matrixStats.benchmarks

---

colCumprods() and rowCumprods() benchmarks on subsetted computation

This report benchmark the performance of colCumprods() and rowCumprods() 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, range = c(-1, 1))

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 5213456 278.5    7916910 422.9  7916910 422.9
Vcells 9827406  75.0   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5200924 277.8    7916910 422.9  7916910 422.9
Vcells 9785878  74.7   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(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	colCumprods_X_S	0.002130	0.0023095	0.0039658	0.0023790	0.0024975	0.156769
2	colCumprods(X, rows, cols)	0.002353	0.0026730	0.0028235	0.0027665	0.0028635	0.005400
3	colCumprods(X[rows, cols])	0.002910	0.0033600	0.0035488	0.0034995	0.0036080	0.007786

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	colCumprods(X, rows, cols)	1.104695	1.157393	0.7119583	1.162884	1.146546	0.0344456
3	colCumprods(X[rows, cols])	1.366197	1.454860	0.8948605	1.470996	1.444645	0.0496654

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(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	rowCumprods_X_S	0.002123	0.0024040	0.0025326	0.0024905	0.0025800	0.004748
2	rowCumprods(X, cols, rows)	0.002473	0.0027365	0.0044240	0.0028100	0.0029145	0.160240
3	rowCumprods(X[cols, rows])	0.003089	0.0034510	0.0035787	0.0035265	0.0036070	0.007463

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.00000
2	rowCumprods(X, cols, rows)	1.164861	1.138311	1.746846	1.128288	1.129651	33.74895
3	rowCumprods(X[cols, rows])	1.455016	1.435524	1.413078	1.415981	1.398062	1.57182

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

Table: Benchmarking of colCumprods_X_S() and rowCumprods_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	colCumprods_X_S	2.130	2.3095	3.96578	2.3790	2.4975	156.769
2	rowCumprods_X_S	2.123	2.4040	2.53257	2.4905	2.5800	4.748

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.0000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	rowCumprods_X_S	0.9967136	1.040918	0.6386058	1.046868	1.033033	0.0302866

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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 5199629 277.7    7916910 422.9  7916910 422.9
Vcells 9454988  72.2   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5199623 277.7    7916910 422.9  7916910 422.9
Vcells 9460071  72.2   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+100x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.016517	0.0172645	0.0173347	0.0173455	0.0174905	0.019696
2	colCumprods(X, rows, cols)	0.023057	0.0240730	0.0242243	0.0241885	0.0243140	0.028630
3	colCumprods(X[rows, cols])	0.027261	0.0286655	0.0293388	0.0288380	0.0290040	0.061713

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.395956	1.394364	1.397446	1.394511	1.390126	1.453595
3	colCumprods(X[rows, cols])	1.650481	1.660372	1.692492	1.662564	1.658272	3.133276

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on integer+100x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.019701	0.0205150	0.0210799	0.0209795	0.021562	0.026141
2	rowCumprods(X, cols, rows)	0.020670	0.0216395	0.0224825	0.0221750	0.022656	0.049934
3	rowCumprods(X[cols, rows])	0.030229	0.0314540	0.0324402	0.0318710	0.033033	0.048070

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.049185	1.054814	1.066538	1.056984	1.050737	1.910179
3	rowCumprods(X[cols, rows])	1.534389	1.533220	1.538916	1.519150	1.532001	1.838874

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

Table: Benchmarking of colCumprods_X_S() and rowCumprods_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	colCumprods_X_S	16.517	17.2645	17.33470	17.3455	17.4905	19.696
2	rowCumprods_X_S	19.701	20.5150	21.07987	20.9795	21.5620	26.141

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.00000	1.000000	1.000000	1.000000
2	rowCumprods_X_S	1.192771	1.188276	1.21605	1.209507	1.232784	1.327224

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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 5200371 277.8    7916910 422.9  7916910 422.9
Vcells 9459045  72.2   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5200365 277.8    7916910 422.9  7916910 422.9
Vcells 9464128  72.3   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(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	colCumprods_X_S	0.016288	0.0171430	0.0174180	0.0173890	0.0175485	0.028024
2	colCumprods(X, rows, cols)	0.023358	0.0244505	0.0261008	0.0254035	0.0256075	0.059107
3	colCumprods(X[rows, cols])	0.027170	0.0285190	0.0305344	0.0296390	0.0299290	0.083092

