Home > Storage > PowerFlex > White Papers > Dell APEX Block Storage for AWS: MongoDB using Kubernetes and PowerFlex CSI > Use case 3: Workload A
Workload A simulates 50 percent read, and 50 percent write operations. Ten million records were written and read from each MongoDB sharded pod. The YCSB-mongo client can read/write a subset of the data and was accomplished by changing the number of threads and the requestdistribution=sequential option:
kubectl -n mongo exec $(kubectl get pod -n mongo | grep YCSB-pod1 | awk '{print $1}') -- \ ./bin/YCSB load mongodb -s -threads 32 \
-P "workloads/workloada" \ -p mongodb.url=mongodb://mongodb:*****@10.244.3.9:27017/YCSB?w=1 2>&1 \
-p recordcount=100000000 \
-p operationcount=100000000 \
-p insertproportion=1 \
-p requestdistribution=sequential \
-p mongodb.writeConcern=acknowledged \
-p writeallfields=true | tee -a /home/ec2-user/mongodblogs/10m_32_workloada_s0n1 &
The following figure shows the results of (50 percent read, 50 percent write) operations across all eight MongoDB sharded pods on PowerFlex two-layer cluster on the AWS public cloud. A cumulative throughput of 571,153 ops/s is achieved at a submillisecond latency for 32 threads on the MongoDB sharded cluster.
Note: Benchmark results depend on the workload type, AWS Instance type, placement of instances in a multi-AZ 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 might vary.