ich habe bei meinem ESXi 2 iSCSI Laufwerke eingebunden.
Netzwerkuplinks sind beides 1GBit, jedoch ist meine Durchsatzrate eher schwach:
Code: Alles auswählen
iSCSI Target1 : Iomega StorCenter
/vmfs/volumes # vmkping -s 9000 192.168.3.123
PING 192.168.3.123 (192.168.3.123): 9000 data bytes
9008 bytes from 192.168.3.123: icmp_seq=0 ttl=64 time=0.689 ms
9008 bytes from 192.168.3.123: icmp_seq=1 ttl=64 time=0.672 ms
9008 bytes from 192.168.3.123: icmp_seq=2 ttl=64 time=0.736 ms
--- 192.168.3.123 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 0.672/0.699/0.736 ms
Results for write throughput:
119.304 Mbit/s with udp,async,wsize=8192,blocksize=8
119.304 Mbit/s with tcp,async,wsize=1024,blocksize=1
107.374 Mbit/s with udp,async,wsize=32768,blocksize=32
107.374 Mbit/s with udp,async,wsize=2048,blocksize=2
107.374 Mbit/s with udp,async,wsize=16384,blocksize=16
107.374 Mbit/s with udp,async,wsize=1024,blocksize=1
107.374 Mbit/s with tcp,async,wsize=8192,blocksize=8
107.374 Mbit/s with tcp,async,wsize=4096,blocksize=4
107.374 Mbit/s with tcp,async,wsize=32768,blocksize=32
97.612 Mbit/s with udp,async,wsize=4096,blocksize=4
97.612 Mbit/s with tcp,async,wsize=2048,blocksize=2
97.612 Mbit/s with tcp,async,wsize=16384,blocksize=16
Results for read throughput:
178.956 Mbit/s with tcp,async,rsize=32768,blocksize=32
153.391 Mbit/s with udp,async,rsize=32768,blocksize=32
153.391 Mbit/s with udp,async,rsize=16384,blocksize=16
153.391 Mbit/s with tcp,async,rsize=2048,blocksize=2
153.391 Mbit/s with tcp,async,rsize=16384,blocksize=16
134.217 Mbit/s with udp,async,rsize=8192,blocksize=8
134.217 Mbit/s with udp,async,rsize=2048,blocksize=2
134.217 Mbit/s with udp,async,rsize=1024,blocksize=1
134.217 Mbit/s with tcp,async,rsize=8192,blocksize=8
134.217 Mbit/s with tcp,async,rsize=1024,blocksize=1
119.304 Mbit/s with tcp,async,rsize=4096,blocksize=4
71.582 Mbit/s with udp,async,rsize=4096,blocksize=4
Code: Alles auswählen
iSCSI Target2 : debian 2.6.26-2-686 #1 SMP Thu Nov 25 01:53:57 UTC 2010 i686 GNU/Linux
/vmfs/volumes # vmkping -s 9000 192.168.1.139
PING 192.168.1.139 (192.168.1.139): 9000 data bytes
9008 bytes from 192.168.1.139: icmp_seq=0 ttl=64 time=0.477 ms
9008 bytes from 192.168.1.139: icmp_seq=1 ttl=64 time=0.494 ms
9008 bytes from 192.168.1.139: icmp_seq=2 ttl=64 time=0.492 ms
--- 192.168.1.139 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 0.477/0.488/0.494 ms
Results for write throughput:
82.595 Mbit/s with udp,async,wsize=8192,blocksize=8
82.595 Mbit/s with udp,async,wsize=32768,blocksize=32
82.595 Mbit/s with udp,async,wsize=16384,blocksize=16
76.695 Mbit/s with udp,async,wsize=2048,blocksize=2
76.695 Mbit/s with tcp,async,wsize=16384,blocksize=16
71.582 Mbit/s with tcp,async,wsize=8192,blocksize=8
71.582 Mbit/s with tcp,async,wsize=4096,blocksize=4
71.582 Mbit/s with tcp,async,wsize=32768,blocksize=32
67.108 Mbit/s with udp,async,wsize=1024,blocksize=1
63.161 Mbit/s with tcp,async,wsize=1024,blocksize=1
53.687 Mbit/s with udp,async,wsize=4096,blocksize=4
53.687 Mbit/s with tcp,async,wsize=2048,blocksize=2
Results for read throughput:
134.217 Mbit/s with udp,async,rsize=32768,blocksize=32
134.217 Mbit/s with tcp,async,rsize=8192,blocksize=8
134.217 Mbit/s with tcp,async,rsize=4096,blocksize=4
134.217 Mbit/s with tcp,async,rsize=32768,blocksize=32
134.217 Mbit/s with tcp,async,rsize=16384,blocksize=16
134.217 Mbit/s with tcp,async,rsize=1024,blocksize=1
119.304 Mbit/s with udp,async,rsize=8192,blocksize=8
119.304 Mbit/s with udp,async,rsize=4096,blocksize=4
119.304 Mbit/s with udp,async,rsize=2048,blocksize=2
119.304 Mbit/s with udp,async,rsize=16384,blocksize=16
119.304 Mbit/s with udp,async,rsize=1024,blocksize=1
119.304 Mbit/s with tcp,async,rsize=2048,blocksize=2
Ist das Nadelöhr in der Konfiguration oder in der Hardwareausstattung zu finden?