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.434062	1.426267	1.498493	1.460895	1.459242	2.109156
3	colCumprods(X[rows, cols])	1.668099	1.663595	1.753037	1.704468	1.705502	2.965030

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(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	rowCumprods_X_S	0.020412	0.0214680	0.0224569	0.0221435	0.0226890	0.034829
2	rowCumprods(X, cols, rows)	0.024611	0.0258885	0.0267093	0.0265215	0.0271425	0.032688
3	rowCumprods(X[cols, rows])	0.033090	0.0344075	0.0370630	0.0355460	0.0369685	0.075511

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowCumprods(X, cols, rows)	1.205712	1.205911	1.189362	1.197710	1.196284	0.9385282
3	rowCumprods(X[cols, rows])	1.621105	1.602734	1.650409	1.605257	1.629358	2.1680496

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

Table: Benchmarking of colCumprods_X_S() and rowCumprods_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	colCumprods_X_S	16.288	17.143	17.41801	17.3890	17.5485	28.024
2	rowCumprods_X_S	20.412	21.468	22.45685	22.1435	22.6890	34.829

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.00000	1.000000	1.00000	1.000000	1.000000
2	rowCumprods_X_S	1.253192	1.25229	1.289289	1.27342	1.292931	1.242828

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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 5200576 277.8    7916910 422.9  7916910 422.9
Vcells 9459899  72.2   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5200570 277.8    7916910 422.9  7916910 422.9
Vcells 9464982  72.3   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(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	colCumprods_X_S	0.017480	0.0184885	0.0201267	0.0187010	0.0189595	0.057510
2	colCumprods(X, rows, cols)	0.025187	0.0263920	0.0287659	0.0268645	0.0271745	0.053122
3	colCumprods(X[rows, cols])	0.031217	0.0324140	0.0350349	0.0327510	0.0332205	0.100567

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colCumprods(X, rows, cols)	1.440904	1.427482	1.429242	1.436527	1.433292	0.9237002
3	colCumprods(X[rows, cols])	1.785870	1.753198	1.740718	1.751297	1.752182	1.7486872

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(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	rowCumprods_X_S	0.018619	0.0195895	0.0200671	0.0201430	0.0203505	0.024133
2	rowCumprods(X, cols, rows)	0.021443	0.0219820	0.0231428	0.0226555	0.0229405	0.059412
3	rowCumprods(X[cols, rows])	0.029786	0.0310765	0.0321629	0.0321695	0.0324460	0.046654

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.151673	1.122132	1.153271	1.124733	1.127270	2.461857
3	rowCumprods(X[cols, rows])	1.599764	1.586386	1.602771	1.597056	1.594359	1.933204

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

Table: Benchmarking of colCumprods_X_S() and rowCumprods_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	colCumprods_X_S	17.480	18.4885	20.12670	18.701	18.9595	57.510
2	rowCumprods_X_S	18.619	19.5895	20.06707	20.143	20.3505	24.133

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.00000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	rowCumprods_X_S	1.06516	1.059551	0.9970373	1.077108	1.073367	0.4196314

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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 5200778 277.8    7916910 422.9  7916910 422.9
Vcells 9482560  72.4   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5200781 277.8    7916910 422.9  7916910 422.9
Vcells 9532658  72.8   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+100x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.093983	0.1022130	0.1169567	0.1153145	0.1277795	0.167969
2	colCumprods(X, rows, cols)	0.135618	0.1493470	0.1696957	0.1664825	0.1842450	0.308803
3	colCumprods(X[rows, cols])	0.162529	0.1818865	0.2049450	0.2005405	0.2204470	0.318674

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.443006	1.461135	1.450927	1.443726	1.441898	1.838452
3	colCumprods(X[rows, cols])	1.729345	1.779485	1.752315	1.739074	1.725214	1.897219

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on integer+100x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.125316	0.1368660	0.1525377	0.1466195	0.1662570	0.213777
2	rowCumprods(X, cols, rows)	0.130144	0.1374480	0.1522626	0.1475275	0.1574455	0.255718
3	rowCumprods(X[cols, rows])	0.194207	0.2080405	0.2350866	0.2261010	0.2506580	0.369761

