The test results are based on a model in which the constant-bandwidth safety and security video traffic remained unaffected during a cluster normal load baseline, disk rebuild, and multiple iterations of node reboots.
We performed all tests as follows:
- With multiple iterations of node reboots.
- With drive failures in place in the A300 cluster to ensure a worst-case scenario for all sizing parameters.
- With a per-camera bandwidth of 6.24 Mb/s.
The following table provides bandwidth-sizing guidelines based on our test results.
Configuration
Component | Details |
VMS | Intellicene Symphia VMS |
Version | 8.0.12315.0 |
Component | Bandwidth MB/s (Mb/s) |
Per Cluster Bandwidth | 754 (6325) |
Per Node Bandwidth | 188 (1577) |
Per Recorder | 62.8 (526) |
Per Cluster Bandwidth (mission critical) | 648 (5435) |
Per Node Bandwidth (mission critical) | 162 (1359) |
Number of Recorders per Node | 3 |
Number of Recorders per cluster | 12 |
Component | Details |
Server Model | PowerEdge R650 |
Hypervisor version | VMware ESXi, 8.0.1 |
Processor Model | Intel(R) Xeon(R) Gold 5318Y @ 2.1GHz |
Number of Processors | 2 |
Cores per socket | 24 |
Logical Processors | 96 |
Memory | 256 |
Disk controller | PERC H755 Front (Embedded) RAID Controller |
VMS Storage Tier | N/A |
Component | Details |
Compute platform | VMware ESXi |
Total No. of VMs per ESXi host | 4 |
Guest OS | Windows Server 2019 Standard |
vCPU | 12 |
vMemory | 30 GB |
vNIC type | VMXNET3 |
Video disk size | 20 TB |
Component | Details |
Storage | PowerScale A300 |
Nodes per cluster | 4 |
OneFS | 9.5.0.2 |
Disk type | SATA HDD |
Protection scheme | + 2d:1n |
Disks per node | 20 |
Disk capacity | 12 TB |
Protocol | SMB |
CA enabled | Yes |
VMS Storage Tier | 1 |
Component | Details |
Camera type | Virtual Camera |
Number of cameras | 70 |
Codec | H264 |
Bitrate | 6.2 Mbps |
FPS | 9 |
Resolution | 2560x1920 |
Component | Category | Details |
ESXi host (%) | Max. CPU utilization | 45% |
Avg. CPU utilization | 48% | |
Max memory utilization | 58% | |
Per VM (%) | Max. CPU utilization | 51% |
Avg. CPU utilization | 48% | |
Max. memory utilization | 16% | |
Windows (%) | Write IO size | 971 KB |
Read IO size | 211 kB | |
Write IOPS | 56 | |
Read IOPS | 48 | |
Disk Rebuild | Tier 1 | N/A |
Mission-critical graphs
The disk throughput rate displays the total amount of data being read from and written to the disks in the cluster.
The external network throughput rate displays the total amount of data that passes through the external network interfaces in the monitored cluster.
The CPU utilization percentage averages 58 percent during failure and recovery scenarios, such as Smartfail and adding the failed node back in.
The disk throughput rate displays the total amount of data being read from and written to the disks in the cluster.
Normal baseline graphs
The disk throughput rate displays the total amount of data being read from and written to the disks in the cluster.
The external network throughput rate displays the total amount of data that passes through the external network interfaces in the monitored cluster.
The CPU utilization percentage averages 60 percent during failure and recovery scenarios, such as Smartfail and adding the failed node back in.
The disk throughput rate displays the total amount of data being read from and written to the disks in the cluster.