4.4 Performance Benchmarking with Synthetic Feed with Replication in OCI Environment

The performance benchmarking test is performed on OCNADD with 5K MPS traffic, which includes 3K MPS SCP traffic, 1K MPS NRF traffic, and 1K MPS SEPP traffic, in a centralized deployment with one worker group. The benchmarking results are provided below:

Note:

An ingress message from a NF is '1' MPS for OCNADD.

Benchmark Test Environment

The test environment specifications are listed below:
  • Ingress Traffic Rate: 5K MPS
  • The latency, resource utilization, and MPS is observed for the period of 3 hrs or more.
  • OCNADD single site deployment with ASM disabled
  • OCNADD Release 24.1.0 deployed over OCI.
  • Kafka Replication Factor (RF): 1
  • OCNADD TCP Feed: TCP/TLS
  • OCNADD Kafka PVC: 50GB
  • 3rd Party Application: OCNADD Simulator with one endpoint
  • cnDBTier Release 24.1.0
  • NRF Release: NA
  • SCP Release: NA
  • SEPP Release: NA
  • SASL/SSL, enabled between OCNADD Simulator and OCNADD
  • CNCC Release 24.1.0
  • Message Size: 3330 Bytes
  • Environment: OCI environment, "6" Worker Nodes, "VM.Standard.E4.Flex" Instance Shape, "194 GB" memory in Worker Node, "50 (CPU: 100)" OCPUs in the Worker Nodes
  • Execution time: 16 Hours

Benchmark Testcase Specifications

The testcase parameters are as follows:
  • OCNADD
    • Message Ingestion Rate: 5K MPS
    Resource Specifications:

    Table 4-10 OCNADD Worker Group Resource Specifications

    Services CPU Request Per Pod CPU Limit Per Pod Memory Request Per Pod (Gi) Memory Limit Per Pod (Gi) min Replicas max Replicas Kafka Topic Partitions and Retention
    ocnaddconfiguration 1 1 2 2 1 1 -
    ocnaddalarm 1 1 2 2 1 1 -
    ocnaddadmin 1 1 2 2 1 1 -
    ocnaddhealthmonitoring 1 1 2 2 1 1 -
    ocnadduirouter 0 0 0 0 0 0 -
    ocnaddscpaggregation 2 2 4 4 1 1

    6 (SCP)/

    retention.ms = 300000
    ocnaddnrfaggregation 2 2 4 4 1 1

    6 (NRF)/

    retention.ms = 300000
    ocnaddseppaggregation 2 2 4 4 1 1

    12 (SEPP)/

    retention.ms = 300000
    ocnaddadapter 3 3 5 5 2 2

    12 (MAIN)/

    retention.ms = 300000
    ocnaddkafka 4 4 32 32 4 4 -
    zookeeper 1 1 3 3 3 3 -
    ocnaddgui 0 0 0 0 0 0 -
  • SCP

    SCP Traffic: 3K TPS in a centralized deployment having one worker group.

  • NRF

    NRF Traffic: 1K TPS in a centralized deployment having one worker group.

  • SEPP

    SEPP Traffic: 1K TPS in a centralized deployment having one worker group.

Benchmark Test Results

Traffic Feed Details

Table 4-11 Traffic Feed Details

NF NF Traffic Copy Trigger Points Traffic Rate Duration in hours E2E Traffic Feed Average Latency E2E Traffic Feed Success Rate
OCNADD Worker Group NA 5K MPS 16

Feed-1: 5 msec

Feed-2: 4 msec

99.99%
NRF 2 - IGW/EGW (Request and Response) 1K MPS 16 NA NA
SCP 2 (Request Ingress/Request Egress) 3K MPS 16 NA NA
SEPP 2 (PLMN IGW) (Request and Response) 1K MPS 16 NA NA

CPU and Memory Utilization

The following table describes the OCNADD CPU and memory utilization:

Table 4-12 OCNADD CPU and Memory Utilization

Microservice/ Container Pod Count CPU Utilization (%) Memory Utilization (%)
ocnaddadminservice 1 54.764% 53.423%
ocnaddalarm 1 0.205% 60.424%
ocnaddconfiguration 1 0.214% 64.145%
ocnaddhealthmonitoring 1 0.275% 54.615%
kafka-broker 4 74.323% 59.958%
zookeeper 3 0.430052465% 53.955%
ocnaddnrfaggregation 1 40.833% 78.980%
ocnaddscpaggregation 1 60.754% 82.635%
ocnaddseppaggregation 1 44.205% 84.717%