Problematic NFS performance when fseek() is used in the client code - c

Question

I am developing a simple parallel application using MPI that involves the loading of a file to memory. That file is exported via NFS to the nodes of the computer cluster. I've noticed that in some cases the performance of NFS drops significantly with thousands of additional TCP packets being trasmitted from the server to the clients and i've pinpointed the problem to the use of fseek() in the code:
//Seek to data and load them to array
fseek ( fp, ( unsigned int ) dec_number + start, SEEK_SET );
for ( i = 0; i < n * mpi_n; i++ ) {
if ( ! feof ( fp ) )
text[i] = fgetc ( fp );
if ( i > 0 && n > mpi_n && i % mpi_n == 0 )
fseek ( fp, n - mpi_n, SEEK_CUR );
}
fclose ( fp );
Since the same code without the fseek() works without problems, is it possible that the server actually resends parts of the file after each fseek() ? How can this performance be improved?
Time with cold NFS cache, without fseek(): ~4 sec
Time with hot NFS cache, without fseek(): ~3 sec
Time with cold NFS cache, with fseek(): ~12 sec
Time with hot NFS cache, with fseek(): ~3 sec
Snapshot of nfswatch with a cluster of 10 nodes, a 300MB file with cold NFS cache and with fseek():
Total packets:
1903459 (network) 544803 (to host) 0 (dropped)
Packet counters:
NFS3 Read: 116290 21%
NFS3 Write: 10 0%
NFS Read: 0 0%
NFS Write: 0 0%
NFS Mount: 0 0%
Port Mapper: 0 0%
RPC Authorization: 29 0%
Other RPC Packets: 0 0%
TCP Packets: 544386 100%
UDP Packets: 17 0%
ICMP Packets: 0 0%
Routing Control: 0 0%
Address Resolution: 0 0%
Reverse Addr Resol: 0 0%
Ethernet Broadcast: 0 0%
Other Packets: 49 0%
Snapshot of nfswatch with a cluster of 10 nodes, a 2GB file with cold NFS cache and without fseek():
Total packets:
251804 (network) 102650 (to host) 0 (dropped)
Packet counters:
NFS3 Read: 37039 36%
NFS3 Write: 1 0%
NFS Read: 0 0%
NFS Write: 0 0%
NFS Mount: 0 0%
Port Mapper: 0 0%
RPC Authorization: 2 0%
Other RPC Packets: 0 0%
TCP Packets: 102543 100%
UDP Packets: 30 0%
ICMP Packets: 1 0%
Routing Control: 0 0%
Address Resolution: 0 0%
Reverse Addr Resol: 0 0%
Ethernet Broadcast: 0 0%
Other Packets: 41 0%
The clients are mounted using the following mount command:
/nfs on /nfs type nfs (rw,rsize=8192,wsize=8192,timeo=14,intr)