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.0000000	1.000000	1.0000000	1.000000
2	rowCumprods(X, cols, rows)	1.038527	1.004252	0.9981966	1.006193	0.9470007	1.196190
3	rowCumprods(X[cols, rows])	1.549738	1.520031	1.5411708	1.542094	1.5076538	1.729658

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

Table: Benchmarking of colCumprods_X_S() and rowCumprods_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	colCumprods_X_S	93.983	102.213	116.9567	115.3145	127.7795	167.969
2	rowCumprods_X_S	125.316	136.866	152.5377	146.6195	166.2570	213.777

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.00000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods_X_S	1.33339	1.339027	1.304223	1.271475	1.301124	1.272717

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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 5200997 277.8    7916910 422.9  7916910 422.9
Vcells 9483360  72.4   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5200991 277.8    7916910 422.9  7916910 422.9
Vcells 9533443  72.8   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+1000x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.098656	0.1078020	0.1197871	0.1148990	0.1273170	0.173663
2	colCumprods(X, rows, cols)	0.142179	0.1583975	0.1746456	0.1702785	0.1862445	0.240443
3	colCumprods(X[rows, cols])	0.167736	0.1863940	0.2069996	0.2021675	0.2245875	0.315618

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.441159	1.469337	1.457967	1.481984	1.462841	1.384538
3	colCumprods(X[rows, cols])	1.700211	1.729040	1.728062	1.759524	1.764002	1.817416

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on integer+1000x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.119969	0.1302785	0.1498232	0.1470475	0.1629580	0.196475
2	rowCumprods(X, cols, rows)	0.127065	0.1420880	0.1609298	0.1559950	0.1732525	0.286304
3	rowCumprods(X[cols, rows])	0.193109	0.2117385	0.2380216	0.2315960	0.2568630	0.319597

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.059149	1.090648	1.074131	1.060848	1.063173	1.457203
3	rowCumprods(X[cols, rows])	1.609657	1.625276	1.588683	1.574974	1.576253	1.626655

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

Table: Benchmarking of colCumprods_X_S() and rowCumprods_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	colCumprods_X_S	98.656	107.8020	119.7871	114.8990	127.317	173.663
2	rowCumprods_X_S	119.969	130.2785	149.8232	147.0475	162.958	196.475

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods_X_S	1.216033	1.208498	1.250746	1.279798	1.279939	1.131358

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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, range = c(-1, 1))

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 5201215 277.8    7916910 422.9  7916910 422.9
Vcells 9574477  73.1   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5201200 277.8    7916910 422.9  7916910 422.9
Vcells 9574645  73.1   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(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	colCumprods_X_S	0.002207	0.002362	0.0025968	0.0024205	0.0025245	0.015899
2	colCumprods(X, rows, cols)	0.002468	0.002749	0.0028896	0.0028520	0.0029600	0.005228
3	colCumprods(X[rows, cols])	0.003273	0.003464	0.0037712	0.0035635	0.0036935	0.018393

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colCumprods(X, rows, cols)	1.118260	1.163844	1.112761	1.178269	1.172509	0.3288257
3	colCumprods(X[rows, cols])	1.483009	1.466554	1.452245	1.472217	1.463062	1.1568652

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(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	rowCumprods_X_S	0.002188	0.0023635	0.0026572	0.0024385	0.0025375	0.019886
2	rowCumprods(X, cols, rows)	0.002542	0.0026690	0.0029880	0.0027685	0.0029205	0.021330
3	rowCumprods(X[cols, rows])	0.003091	0.0034915	0.0036114	0.0035800	0.0036940	0.006012

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowCumprods(X, cols, rows)	1.161792	1.129258	1.124512	1.135329	1.150936	1.0726139
3	rowCumprods(X[cols, rows])	1.412706	1.477258	1.359091	1.468116	1.455763	0.3023232

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

Table: Benchmarking of colCumprods_X_S() and rowCumprods_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	colCumprods_X_S	2.207	2.3620	2.59682	2.4205	2.5245	15.899
2	rowCumprods_X_S	2.188	2.3635	2.65718	2.4385	2.5375	19.886

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods_X_S	0.991391	1.000635	1.023244	1.007437	1.005149	1.250771

