R中大df的距离distHaversine模型的优化

install.packages("geosphere")
install.packages("data.table")
install.packages("parallel")
install.packages("foreach")
install.packages("doParallel")

library(geosphere)
library(data.table)
library(parallel)
library(foreach)
library(doParallel)

set.seed(123)
n <- 10000  # 数据点数量
df <- data.frame(
  id = 1:n,
  lat = runif(n, -90, 90),
  lon = runif(n, -180, 180)
)

coords <- as.matrix(df[, c("lon", "lat")])
dist_matrix <- distm(coords, fun = distHaversine)

numCores <- detectCores() - 1  # 使用可用核心数减1
cl <- makeCluster(numCores)
registerDoParallel(cl)

dist_parallel <- function(coords) {
  n <- nrow(coords)
  dist_matrix <- matrix(NA, n, n)
  
  foreach(i = 1:n, .combine = rbind, .packages = "geosphere") %dopar% {
    dist_row <- distHaversine(coords[i, ], coords)
    return(dist_row)
  }
}

coords <- as.matrix(df[, c("lon", "lat")])
dist_matrix <- dist_parallel(coords)

stopCluster(cl)