Packet Core Gateway Throughput and Capacity Metrics
This dataset provides granular, timestamped performance metrics for 4G (PGW/SGW) and 5G (UPF) packet core gateways, including throughput, session counts, setup rates, packet loss, latency, resource utilization, and tunneling efficiency. It enables network operators and engineers to monitor user plane performance, optimize capacity, and troubleshoot issues across multiple gateway types and locations.
Sample rows
preview · 8 of 150 rows · all 14 columns| record_idstring | gateway_typestring | session_setup_ratefloat | notesstring | gateway_idstring | locationstring | timestampdatetime | total_throughput_gbpsfloat | concurrent_sessionsinteger | packet_loss_percentfloat | latency_msfloat | cpu_utilization_percentfloat | memory_utilization_percentfloat | gtpu_tunneling_efficiency_percentfloat |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A001 | PGW | 274.3 | Routine snapshot, healthy metrics | PGW-US-EAST-01 | US-East-1 | 2024-06-01T08:05:00Z | 14.67 | 11245 | 0.07 | 17.6 | 61.2 | 68.4 | 97.2 |
| A002 | UPF | 3150.7 | Boundary test: maximum load | UPF-SG-EDGE-01 | SG-Edge | 2024-06-01T08:10:00Z | 74.55 | 71235 | 0.02 | 7.4 | 98.9 | 94.3 | 99.9 |
| A003 | SGW | 307.5 | Routine snapshot, healthy metrics | SGW-EU-WEST-03 | EU-West-3 | 2024-06-01T08:15:00Z | 17.23 | 17483 | 0.11 | 22.2 | 58.7 | 69.2 | 96.6 |
| A004 | PGW | 401.9 | blank | PGW-CA-TOR-02 | CA-Toronto | 2024-06-01T08:20:00Z | 18.03 | 19841 | 0.09 | 13.9 | 57.1 | 62.7 | 98.1 |
| A005 | SGW | 0 | Gateway offline during maintenance | SGW-US-WEST-01 | US-West-1 | 2024-06-01T08:25:00Z | 0 | 0 | 0 | 0 | 0 | 0 | blank |
| A006 | UPF | 0.03 | Catastrophic failure | UPF-EU-FR-01 | EU-France | 2024-06-01T08:30:00Z | 0.06 | 4 | 99.8 | 9125.7 | 0.4 | 0.7 | 0.2 |
| A007 | PGW | 405.6 | blank | PGW-IN-DEL-01 | IN-Delhi | 2024-06-01T08:35:00Z | 21.43 | 23102 | 0.13 | 15.3 | 70.5 | 66.8 | 99.2 |
| A008 | SGW | 321.5 | blank | SGW-LATAM-BR-02 | LATAM-BR | 2024-06-01T08:40:00Z | 16.84 | 16203 | 0.1 | 19.1 | 64.7 | 70.2 | 97.9 |
| A009 | PGW | 239.8 | Routine snapshot, healthy metrics | PGW-AU-SYD-01 | AU-Sydney | 2024-06-01T08:45:00Z | 11.92 | 9332 | 0.06 | 20.7 | 52.2 | 58.9 | 96.8 |
| A010 | UPF | 2085.2 | blank | UPF-US-EAST-02 | US-East-2 | 2024-06-01T08:50:00Z | 55.32 | 54721 | 0.03 | 11.1 | 97.6 | 92.1 | 99.7 |
| A011 | SGW | 0.01 | Catastrophic failure | SGW-DE-BER-01 | DE-Berlin | 2024-06-01T08:55:00Z | 0.02 | 2 | 100 | 9999.9 | 0 | 0.1 | 0 |
| A012 | PGW | 295.2 | blank | PGW-EU-IT-01 | EU-Italy | 2024-06-01T09:00:00Z | 13.54 | 10254 | 0.08 | 18.9 | 61.8 | 65.5 | 98.4 |
| A013 | UPF | 4821.6 | Boundary test: maximum load | UPF-ZA-JHB-01 | ZA-Johannesburg | 2024-06-01T09:05:00Z | 99.83 | 98593 | 0.01 | 8.6 | 99.3 | 98.2 | 100 |
| A014 | SGW | 428.2 | blank | SGW-IN-MUM-01 | IN-Mumbai | 2024-06-01T09:10:00Z | 19.67 | 18762 | 0.15 | 17.8 | 68.5 | 70.7 | 97.5 |
| A015 | PGW | 312.8 | Routine snapshot, healthy metrics | PGW-EU-SE-01 | EU-Sweden | 2024-06-01T09:15:00Z | 15.23 | 13456 | 0.12 | 16.3 | 54.9 | 60.4 | 98.9 |
| A016 | UPF | 3082.9 | blank | UPF-US-WEST-03 | US-West-3 | 2024-06-01T09:20:00Z | 69.34 | 65872 | 0.04 | 9.3 | 98.6 | 97.4 | 99.6 |
| A017 | SGW | 282.1 | blank | SGW-ME-RIY-01 | ME-Riyadh | 2024-06-01T09:25:00Z | 14.87 | 11632 | 0.08 | 21.4 | 60.8 | 65.2 | 97.7 |
| A018 | PGW | 389.4 | blank | PGW-US-CHI-01 | US-Chicago | 2024-06-01T09:30:00Z | 18.99 | 21035 | 0.11 | 14.7 | 62.6 | 66.3 | 98.3 |
| A019 | UPF | 3962.4 | Boundary test: maximum load | UPF-APAC-TOK-01 | APAC-Tokyo | 2024-06-01T09:35:00Z | 88.45 | 89324 | 0.05 | 10.2 | 98.8 | 99.6 | 99.8 |