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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 5201414 277.8    7916910 422.9  7916910 422.9
Vcells 9580440  73.1   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5201408 277.8    7916910 422.9  7916910 422.9
Vcells 9590523  73.2   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(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
1	colCumprods_X_S	0.007957	0.0083910	0.0089477	0.0087455	0.009227	0.014468
2	colCumprods(X, rows, cols)	0.012504	0.0130610	0.0139308	0.0135035	0.013929	0.030816
3	colCumprods(X[rows, cols])	0.023097	0.0234925	0.0252585	0.0245470	0.025045	0.058668

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.571446	1.556549	1.556917	1.544051	1.509591	2.129942
3	colCumprods(X[rows, cols])	2.902727	2.799726	2.822910	2.806815	2.714317	4.055018

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(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	rowCumprods_X_S	0.006403	0.007585	0.0098473	0.0085590	0.0125505	0.014323
2	rowCumprods(X, cols, rows)	0.011944	0.012908	0.0187792	0.0133240	0.0266820	0.052664
3	rowCumprods(X[cols, rows])	0.020735	0.022745	0.0286560	0.0238975	0.0364430	0.045024

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.865376	1.701780	1.907035	1.556724	2.125971	3.676883
3	rowCumprods(X[cols, rows])	3.238326	2.998682	2.910029	2.792090	2.903709	3.143476

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

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

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	6.403	7.585	9.84733	8.5590	12.5505	14.323
1	colCumprods_X_S	7.957	8.391	8.94769	8.7455	9.2270	14.468

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	1.000000	1.000000	1.0000000	1.00000	1.0000000	1.000000
1	colCumprods_X_S	1.242699	1.106262	0.9086412	1.02179	0.7351898	1.010124

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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 5201613 277.8    7916910 422.9  7916910 422.9
Vcells 9581854  73.2   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5201607 277.8    7916910 422.9  7916910 422.9
Vcells 9591937  73.2   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(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	colCumprods_X_S	0.012348	0.012643	0.0129564	0.0128440	0.0130935	0.017314
2	colCumprods(X, rows, cols)	0.016246	0.016648	0.0171531	0.0168325	0.0171245	0.032182
3	colCumprods(X[rows, cols])	0.029210	0.029672	0.0305601	0.0299235	0.0305715	0.062543

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.315679	1.316776	1.323905	1.310534	1.307863	1.858727
3	colCumprods(X[rows, cols])	2.365565	2.346911	2.358680	2.329765	2.334861	3.612279

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(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	rowCumprods_X_S	0.011313	0.0119110	0.0124909	0.0121905	0.0125965	0.027712
2	rowCumprods(X, cols, rows)	0.019090	0.0195905	0.0200319	0.0198690	0.0202420	0.023333
3	rowCumprods(X[cols, rows])	0.030922	0.0319430	0.0333944	0.0321860	0.0329190	0.068596

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowCumprods(X, cols, rows)	1.687439	1.644740	1.603713	1.629876	1.606954	0.8419818
3	rowCumprods(X[cols, rows])	2.733316	2.681807	2.673488	2.640253	2.613345	2.4753176

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

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

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	11.313	11.911	12.49093	12.1905	12.5965	27.712
1	colCumprods_X_S	12.348	12.643	12.95644	12.8440	13.0935	17.314

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	colCumprods_X_S	1.091488	1.061456	1.037268	1.053607	1.039455	0.6247835

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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 5201818 277.9    7916910 422.9  7916910 422.9
Vcells 9581990  73.2   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5201812 277.9    7916910 422.9  7916910 422.9
Vcells 9592073  73.2   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(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	colCumprods_X_S	0.007019	0.007859	0.0087901	0.0084445	0.0089715	0.025116
2	colCumprods(X, rows, cols)	0.014859	0.015518	0.0167590	0.0160210	0.0167835	0.047378
3	colCumprods(X[rows, cols])	0.025477	0.026918	0.0280158	0.0281410	0.0289230	0.034762

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	2.116968	1.974552	1.906580	1.897211	1.870757	1.886367
3	colCumprods(X[rows, cols])	3.629719	3.425118	3.187209	3.332465	3.223876	1.384058

