Comware

Hello,

i want to connect my Server with 4 NICs with my 1910 Switch. I have succesfully created one LACP-Group over 4 Ports at the Switch. I turned the flowcontrol at this Ports on. The Problem is that the traffic isnt evenly distributed over the 4 NICs and i can reach only 1.8Gbit/s. I think the problem is the method which uses the switch. I think i must use RR (Round Robin) for the best Speed and Load Balancing, but i dont know how to configure this. Maybe someone can Help me =)

Thanks

P.S. This thread has been moevd from Switches, Hubs, Modems (Legacy ITRC forum) to Web and Unmanaged - Hp forum moderator

According to the JE009A specs it seems to indicate that you can only have a 2x1Gbit link aggregation: 

Link aggregation: groups together multiple ports (up to a maximum of 2 ports) automatically using Link Aggregation Control
Protocol (LACP), or manually, to form an ultra-high-bandwidth connection to the network backbone; helps prevent traffic bottleneck.

That would account for your 1.8Gbit/s pretty well. 

Hi all! Is there any way to properly configure link aggregation between 2 servers with 2 or more gigabit adapters each with this switch?

I have 2 asus rs-300 e7/ps4 servers with 4xgigabit intel NIC, running with with Windows Server 2012R2. Each server have configured NIC teaming with LACP Dynamic. 

On 1910-48g switch LACP is configured too. Switch says that LACP is configured properly, load balancing works, but data transfer speed is not aggregated between servers. Nic teaming on windows server 2012R2 can use several algoritms, like hashes etc, but none of them aggregate data transfer speed between servers - only 1GBE can be reached. 

Is there any way to tune LACP on switch?

Hi,

on LACP Trunks a connection between two Ip endpoints can only use one interface. So with a connection between two servers you always get the 1G. But if there are 4 individual clients the server would be able to feed all with 1G.....

Thats how I know LACP Trunks (mainly from extreme network switches)..

hth Micha