| A020 | SGW | 0 | Gateway offline during maintenance | SGW-APAC-SIN-01 | APAC-Singapore | 2024-06-01T09:40:00Z | 0 | 0 | 0 | 0 | 0 | 0 | blank |
What the 150 rows show
from the 150-row sampleUPF (gateway type) stands out: mean session_
- 357.9median session_
setup_ rate - 17.0median total_
throughput_ gbps - 15,967median concurrent_
sessions - 0.08median packet_
loss_ percent - 15.3median latency_
ms - 61.5median cpu_
utilization_ percent
Median 357.9, from 0.0 to 4,822.
- string 5
- integer 1
- float 7
- datetime 1
Columns
14 columns in three groups| column | type | description | example |
|---|---|---|---|
| Text 5 columns | |||
record_id | string | Unique identifier for each metric recordunique | A001 |
gateway_type | string | Type of gateway (PGW, SGW, UPF)PGW · SGW · UPF | PGW |
gateway_id | string | Unique identifier for the specific gateway instance | PGW-US-EAST-01 |
location | string | Physical or logical location of the gateway (e.g., data center, region)optional | US-East-1 |
notes | string | Optional field for comments or additional information about the recordoptional | Catastrophic failure |
| Numbers 8 columns | |||
total_throughput_gbps | float | Total user plane throughput in gigabits per second (Gbps)0 or more | 14.67 |
concurrent_sessions | integer | Number of concurrent user sessions handled by the gateway0 or more | 11245 |
session_setup_rate | float | Rate of new session setups per second0 or more | 274.3 |
packet_loss_percent | float | Percentage of packets lost during transmission0 to 100 | 0.07 |
latency_ms | float | Average user plane latency in milliseconds0 or more | 17.6 |
cpu_utilization_percent | float | CPU utilization as a percentage0 to 100 | 61.2 |
memory_utilization_percent | float | Memory utilization as a percentage0 to 100 | 68.4 |
gtpu_tunneling_efficiency_percent | float | GTP-U tunneling efficiency as a percentage0 to 100 · optional | 97.2 |
| Dates and times 1 column | |||
timestamp | datetime | Date and time when the metrics were collected | 2024-06-01T08:05:00Z |
Use it for
A telecommunications dashboard
Session_
setup_ rate by gateway_ type and a breakdown of notes. Excel, Power BI or Tableau. Why do the 47 UPF rows have a mean session_
setup_ rate of 2,347? A root-cause class exercise
Hand out the rows and one question. The answer is in the data, not in the brief.
- Records150A001274.3PGWA0023150.7UPFA003307.5SGW
A software demo
Believable records with gateway_
type, gateway_ id and location to fill a screen in front of a buyer.
This dataset150 rows14 columns
Yours1,000 rows14 columnslocation: UK only
blueprint · packet-core-gateway-throughput-and-capacity-metrics
Behind this dataset
Same schema. As many rows as you need.
These 150 rows came out of a blueprint — 14 columns with generation rules behind each one. Open it in Data Factory to retune a column, add your own, wire in foreign keys, and run it at the size you actually need.
Rules it was built with
- Gateway instance ID and deployment (physical, virtual, cloud-native)
- Total throughput: uplink and downlink Gbps aggregated
- Concurrent active sessions (bearers in 4G, PDU sessions in 5G)
- Session setup rate: sessions per second (SPS)
- Packet loss percentage (<0.1% target for carrier-grade)
- End-to-end latency (UE to internet gateway) in milliseconds
- CPU and memory utilization percentage for capacity planning
- GTP-U encapsulation overhead and tunneling efficiency
Open the blueprint in Data Factory
1 credit per row. Preview 20 rows free. Buy rows for the full run.
- Exports
- CSV, JSON, JSONL, Parquet, SQL, Excel, TSV, XML
- Licence
- yours to use, including commercially
- API slug
- packet-core-gateway-throughput-and-capacity-metrics