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(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	rowCumprods_X_S	0.007061	0.0082845	0.0115762	0.010345	0.0137015	0.028760
2	rowCumprods(X, cols, rows)	0.015604	0.0163425	0.0234464	0.017435	0.0313690	0.112562
3	rowCumprods(X[cols, rows])	0.025231	0.0277925	0.0335552	0.030111	0.0414405	0.062197

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	2.209885	1.972660	2.025395	1.685355	2.289457	3.913839
3	rowCumprods(X[cols, rows])	3.573290	3.354759	2.898629	2.910681	3.024523	2.162622

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

Table: Benchmarking of colCumprods_X_S() and rowCumprods_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
1	colCumprods_X_S	7.019	7.8590	8.79009	8.4445	8.9715	25.116
2	rowCumprods_X_S	7.061	8.2845	11.57623	10.3450	13.7015	28.760

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods_X_S	1.005984	1.054142	1.316964	1.225058	1.527225	1.145087

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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 5202029 277.9    7916910 422.9  7916910 422.9
Vcells 9627459  73.5   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5202023 277.9    7916910 422.9  7916910 422.9
Vcells 9727542  74.3   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(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	colCumprods_X_S	0.049704	0.052417	0.0756478	0.065157	0.0720310	0.255481
2	colCumprods(X, rows, cols)	0.088915	0.092421	0.1254287	0.116584	0.1484445	0.307144
3	colCumprods(X[rows, cols])	0.168890	0.175132	0.2300729	0.199742	0.2688120	0.454898

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.788890	1.763187	1.658061	1.789278	2.060842	1.202219
3	colCumprods(X[rows, cols])	3.397916	3.341130	3.041369	3.065549	3.731893	1.780555

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on double+100x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.034935	0.0377615	0.0431616	0.039901	0.0436150	0.086740
2	rowCumprods(X, cols, rows)	0.069927	0.0724895	0.0810520	0.075033	0.0821490	0.190901
3	rowCumprods(X[cols, rows])	0.136474	0.1427095	0.1520762	0.146919	0.1560295	0.272534

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	2.001632	1.919667	1.877873	1.880479	1.883503	2.200842
3	rowCumprods(X[cols, rows])	3.906512	3.779233	3.523415	3.682088	3.577428	3.141964

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

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

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	34.935	37.7615	43.16159	39.901	43.615	86.740
1	colCumprods_X_S	49.704	52.4170	75.64781	65.157	72.031	255.481

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCumprods_X_S	1.422756	1.388107	1.752665	1.632967	1.651519	2.945365

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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 5202239 277.9    7916910 422.9  7916910 422.9
Vcells 9627600  73.5   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5202233 277.9    7916910 422.9  7916910 422.9
Vcells 9727683  74.3   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(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
1	colCumprods_X_S	0.068627	0.0749925	0.1291245	0.0869250	0.189222	0.287165
2	colCumprods(X, rows, cols)	0.087090	0.0926745	0.1450763	0.1039750	0.211426	0.365470
3	colCumprods(X[rows, cols])	0.172317	0.1843195	0.2874315	0.1964725	0.415401	0.654066

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.269034	1.235784	1.123538	1.196146	1.117344	1.272683
3	colCumprods(X[rows, cols])	2.510921	2.457839	2.226002	2.260253	2.195310	2.277666

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on double+1000x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.054785	0.057548	0.0707300	0.0642870	0.0762220	0.155719
2	rowCumprods(X, cols, rows)	0.107062	0.111125	0.1234247	0.1140990	0.1270610	0.251079
3	rowCumprods(X[cols, rows])	0.178404	0.182863	0.2044453	0.1928805	0.2152165	0.363979

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.954221	1.930997	1.745012	1.774838	1.666986	1.612385
3	rowCumprods(X[cols, rows])	3.256439	3.177573	2.890504	3.000303	2.823548	2.337409

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

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

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	54.785	57.5480	70.72998	64.287	76.222	155.719
1	colCumprods_X_S	68.627	74.9925	129.12454	86.925	189.222	287.165

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	1.00000	1.00000	1.000000	1.00000	1.000000	1.000000
1	colCumprods_X_S	1.25266	1.30313	1.825598	1.35214	2.482512	1.844123

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_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_0.60.0     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 23.27 secs.

Reproducibility

To reproduce this report, do:

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

---

Copyright Dongcan Jiang. Last updated on 2021-08-25 22:13:55 (+0200 UTC). Powered by RSP.