584	1.71919	1.693484	2.036989
5	genefilter::rowVars(t(.))	6.537741	6.193673	6.202164	6.52325	6.189036	4.720777
2	colVarsCenter	32.854795	30.757168	29.112468	30.11409	28.185294	18.746695
4	apply+var	316.563335	297.189739	289.694010	286.37550	272.138237	247.591609

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.023416	0.0251300	0.0878095	0.0272560	0.0297025	5.961206
3	rowVarRowMeans	0.069447	0.0733340	0.0804046	0.0771735	0.0849200	0.143452
2	rowVarsCenter	0.093996	0.1007760	0.1149330	0.1098630	0.1198065	0.258423
5	genefilter::rowVars	0.135324	0.1445095	0.2198861	0.1519365	0.1646340	6.212482
4	apply+var	7.574588	7.9150710	8.5796664	8.1310805	8.7260745	15.537949

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
3	rowVarRowMeans	2.965793	2.918185	0.9156709	2.831432	2.859019	0.0240643
2	rowVarsCenter	4.014178	4.010187	1.3088902	4.030782	4.033549	0.0433508
5	genefilter::rowVars	5.779125	5.750477	2.5041257	5.574424	5.542766	1.0421519
4	apply+var	323.479159	314.965022	97.7076906	298.322589	293.782493	2.6065110

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on double+10x1000 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	rowVars	23.416	25.130	87.80953	27.256	29.7025	5961.206
1	colVars	23.794	26.821	30.25482	28.473	32.2055	61.424

	expr	min	lq	mean	median	uq	max
2	rowVars	1.000000	1.00000	1.0000000	1.000000	1.000000	1.000000
1	colVars	1.016143	1.06729	0.3445505	1.044651	1.084269	0.010304

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

100x1000 double matrix

> X <- data[["100x1000"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5324592 284.4    7916910 422.9  7916910 422.9
Vcells 10427217  79.6   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5324574 284.4    7916910 422.9  7916910 422.9
Vcells 10527304  80.4   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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
3	colVarColMeans	0.217100	0.2541795	0.2836950	0.2733770	0.3066210	0.532291
1	colVars	0.276846	0.2965845	0.3296974	0.3253395	0.3629285	0.407909
5	genefilter::rowVars(t(.))	1.189254	1.2876725	1.6077946	1.4054090	1.5729725	15.897295
2	colVarsCenter	1.504681	1.6610540	1.8318492	1.8216455	1.9761710	2.701192
4	apply+var	8.455217	9.0191355	10.3333435	9.3339725	10.6362330	23.139245

	expr	min	lq	mean	median	uq	max
3	colVarColMeans	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	colVars	1.275200	1.166831	1.162155	1.190076	1.183639	0.7663271
5	genefilter::rowVars(t(.))	5.477909	5.065997	5.667336	5.140919	5.130022	29.8657971
2	colVarsCenter	6.930820	6.534965	6.457108	6.663492	6.444996	5.0746528
4	apply+var	38.946186	35.483332	36.424134	34.143225	34.688534	43.4710431

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.286441	0.3034615	0.3443770	0.3196560	0.3715660	0.865443
3	rowVarRowMeans	0.510146	0.5593850	0.5929959	0.5695750	0.6114635	0.828932
2	rowVarsCenter	0.715315	0.7583490	0.8288214	0.7897215	0.8284805	1.622231
5	genefilter::rowVars	1.069789	1.1644100	1.2231647	1.1932250	1.2426030	1.825741
4	apply+var	8.558351	9.0009320	9.9293128	9.2546195	9.8413630	18.806774

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
3	rowVarRowMeans	1.780981	1.843347	1.721938	1.781837	1.645639	0.9578124
2	rowVarsCenter	2.497251	2.498996	2.406727	2.470535	2.229699	1.8744516
5	genefilter::rowVars	3.734762	3.837093	3.551819	3.732841	3.344232	2.1096028
4	apply+var	29.878233	29.660870	28.832685	28.951809	26.486177	21.7308061

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on double+100x1000 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	rowVars	286.441	303.4615	344.3770	319.6560	371.5660	865.443
1	colVars	276.846	296.5845	329.6974	325.3395	362.9285	407.909

	expr	min	lq	mean	median	uq	max
2	rowVars	1.0000000	1.0000000	1.0000000	1.00000	1.0000000	1.0000000
1	colVars	0.9665027	0.9773381	0.9573737	1.01778	0.9767538	0.4713297

