Below I have write my awk script to calculate throughput.
The columns that you need are:
column 7: TCP bytes sent
column 8: interval
then the throughput is
column7/column8
And, this script also calculates: bytes drops, packets drops, bytes
arrivals als packets arrivals.
I'm sorry, but the names of variables are in Spanish.
/salva
#:::::::::::::::::::::::::::::::::::::::::::::::::::::::::
echo "[$1:$2:$3] ...Obteniendo bytes/packets/flows sent/dropped TCP "
cat $TMP/out.tr | perl column 3 0 2 1 | sort -g |
awk '
BEGIN { max = 0 }
{
# i = fid
# flujo[i,1] = packets_arrivals (+)
# flujo[i,2] = packets_drops (d)
# flujo[i,3] = packets_sent (-)
# flujo[i,4] = bytes_arrivals
# flujo[i,5] = bytes_drops
# flujo[i,6] = bytes_sent
# flujo[i,7] = instante inicio
# flujo[i,8] = instante final
i = $1;
tiempo = $4;
evento = $2;
max = ((i > max) ? i : max);
contabilizar = 0;
if (evento == "+") {
flujo[i,1]++;
flujo[i,4] += $3;
contabilizar = 1;
} else if (evento == "d") {
flujo[i,2]++;
flujo[i,5] += $3;
contabilizar = 1;
} else if (evento == "-") {
flujo[i,3]++;
flujo[i,6] += $3;
contabilizar = 1;
}
if (contabilizar == 1) {
tiempo = $4;
if (flujo[i,7] == 0) { flujo[i,7]= tiempo };
if (flujo[i,8] == 0) { flujo[i,8]= tiempo };
if ( tiempo < flujo[i,7]) { flujo[i,7] = tiempo};
if ( tiempo > flujo[i,8]) { flujo[i,8] = tiempo};
}
}
END {
for (i=0; i<=max ; i++) {
lapsus = flujo[i,8] - flujo[i,7];
if ( lapsus > 0 ) { printf("%d\t%d\t%d\t%d\t%d\t%d\t%d\t%f\n",i,flujo[i,1],flujo[i,2],flujo[i,3],flujo[i,4],flujo[i,5],flujo[i,6],lapsus) }
}
}' > $TMP/tcp_flows.dat
#:::::::::::::::::::::::::::::::::::::::::::::::::::::::::
#::: FIN
#:::::::::::::::::::::::::::::::::::::::::::::::::::::::::
On Mon, 26 Mar 2007, paul yoong wrote:
>
> Hi all,
>
> I want to calculate the throughput of a link using awk. Can anyone tell me how to do this for a simple 2 node link. If anyone could direct me to any example on the web that would be great.
>
> Thanks
출처 : http://aza97.egloos.com/
Comments List
移
嫄몄