Service aware load imbalance detection and root cause identification
A system can include a network analysis platform that applies performance models to determine if a load imbalance exists at a cell, such as at a base station. The performance models are pre-trained based on network telemetry data. For a session at a cell, an expected load can be compared to an actual load to determine whether the session is impacted by a load imbalance. If the number of impacted sessions exceeds a threshold, the base station can be highlighted on a GUI. Additionally, the network analysis platform can perform root cause analysis of a victim cell based on session handoff analysis to determine how to decrease the imbalance impacts.
1. A method for identifying load imbalances in a telco network, comprising:
receiving telemetry data;
determining an actual throughput for a first session at a first base station of a plurality of base stations;
predicting an expected throughput for the first session based on a normalized load at the first base station, wherein the normalized load relates to loads across the plurality of base stations;
determining that the expected throughput exceeds the actual throughput by at least a threshold amount; and
indicating, on a graphical user interface (“GUI”), a load imbalance exists with respect to the first base station.
2. The method of claim 1 , wherein the normalized load is determined based on features corresponding to at least one of: a utilization level of a downlink control channel across the plurality of base stations, an average number of users waiting in a queue of any of the plurality of base stations, a fraction of resources available on average to a user, an average utilization of a data channel of the base station, and a number of physical resource blocks available, on average, at the plurality of base stations.
3. The method of claim 1 , further comprising:
determining that a number outgoing of session handoffs from the base station exceeds a number of incoming session handoffs to the base station; and
indicating that improved load balancing for the base station can be achieved by at least one of offloading additional traffic from the base station and reconfiguring load balancing parameters for the base station.
4. The method of claim 1 , further comprising:
determining that a number of incoming session handoffs to the base station exceeds a number of outgoing session handoffs from the base station; and
grouping the incoming session handoffs as triggered by a vendor or based on a Long Term Evolution (LTE) standard.
5. The method of claim 4 , wherein, based on a majority of the incoming session handoffs being grouped as triggered by a vendor, indicating that the vendor's load balancing parameters is a root cause.
6. The method of claim 4 , wherein, based on a majority of the incoming session handoffs being grouped as based on the LTE standard, indicating that a coverage footprint of the base station is a root cause.
7. The method of claim 1 , wherein a same performance model is used to determine actual throughput and predict the expected throughput.
8. A non-transitory, computer-readable medium containing instructions that, when executed by a hardware-based processor, performs stages for identifying load imbalances in a telco network, the stages comprising:
receiving telemetry data;
determining an actual throughput for a first session at a first base station of a plurality of base stations;
predicting an expected throughput for the first session based on a normalized load at the first base station, wherein the normalized load relates to loads across the plurality of base stations;
determining that the expected throughput exceeds the actual throughput by at least a threshold amount; and
indicating, on a graphical user interface (“GUI”), a load imbalance exists with respect to the first base station.
9. The non-transitory, computer-readable medium of claim 8 , wherein the normalized load is determined based on features correspond to at least one of: a utilization level of a downlink control channel across the plurality of base stations, an average number of users waiting in a queue of any of the plurality of base stations, a fraction of resources available on average to a user, an average utilization of a data channel of the base station, and a number of physical resource blocks available, on average, at the plurality of base stations.
10. The non-transitory, computer-readable medium of claim 8 , the stages further comprising:
determining that a number outgoing of session handoffs from the base station exceeds a number of incoming session handoffs to the base station; and
indicating that improved load balancing for the base station can be achieved by at least one of offloading additional traffic from the base station, and reconfiguring load balancing parameters for the base station.
11. The non-transitory, computer-readable medium of claim 8 , the stages further comprising:
determining that a number of incoming session handoffs to the base station exceeds a number of outgoing session handoffs from the base station; and
grouping the incoming session handoffs as triggered by a vendor or based on a Long Term Evolution (LTE) standard.
12. The non-transitory, computer-readable medium of claim 11 , wherein, based on a majority of the incoming session handoffs are grouped as triggered by a vendor, indicating that the vendor's load balancing parameters is a root cause.
13. The non-transitory, computer-readable medium of claim 11 , wherein, based on a majority of the incoming session handoffs are grouped as based on the LTE standard, indicating that a coverage footprint of the base station is a root cause.
14. The non-transitory, computer-readable medium of claim 8 , wherein a same performance model is used to determine actual throughput and predict the expected throughput.
15. A system for identifying load imbalances in a telco network, comprising:
a memory storage including a non-transitory, computer-readable medium comprising instructions; and
a computing device including a hardware-based processor that executes the instructions to carry out stages comprising:
receiving telemetry data;
determining an actual throughput for a first session at a first base station of a plurality of base stations;
predicting an expected throughput for the first session based on a normalized load at the first base station, wherein the normalized load relates to loads across the plurality of base stations;
determining that the expected throughput exceeds the actual throughput by at least a threshold amount; and
indicating, on a graphical user interface (“GUI”), a load imbalance exists with respect to the first base station.
16. The system of claim 15 , wherein the normalized load is determined based on features correspond to at least one of: a utilization level of a downlink control channel across the plurality of base stations, an average number of users waiting in a queue of any of the plurality of base stations, a fraction of resources available on average to a user, an average utilization of a data channel of the base station, and a number of physical resource blocks available, on average, at the plurality of base stations.
17. The system of claim 15 , the stages further comprising:
determining that a number outgoing of session handoffs from the base station exceeds a number of incoming session handoffs to the base station; and
indicating that improved load balancing for the base station can be achieved by at least one of offloading additional traffic from the base station, and reconfiguring load balancing parameters for the base station.
18. The system of claim 15 , the stages further comprising:
determining that a number of incoming session handoffs to the base station exceeds a number of outgoing session handoffs from the base station; and
grouping the incoming session handoffs as triggered by a vendor or based on a Long Term Evolution (LTE) standard.
19. The system of claim 18 , wherein, based on a majority of the incoming session handoffs are grouped as triggered by a vendor, indicating that the vendor's load balancing parameters is a root cause.
20. The system of claim 18 , wherein, based on a majority of the incoming session handoffs are grouped as based on the LTE standard, indicating that a coverage footprint of the base station is a root cause.