Hi All 5G guys:
What is advantage of split bearer at PDCP layer if we don’t use it?
Is there any negative impact on 5G?
Hi All 5G guys:
If you didn’t use split bearer at PDCP, that means lower throughput at 5G side as the throughput in this case will be the data transferred over 5G air interface only.
While using split bearer, throughput will be the sum of both legs 4G+5G.
Thanks a lot!
Actually I’m reading split happen at condition of Throughput parameter once condition reached for configured parameter.
But I’ve seen in one of 5G book that traffic flow happen on QCI configured at 5G end.
Not able to seen any traffic at 4G end parallely.
Test is done with single UE.
May be split parameter differ vendor to vendor.
Any expert from Ericsson, Huawei can please confirm how split bearer happening in their system.
So, that means you need to revise all parameters needed to activate split bearer from both 4G and 5G sides.
Also, assuming all of the above parameters are ok, you need to revise the split bearer parameters needed to to trigger the split bearer process as it differs from one vendor to another.
Yes agree thanks
I’m checking if is there some counters indicate split bearer working or not on site.
Yes, there are some counters, but of course differ from one vendor to another again.
Even these counters are used to measure the percentage of traffic split between 4G and 5G.
To know which leg is utilized more from 5G point of view.
Ok let me check
As far as I am aware All vendors have Split bearer configured.
But data flow is on 5G channel until there is an overflow then PDCP split happens and overflow data comes over LTE.
For example if I am trying to do iPerf of 2Gbps and NR supports max throughput of 1.8 Gbps then the remaining 200 Mbps comes over LTE.
Any data throughput that comes under 1.8 Gbps will be on NR leg.
This is config deployed by All US operators by any vendor.
Initial trigger for bearer split typically occurs if the data in the buffer is above a threshold. Afterwards, the amount of data split depends on how fast each leg serves (highly dependent on total BW on each leg) with goal of minimizing the reordering of packets at UE side