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

1000x100 double matrix

> X <- data[["1000x100"]]
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5324835 284.4    7916910 422.9  7916910 422.9
Vcells 10427379  79.6   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colVars = colVars(X, na.rm = FALSE), colVarsCenter = colVars(X, center = colMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), colVarColMeans = colVarColMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 2L, FUN = var, na.rm = FALSE), `genefilter::rowVars(t(.))` = genefilter_colVars(X, na.rm = FALSE), 
+     unit = "ms")
> X <- t(X)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5324817 284.4    7916910 422.9  7916910 422.9
Vcells 10527466  80.4   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowVars = rowVars(X, na.rm = FALSE), rowVarsCenter = rowVars(X, center = rowMeans(X, 
+     na.rm = FALSE), na.rm = FALSE), rowVarRowMeans = rowVarRowMeans(X, na.rm = FALSE), `apply+var` = apply(X, 
+     MARGIN = 1L, FUN = var, na.rm = FALSE), `genefilter::rowVars` = genefilter_rowVars(X, na.rm = FALSE), 
+     unit = "ms")

Table: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() 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
3	colVarColMeans	0.235789	0.2501695	0.3545995	0.2654355	0.3072345	7.212736
1	colVars	0.269846	0.2757440	0.3202453	0.2955760	0.3484210	0.505107
2	colVarsCenter	0.845354	0.8825620	1.0393725	0.9923470	1.1263665	1.902730
5	genefilter::rowVars(t(.))	1.285008	1.3302660	1.7078352	1.4647050	1.7319530	9.753642
4	apply+var	1.611236	1.6562830	1.9752686	1.7862820	2.0936840	9.164259

	expr	min	lq	mean	median	uq	max
3	colVarColMeans	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
1	colVars	1.144438	1.102229	0.9031182	1.113551	1.134056	0.0700299
2	colVarsCenter	3.585214	3.527856	2.9311164	3.738562	3.666146	0.2638014
5	genefilter::rowVars(t(.))	5.449822	5.317459	4.8162364	5.518120	5.637235	1.3522805
4	apply+var	6.833381	6.620643	5.5704209	6.729627	6.814612	1.2705663

Table: Benchmarking of rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() 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	rowVars	0.298175	0.308941	0.3325839	0.3199880	0.3469940	0.494310
3	rowVarRowMeans	0.508590	0.548085	0.5971648	0.5733280	0.6302175	0.848073
2	rowVarsCenter	0.716750	0.754393	0.8309605	0.7760755	0.8706715	1.443821
5	genefilter::rowVars	1.127413	1.209552	1.5236509	1.2495070	1.4158060	8.546254
4	apply+var	1.670014	1.726946	1.8908866	1.8097425	2.0256590	2.828460

	expr	min	lq	mean	median	uq	max
1	rowVars	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	rowVarRowMeans	1.705676	1.774077	1.795532	1.791717	1.816220	1.715670
2	rowVarsCenter	2.403790	2.441867	2.498499	2.425327	2.509183	2.920882
5	genefilter::rowVars	3.781045	3.915155	4.581253	3.904856	4.080203	17.289260
4	apply+var	5.600785	5.589889	5.685443	5.655657	5.837735	5.722037

Figure: Benchmarking of colVars(), colVarsCenter(), colVarColMeans(), apply+var() and genefilter::rowVars(t(.))() on double+1000x100 data as well as rowVars(), rowVarsCenter(), rowVarRowMeans(), apply+var() and genefilter::rowVars() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colVars() and rowVars() 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	colVars	269.846	275.744	320.2453	295.576	348.421	505.107
2	rowVars	298.175	308.941	332.5839	319.988	346.994	494.310

	expr	min	lq	mean	median	uq	max
1	colVars	1.000000	1.000000	1.000000	1.000000	1.0000000	1.0000000
2	rowVars	1.104982	1.120391	1.038529	1.082591	0.9959044	0.9786243

Figure: Benchmarking of colVars() and rowVars() 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_0.60.0     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] rstudioapi_0.13         rappdirs_0.3.3          startup_0.15.0         
[67] labeling_0.4.2          bitops_1.0-7            base64enc_0.1-3        
[70] boot_1.3-28             gtable_0.3.0            DBI_1.1.1              
[73] markdown_1.1            R6_2.5.1                lpSolveAPI_5.5.2.0-17.7
[76] rle_0.9.2               dplyr_1.0.7             fastmap_1.1.0          
[79] bit_4.0.4               utf8_1.2.2              parallel_4.1.1         
[82] Rcpp_1.0.7              vctrs_0.3.8             png_0.1-7              
[85] DEoptimR_1.0-9          tidyselect_1.1.1        xfun_0.25              
[88] coda_0.19-4            

Total processing time was 37.33 secs.

Reproducibility

To reproduce this report, do:

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

---

Copyright Henrik Bengtsson. Last updated on 2021-08-25 22:31:39 (+0200 UTC). Powered by RSP.