Overview

• The primary objective of testing was to evaluate the performance of the Baseline MCPT Environment configuration while attempting to optimize costs by adjusting instance types and reducing the number of instances. The tests were designed to compare the performance outcomes across different configurations, including variations in instance types and counts within multiple Auto Scaling Groups (ASGs). By systematically modifying these variables, the goal was to maintain or improve the performance observed in the baseline configuration while achieving cost efficiency.

PERF-961 - Getting issue details... STATUS  

Summary

• Through a series of experiments involving different placement strategies, instance types, and total instance counts, we found that the performance remained consistent when using these two configuration: 
  • six c7g.large instances (three dedicated to the okapi service and three allocated to mod-courses, mod-sender, mod-tasks-list, mod-gobi, edge-dematic, mod-erm-usage, mod-eusage-reports, mod-notify, and mod-data-import services) alongside ten r7g.xlarge instances for all other services, with the CPU parameter set to 2 for all services.
  • three c7g.large instances dedicated to the okapi service alongside five r7g.2xlarge instances for all other services, with the CPU parameter set to 2 for all services.
• Notably, both environment configuration offers a 40% cost reduction compared to the existing setup, making it a more economical option without compromising on performance.
• Configurations with three c7g.large instances for the okapi service and five r7g.2xlarge instances for all other services show the best performance across all experiments. So it will be more tests in this ticket.
• Tests had 100% errors count for AIE_TC: Create Invoices, AIE_TC: Invoices Approve, AIE_TC: Paying Invoices, TC: Receiving-an-Order-Line, Unreceiving-a-Piece and Unreceiving-a-Piece Workflows because data was not regenerated.  

Test Runs

Test Results

This table contains durations for all Workflows. 

Comparison

This graph shows the durations of all workflows compared between the best test result from Test №8, MCPT Basile configuration with 14 instances m6g.2xlarge and result from mcpt tunning report.

Average Case workflows: Part 1.

Average Case workflows: Part 2.

Average Case workflows: Part 3.

Instance CPU Utilization

Service CPU Utilization

Here we can see that okapi modules used 42k% CPU power parameter CPU=2 for module.

Service Memory Utilization

Here we can't see any sign of memory leaks on every module. Memory shows stable trend.

Kafka metrics

DB CPU Utilization

DB CPU was 99% average with ERW: Exporting Receiving Information.

DB Connections

Max number of DB connections was 5600 in maximum.

Instance CPU Utilization

Service CPU Utilization

Here we can see that okapi used 47k% of the CPU power parameter CPU=2 for module.

Service Memory Utilization

Here we can't see any sign of memory leaks on every module. Memory shows stable trend.

Kafka metrics

DB CPU Utilization

DB CPU was 99% average with ERW: Exporting Receiving Information

DB Connections

Max number of DB connections was 5710.

Instance CPU Utilization

Service CPU Utilization

Here we can see that okapi used 45k% of the CPU power of the parameter CPU=2.

Service Memory Utilization

Here we can't see any sign of memory leaks on every module. Memory shows stable trend.

Kafka metrics

DB CPU Utilization

DB CPU was 99% maximum.

DB Connections

Max number of DB connections was 5500.

Instance CPU Utilization

Service CPU Utilization

Here we can see that okapi used 43k% of the CPU power of parameter CPU=2.

Service Memory Utilization

Here we can't see any sign of memory leaks on every module. Memory shows stable trend.

Kafka metrics

DB CPU Utilization

DB CPU was 99%.

DB Connections

Max number of DB connections was 5810.

DB load

Top SQL-queries

Instance CPU Utilization

Service CPU Utilization

Here we can see that okapi used 46k% of the CPU power of parameter CPU=2.

Service Memory Utilization

Here we can't see any sign of memory leaks on every module. Memory shows stable trend.

Kafka metrics

DB CPU Utilization

DB CPU was 99%.

DB Connections

Max number of DB connections was 5500.

Top SQL-queries

Instance CPU Utilization

Service CPU Utilization

Here we can see that okapi used 46k% of the CPU power of parameter CPU=2.

