6 Billing Care REST API Performance

This chapter provides an analysis of the performance of Oracle Communications Billing Care REST API.

The test focuses on the API’s ability to handle concurrent requests and its processing speed. Apache JMeter, a load testing tool, was used to generate and manage the API requests.

Topics in this document:

Scenario 1: Create 100,000 Accounts Using 50 Threads in 2 Schemas

This scenario evaluates the Billing Care REST API’s ability to create 100,000 accounts using 50 threads in 2 schemas.

Table 6-1 shows the client configuration used for the test.

Table 6-1 Scenario 1 Test Client Configuration

Parameter Description
CPU Cores 4 OCPU
Memory 64 GB
Network Bandwidth 8 Gbps

Table 6-2 shows the various parameters used for the test.

Table 6-2 Scenario 1 Test Case and Setup Details

Parameter Description
Account volume 20,000,000
Number of Accounts to Create 100,000
Schemas 2
Threads 50
CM Replicas 10
Oracle DM Replicas 5
Oracle DM2 Replicas 5
OKE Nodes 12 (10 OCPUs each)
Shape VM.Standard.E4.Flex

Scenario 2: Get 100,000 Accounts Using 50 Threads in 2 Schemas

This scenario evaluates the Billing Care REST API’s ability to retrieve 100,000 accounts using 50 threads in 2 schemas.

Table 6-3 shows the client configuration used for the test.

Table 6-3 Scenario 2 Test Client Configuration

Parameter Description
CPU Cores 4 OCPU
Memory 64 GB
Network Bandwidth 8 Gbps

Table 6-4 shows the various parameters used for the test.

Table 6-4 Scenario 2 Test Case and Setup Details

Parameter Description
Account volume 20,000,000
Number of Accounts to Retrieve 100,000
Schemas 2
Threads 50
CM Replicas 10
Oracle DM Replicas 5
Oracle DM2 Replicas 5
OKE Nodes 12 (10 OCPUs each)
Shape VM.Standard.E4.Flex

Performance Results

Table 6-5 shows the performance results for the Billing Care REST API.

Table 6-5 Performance Results

Test Number of Accounts Number of Threads Number of Schemas Throughput (transactions/second) Latency 95th Percentile (ms)
Scenario 1 100,000 50 2 142 423
Scenario 2 100,000 50 2 112 772

See "System Performance Benchmarking Metrics" for more information on how these metrics are calculated.