Performance reference for the Azure Storage Table input in the Splunk Add-on for Microsoft Cloud Services¶
The following tables show reference information about performance testing of the Azure storage input in the Splunk Add-on for Microsoft Cloud Services. The testing occurred with version 2.0.0, when the Azure storage input was first introduced. Use this information to enhance the performance of your own Azure storage data collection tasks.
Many factors impact performance results, including file size, file compression, event size, deployment architecture, and hardware. These results represent reference information and do not represent performance in all environments.
Testing architecture¶
Splunk tested the performance of the Storage input using a single-instance Splunk Enterprise 6.4.3 on an C4 High-CPU Double Extra Large instance to ensure CPU, memory, storage, and network do not introduce any bottlenecks. See the following instance specifications:
The EC2 in the testing environment is in the same area of Azure storage input, the network latency is low.
Instance type | C4 High-CPU Double Extra Large |
---|---|
Memory | 15 GB |
Compute Units (ECU) | 31 Units |
Cores | 8 |
Storage Type | GP2(SSD) |
Architecture | 64-bit |
Network performance | High |
EBS Optimized: Max Bandwidth | 1000 Mbps |
Test environment¶
Deployment Type | Role | EC2 Type | Count |
---|---|---|---|
Standalone Deployment | Standalone | C4 High-CPU Double Extra Large | 1 |
Testing result¶
The detailed test result is as follows.
- The input number stands for the number of the inputs, where one input collects one table.
- Each table contains 2,131,200 events.
- Each event is 500 Bytes.
Input number | Throughput (MB/s) | Throughput (GB/day) | Throughput (Event/s) |
---|---|---|---|
1 | 3.44 | 290 | 7045 |
2 | 5.7 | 480 | 11670 |
4 | 6.84 | 577 | 14000 |
8 | 6.12 | 516 | 12533 |
The maximum throughput is 6.8 MB/s with 4 data inputs, which is 100% higher than one data input. The data input can scale by 100% with max throughput 577 GB/day for single instances.