Service Memory Utilization

Here we can't see any sign of memory leaks on every module. Memory shows stable trend.

Kafka metrics

DB CPU Utilization

DB CPU was 99%.

DB Connections

Max number of DB connections was 6000.

Top SQL-queries

Appendix

Infrastructure

PTF - Baseline MCPT environment configuration

• 14 m6g.2xlarge EC2 instances located in US East (N. Virginia)us-east-1

• 1 database  instance, writer

  Name	Memory GIB	vCPUs
  db.r6g.4xlarge	128 GiB	16 vCPUs

• Open Search ptf-test 
  • Data nodes
    • Instance type - r6g.2xlarge.search
    • Number of nodes - 4
    • Version: OpenSearch_2_7_R20240502
  • Dedicated master nodes
    • Instance type - r6g.large.search
    • Number of nodes - 3
• MSK ptf-KRaft-mode
  
  • 2 brokers, kafka.m7g.xlarge brokers in 2 zones

  • Apache Kafka version 3.7.x 

  • EBS storage volume per broker 300 GiB

  • auto.create.topics.enable=true
  • log.retention.minutes=480
  • default.replication.factor=3

Methodology/Approach

MOBIUS Tests: scenarios were started by JMeter script from load generator. We had 100% error count for AIE_TC: Create Invoices, AIE_TC: Invoices Approve, AIE_TC: Paying Invoices, TC: Receiving-an-Order-Line, Unreceiving-a-Piece and Unreceiving-a-Piece Workflows because data was not regenerated.  

Baseline MCPT Environment configuration according to tunning environment from previous report:  task count: 4 for services: mod-permissions, mod-search, mod-patron, mod-inventory, mod-inventory-storage, mod-circulation, mod-circulation-storage, mod-order, mod-order-storage, mod-invoice, mod-invoice-storage, for mod-users and mod-authtoken task count 6. Parameter srs.marcIndexers.delete.interval.seconds=86400 for mod-source-record-storage. Instance type: m6g.2xlarge. Instances count: 14. Database r6g.4xlarge, Amazon OpenSearch Service  ptf-test: r6g.2хlarge.search (4 nodes).

