VM Network speeds on same host are really slow. External network to VM are fast.

[CloudIx]

Esxi 5.5

One Host (Dell r710) 2 Nics are connected to Dell switch with LAG set up. One computer also connected to this dell switch with 2 nics and also LAG setup. The vSwitch is setup with 2 NIC and Teaming is setup with "Route based on IP hash" I have 2 Vm's set up with vmxnet3. One VM is a server 2003 (32bit) the other is Server 2008 (32bit).

Using iperf with these arguments on the server "iperf.exe -P -r -s -i 2" and this on the Client "iperf.exe -P -d -r -c 192.168.0.1 -t 60 -i 2"

When i transfer a 6 gig file from either server to the windows 7 machine connected to the dell switch i get about a 675 Mbit/sec rate. "Esxtop n" Shows no dropped packet on the Tx or Rx side.

When I transfer the same file from server to server on the same vSwitch and same subnet i get about 5Mbit/sec rate.

Just to make sure it was not an OS issue, i created a new VM of windows 10 and did the same test. Again a slow rate of about 10Mbit/sec rate. It really seems to me that something is up with the vSwitch.

Also both of these servers was a physical that I converted to a VM. When it was a physical computer network speeds were as expected.

[dhanarajramesh]

May I know how did you configure? configure Enhanced LACP with vSphere Distributed Switches in vSphere 5.5, you must use the vSphere Web Client.  and  Configuring LACP Support in vSphere 5 5 - YouTube

Another enhanced LACP feature in vSphere 5.5 is support for all LACP load balancing types, whereas vSphere 5.1 supports only Source/Destination IP Hash based load balancing.

vSphere 5.5 supports these load balancing types:

1. Destination IP address
2. Destination IP address and TCP/UDP port
3. Destination IP address and VLAN
4. Destination IP address, TCP/UDP port and VLAN
5. Destination MAC address
6. Destination TCP/UDP port
7. Source IP address
8. Source IP address and TCP/UDP port
9. Source IP address and VLAN
10. Source IP address, TCP/UDP port and VLAN
11. Source MAC address
12. Source TCP/UDP port
13. Source and destination IP address
14. Source and destination IP address and TCP/UDP port
15. Source and destination IP address and VLAN
16. Source and destination IP address, TCP/UDP port and VLAN
17. Source and destination MAC address
18. Source and destination TCP/UDP port
19. Source port ID
20. VLAN

[CloudIx]

Not using LACP on the group.. only using LAG, not using a distrubuted vswitch just the standard switch with load balanceing set to "route based on Ip Hash". From the physical switch the team forms and like i said I'm getting good speeds external. Only Internal communication from VM to VM suffers.

[dhanarajramesh]

can you check STP is disabled on both link?

[CloudIx]

STP is disabled. but the external switch should have nothing to do with this problem, since the traffic never leaves the host. When the traffic is leaving the host it is good and speeds are fine. Everything that i have read about a VM to VM copy on the same host is dependent on the resources avail. The server only running the 2 VM's and has plenty of resources. The Datastores the VM's are on is a Direct attached SCSI link.