Home > Storage > PowerFlex > White Papers > MongoDB on Dell PowerFlex with NVMe over TCP > Use case 4: Workload E
Workload E simulates 95% percent scan, and 5 percent insert operations. This workload is focused on sequential read bandwidth. For testing workload E, drop the existing database and reload with workload E. This workload inserts records to increase the size of the database. Twenty million records were scan/inserted on each Mongod VM using the following command. This workload is used to verify the PowerFlex storage bandwidth usage.
# ./bin/ycsb run mongodb -s -threads <number of threa“s> -P "workloads/”orkloade" -p mon”odb.url="mongodb://<hostname/IP>:27018/<database name>?compressors=s”appy&w=0" -p recordcount=20000000 -p operationcount=20000000.
The following table shows the results of the operations across all six Mongod VMs on the PowerFlex two-layer cluster. A cumulative bandwidth of 5.85 GB/s is achieved at submillisecond latency for 64 threads.
Parameters | Results |
Operation Count | 20M records |
YCSB Throughput | 11,284 OPS |
YCSB Insert Avg Latency | 133 us |
YCSB Scan Avg Latency | 35,526 us |
PowerFlex IOPS | 378,500 |
PowerFlex Read Latency | 0.46 millisecond |
PowerFlex Write Latency | 0.64 millisecond |
vCPU Usage | 75% |
vMem Usage | 98% |
Note: The benchmark results depend on the workload type, hardware (VM or host) configuration, specific application requirements, and system design implementation. All performance data in this report was obtained in a rigorously controlled environment. Relative system performance results obtained in other operating environments may vary.