| Table of Contents |
|---|
...
- Load tests showed that ListApp refreshing duration is
- 1.5 min for 3 users test (1 user on each of three tenants);
- 2.3 min min for 10 users test (3-4 user on each of three tenants).
- Load test for 30 users (10 users per tenant) failed due to DB overload.
- Load tests showed that ListApp refreshing duration increases 2-3 times when conducted with the same users user number but on 1 tenant instead of 3 tenants. For example, during 3 users test duration increased from 1.5 min to 2.8 min, for 10 users - from 2.3 min to 6.1 min.
- During the 10 users test CPU utilization reached 200% for for mod-fqm-manager and 111% for mod-lists. Also, mod-permissions CPU utilization exceeded 100% during 30 users test.
- Maximum DB CPU utilization reached 83% (writer instance) and 99% (reader instance) during 30 users user test.
- Memory utilization for mod-permissions increased from 48% to 76% during the tests. No memory leak is suspected for all of the modules.
Test runs
Query used in lists - "Item status != Available". List refresh result is about 200K records.
Scenario | Data quantity |
|---|---|
List App 3 users refresh multiple tenants | tenant 1 - 1 user tenant 2 - 1 user tenants 3 - 1 user |
List App 3 users refresh single tenant* | tenant 1 - 3 users |
List App 10 users refresh multiple tenants | tenant 1 - 3 users tenant 2 - 3 users tenants 3 - 4 users |
List App 10 users refresh single tenant* | tenant 1 - 10 users |
List App 30 users refresh multiple tenants | tenant 1 - 10 users tenant 2 - 10 users tenants 3 - 10 users |
*additional Additional tests run to analyze dependancy dependency between tenant quantity and ListApp refresh duration
...
| Transaction | Duration, avg | Release | TnenantsTenants | Number of users | R/W split | Other conditions |
|---|---|---|---|---|---|---|
ListApp refresh | 10 min 40 sec | [Orchid] | 1 tenant | 10 users | disabled | |
| 8.5 min | [Poppy] | 1 tenant | 10 users | disabled | ||
| 17.7 min | [Poppy] | 1 tenant | 10 users | disabled | Testing in parallel with DI and CICO | |
| ListApp refresh current test results** | 1.5 min | [Poppy] | 3 tenants | 3 users | enabled | |
| 2.8 min | [Poppy] | 1 tenant | 3 users | enabled | ||
| 2.3 min | [Poppy] | 3 tenants | 10 users | enabled | ||
| 6.1 min | [Poppy] | 1 tenant | 10 users | enabled | ||
| Server error | [Poppy] | 3 tenants | 30 users | enabled |
...
During the 10 users test CPU utilization reached 200% for for mod-fqm-manager and 111% for mod-lists. Also, mod-permissions CPU utilization exceeded 100% during 30 users test.
...
Memory utilization for mod-permissions increased from 48% to 76% during the tests. No memory leak is suspected for all of the modules.
*results for test with 3-30 users, 3 tenants, R/W split enabled
...
Maximum DB CPU utilization reached 83% (writer instance) and 99% (reader instance) during 30 users user test.
*results for test with 3-30 users, 3 tenants, R/W split enabled
...
Infrastructure
PTF -environment pcp1
- 10 m6i.2xlarge EC2 instances located in US East (N. Virginia)us-east-1
1 database instance, writer
Name API Name Memory GIB vCPUs max_connections R6G Extra Large db.r6g.xlarge 32 GiB 4 vCPUs 2731 - MSK tenant
- 4 m5.2xlarge brokers in 2 zones
Apache Kafka version 2.8.0
EBS storage volume per broker 300 GiB
- auto.create.topics.enable=true
- log.retention.minutes=480
- default.replication.factor=3
...