• Test 1: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all services, used two autoscaling groups, 1st with 3 Instance Type: c7g.large for okapi service and 10 Instance Type: r7g.xlarge for others services. Fixed Load (average case) MOBIUS test was run.
• Test 2: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all services, used two autoscaling groups, 1st with 3 Instance Type: c7g.large for okapi service and 10 Instance Type: m6g.2xlarge for others modules. Fixed Load (average case) MOBIUS test was run.
• Test 3:  The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all services, used two autoscaling groups, 1st with 3 Instance Type: c7g.large for okapi service and 10 Instance Type: m6g.2xlarge for others services. Fixed Load (average case) MOBIUS test was run (Repeat Test 2).
• Test 4: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all modules, used two autoscaling groups, 1st with 6 Instance Type: c7g.large, 3 of them for okapi service and other 3 for mod-cources, mod-sender, mod-tasks-list, mod-gobi, edge-dematic, mod-erm-usage, mod-eusage-reports, mod-notify, mod-data-import services and 10 Instance Type: r7g.xlarge for others modules. Fixed Load (average case) MOBIUS test was run.
• Test 5: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all modules, used two autoscaling groups, 1st with 6 Instance Type: c7g.large, 3 of them for okapi service and other 3 for mod-cources, mod-sender, mod-tasks-list, mod-gobi, edge-dematic, mod-erm-usage, mod-eusage-reports, mod-notify, mod-data-import services and 10 Instance Type: r7g.xlarge for others modules. Fixed Load (average case) MOBIUS test was run (Repeat Test 4).
• Test 6: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all modules, used two autoscaling groups, 1st with 6 Instance Type: c7g.large, 3 of them for okapi service and other 3 for mod-cources, mod-sender, mod-tasks-list, mod-gobi, edge-dematic, mod-erm-usage, mod-eusage-reports, mod-notify, mod-data-import services and 10 Instance Type: r7g.xlarge for others modules. Fixed Load (average case without DE_Exporting MARC Bib records custom workflow, DE_Exporting MARC Bib records workflow, and OPIH_/oai/records workflow) MOBIUS test was run.
• Test 7: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all modules, used two autoscaling groups, 1st with 6 Instance Type: c7g.large, 3 of them for okapi service and other 3 for mod-cources, mod-sender, mod-tasks-list, mod-gobi, edge-dematic, mod-erm-usage, mod-eusage-reports, mod-notify, mod-data-import services and 10 Instance Type: r7g.xlarge for others modules. Fixed Load (average case without DE_Exporting MARC Bib records custom workflow, DE_Exporting MARC Bib records workflow, and OPIH_/oai/records workflow) MOBIUS test was run (Repeat Test 6).
• Test 8: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all modules, used two autoscaling groups, 1st with 6 Instance Type: c7g.large, 3 of them for okapi service and other 3 for mod-cources, mod-sender, mod-tasks-list, mod-gobi, edge-dematic, mod-erm-usage, mod-eusage-reports, mod-notify, mod-data-import services and 10 Instance Type: r7g.xlarge for others modules. Fixed Load (average case) MOBIUS test was run
• Test 9: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all modules, used two autoscaling groups, 1st with 6 Instance Type: c7g.large, 3 of them for okapi service and other 3 for mod-cources, mod-sender, mod-tasks-list, mod-gobi, edge-dematic, mod-erm-usage, mod-eusage-reports, mod-notify, mod-data-import services and 10 Instance Type: r7g.xlarge for others modules. Fixed Load (average case without DE_Exporting MARC Bib records custom workflow, DE_Exporting MARC Bib records workflow, and OPIH_/oai/records workflow) MOBIUS test was run (Repeat Test 7 after Terminate Instances r7g.xlarge).
• Test 10: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all modules, used two autoscaling groups, 1st with 6 Instance Type: c7g.large, 3 of them for okapi service and other 3 for mod-cources, mod-sender, mod-tasks-list, mod-gobi, edge-dematic, mod-erm-usage, mod-eusage-reports, mod-notify, mod-data-import services and 10 Instance Type: r7g.xlarge for others modules. Fixed Load (average case) MOBIUS test was run (Repeat Test 8 after Terminate Instances r7g.xlarge).
• Test 11: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all services, used two autoscaling groups, 1st with 3 Instance Type: c7g.large for okapi service and 11 Instance Type: r7g.xlarge for others services. Fixed Load (average case) MOBIUS test was run.
• Test 12: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all services, used two autoscaling groups, 1st with 3 Instance Type: c7g.large for okapi service and 11 Instance Type: r7g.xlarge for others services. Fixed Load (average case) MOBIUS test was run (Repeat Test 11).
• Test 13: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all services, used two autoscaling groups, 1st with 3 Instance Type: c7g.large for okapi service and 11 Instance Type: r7g.xlarge for others services. Fixed Load (average case) MOBIUS test was run (Repeat Test 11).
• Test 14: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all services, used two autoscaling groups, 1st with 3 Instance Type: c7g.large for okapi service and 11 Instance Type: r7g.xlarge for others services. Fixed Load (average case) MOBIUS test was run (Repeat Test 11 after terminate 11 instances).
• Test 15: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all services, used two autoscaling groups, 1st with 3 Instance Type: c7g.large for okapi and mod-authtoken services and 11 Instance Type: r7g.xlarge for others services. Fixed Load (average case) MOBIUS test was run.
• Test 16: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all services, used two autoscaling groups, 1st with 3 Instance Type: c7g.large for okapi service and 5 Instance Type: r7g.2xlarge for others services. Fixed Load (average case) MOBIUS test was run.
• Test 17: The Baseline MCPT Environment configuration was applied, and CPU=2 was set for all services, used two autoscaling groups, 1st with 3 Instance Type: c7g.large for okapi service and 5 Instance Type: r7g.2xlarge for others services. Fixed Load (average case) MOBIUS test was run (Repeat Test 16).