a.c

#include <stdio.h>
int main(){
  char bytes[65536];
  int len;
  int count = 0;
  int col = 0;
  int filter[256]={0};
  for(int i=0;i<5;i++)filter["aeiou"[i]]=1;
  while((len = fread(bytes, 1, 65536, stdin)) > 0){
    for(int i=0; i<len; i++){
      char b=bytes[i];
      if(b=='\n'){
        col=0;
      }else if(b==','){
        col++;
      }else if(col==1){
        count+=filter[b];
      }
    }
  }
  printf("%d\n", count);
  return 0;
}

a.go

package main

import (
	"fmt"
	"os"
)

func main() {
	var count, col int
	buf := make([]byte, 65536)
	arr := make([]int, 256)
	for _, c := range "aeiou" {
		arr[c] = 1
	}
	for {
		len, err := os.Stdin.Read(buf)
		if err != nil {
			break
		}
		for i := 0; i < len; i++ {
			c := buf[i]
			if c == ',' {
				col++
			} else if c == '\n' {
				col = 0
			} else if col == 1 {
				count += arr[c]
			}
		}
	}
	fmt.Println(count)
}

a.rb

count = 0
target = 'aeiou'
s = ''
STDIN.each_line do |line|
  s << line.split(',', 2).last
  if s.size > 0x10000
    count += s.count target
    s = ''
  end
end
puts count + s.count(target)

b.go

package main

import (
	"fmt"
	"os"
)

type Input struct {
	id, len int
	buf     []byte
}

type TempResult struct {
	id, first, second, count, col int
}

func partialCount(in Input, chBuf chan []byte, chResult chan TempResult) {
	firstRow := true
	var col int
	var result TempResult
	filter := make([]int, 256)
	for _, c := range "aeiou" {
		filter[c] = 1
	}
	for i := 0; i < in.len; i++ {
		c := in.buf[i]
		if c == ',' {
			col++
		} else if c == '\n' {
			col = 0
			firstRow = false
		}
		if firstRow {
			if col == 0 {
				result.first += filter[c]
			} else if col == 1 {
				result.second += filter[c]
			}
		} else if col == 1 {
			result.count += filter[c]
		}
	}
	result.id = in.id
	result.col = col
	chBuf <- in.buf
	chResult <- result
}

func main() {
	allResults := make([]TempResult, 1024*1024)
	cores := 4
	chBuf := make(chan []byte, cores)
	for i := 0; i < cores; i++ {
		chBuf <- make([]byte, 1024*1024)
	}
	chInput := make(chan Input, cores)
	chResult := make(chan TempResult, cores)
	id := 0
	totalCountChannel := make(chan int)
	go func() {
		for {
			buf := <-chBuf
			len, err := os.Stdin.Read(buf)
			if err != nil {
				break
			}
			chInput <- Input{id, len, buf}
			id++
		}
		totalCountChannel <- id
	}()

	for i := 0; i < cores; i++ {
		go func() {
			for {
				partialCount(<-chInput, chBuf, chResult)
			}
		}()
	}
	count := 0
	consumed := 0
	totalCount := -1
	for {
		var result TempResult
		select {
		case result = <-chResult:
		case totalCount = <-totalCountChannel:
			continue
		}
		rid := result.id
		result.id = 1
		allResults[rid] = result
		count += result.count
		if rid == 0 {
			count += result.second
		} else {
			prev := allResults[rid-1]
			if prev.id == 1 {
				if prev.col == 0 {
					count += result.second
				} else if prev.col == 1 {
					count += result.first
				}
			}
		}
		next := allResults[rid+1]
		if next.id == 1 {
			if result.col == 0 {
				count += next.second
			} else if result.col == 1 {
				count += next.first
			}
		}
		consumed++
		if consumed == totalCount {
			break
		}
	}
	fmt.Println(count)
}

b.rb

path = ARGV[0]
filesize = File.size(path)
cores = 4
pipes = Array.new cores do |id|
  per = (filesize + cores - 1) / cores
  from = per * id
  to = [per * (id + 1), filesize].min
  rio, wio = IO.pipe
  fork do
    file = File.open(path, 'r')
    file.seek from
    length = to - from
    length -= file.gets.size if id != 0
    count = 0
    target = 'aeiou'
    s = ''
    file.each_line do |line|
      length -= line.size
      s << line.split(',', 2).last
      if s.size > 0x10000
        count += s.count target
        s = ''
      end
      break if length < 0
    end
    wio.puts count + s.count(target)
  end
  rio
end
puts pipes.map { |io| io.gets.to_i }.sum

gen.rb

File.write 'data.dat', ([%(abc,atmark,123), %(def,colon,456)] * 100 * 10000).join("\n")

result.txt

% cc -O3 a.c && time ./a.out < data.dat
% go build a.go && time ./a < data.dat
% go build b.go && time ./b < data.dat
% time ./v5.sh < data.dat
% time ruby a.rb < data.dat
% time ruby b.rb data.dat　# これだけSTDINじゃなくfileをコマンドライン引数で渡してる

GO B:   0.13s user 0.01s system 327% cpu 0.043 total 25.1倍
C:      0.04s user 0.01s system  93% cpu 0.053 total 20.4倍
GO A:   0.05s user 0.01s system  93% cpu 0.062 total 17.4倍
RUBY B: 2.51s user 0.07s system 329% cpu 0.782 total 1.38倍
SHELL:  1.86s user 0.03s system 175% cpu 1.079 total 基準
RUBY A: 1.28s user 0.05s system  97% cpu 1.363 total 0.79倍

v5.sh

#!/bin/sh
cut -d, -f2 |tr -dc 'aeiou' |wc -c