Test Method

Test Environment

• Region: AP-Singapore
• AZ: AZ1
• Elastic Cloud Server (ECS):
  • For a general-purpose DB instance, select an ECS with s3.2xlarge.2 flavor with 8 vCPUs, 16 GB of memory, and CentOS7.5 64 bit image.
  • For an enhanced II DB instance, select an ECS with c3.2xlarge.2 flavor with 8 vCPUs, 16 GB of memory, and CentOS 7.6 x86_64 bit image.
• DB instances to be tested:
  • The cluster instance has two mongos nodes and two shard nodes, and the storage space is 600 GB.
  • The replica set instance has three nodes, and the storage space is 600 GB.

Test Tool

YCSB is an open-source database performance test tool. In this test, YCSB 0.17.0 is used.

For details on how to use this tool, see YCSB.

Test Metrics

Operations per Second (OPS): number of operations executed per second by a database

Test Procedure

1. Use YCSB to run the following command to load the test data:

   ./bin/ycsb load mongodb -s -p workload=site.ycsb.workloads.CoreWorkload -p recordcount=${recordcount} -p mongodb.url="mongodb://${userName}:${password}@${mongosIP}:${port}/ycsb?authSource=admin" -threads ${threadNum}

   

   • recordcount=${recordcount} indicates the total amount of data loaded to the DDS instance.
   • mongodb.url="mongodb://${userName}:${password}@${mongosIP}:${port}/ycsb?authSource=admin" indicates the HA connection address of the DDS instance.
   • threads ${threadNum} indicates the number of concurrent threads on the client.
2. Use YCSB to run the following command to test the performance:
   ./bin/ycsb run mongodb -s -p workload=site.ycsb.workloads.CoreWorkload -p recordcount=${recordcount} -p operationcount=${operationcount} -p insertproportion=0 -p readproportion=1 -p updateproportion=0 -p requestdistribution=zipfian -p mongodb.url="mongodb://${userName}:${password}@${mongosIP}:${port}/ycsb?authSource=admin" -threads ${threadNum}

   

   • recordcount=${recordcount} indicates the total amount of data loaded to the DDS instance.
   • operationcount=${operationcount} indicates the total number of read/write operations.
   • insertproportion=0 indicates the proportion of data loading operations.
   • readproportion=1 indicates the proportion of read operations.
   • updateproportion=0indicates the proportion of update operations.
   • mongodb.url="mongodb://${userName}:${password}@${mongosIP}:${port}/ycsb?authSource=admin" indicates the HA connection address of the DDS instance.
   • threads ${threadNum} indicates the number of concurrent threads on the client.
   • The values of recordcount and operationcount are the same.
   • The sum of the values of insertproportion, readproportion, and updateproportion is 1.

Testing Models

• Service model

  Table 1 Service model

  Service Model No.	Service Model
  S1	100% insert
  S2	90% update ,10% read
  S3	65% read ,25% insert, 10% update
  S4	90% read ,5% insert, 5% update
  S5	50% update, 50% read
  S6	100% read
  S7	95% read ,5% update
  S8	5% read ,95% update
  S9	100% update

• Document model

  Use the default configuration of YCSB: The size of each document is 1 KB, and the default index is _id.

• Data volume to be prepared

  In this test, prepare two types of data volume for each cluster instance.

  For details, see the following table.

  Table 2 Data volume to be prepared

  Specifications	Low-Level Data Volume	High-Level Data Volume	Number of Concurrent Threads
  1 vCPUs | 4 GB	Records: 1,000,000	Records: 10,000,000	100
  2 vCPUs | 4 GB	Records: 1,000,000	Records: 10,000,000	100
  2 vCPUs | 8 GB	Records: 2,000,000	Records: 20,000,000	100
  2 vCPUs | 16 GB	Records: 4,000,000	Records: 40,000,000	100
  4 vCPUs | 8 GB	Records: 2,000,000	Records: 20,000,000	100
  4 vCPUs | 16 GB	Records: 4,000,000	Records: 40,000,000	100
  4 vCPUs | 32 GB	Records: 10,000,000	Records: 100,000,000	100
  8 vCPUs | 16 GB	Records: 4,000,000	Records: 40,000,000	100
  8 vCPUs | 32 GB	Records: 10,000,000	Records: 100,000,000	100
  8 vCPUs | 64 GB	Records: 10,000,000	Records: 100,000,000	100
  16 vCPUs | 32 GB	Records: 10,000,000	Records: 100,000,000	100
  16 vCPUs | 64 GB	Records: 10,000,000	Records: 100,000,000	100
  16 vCPUs | 128 GB	Records: 10,000,000	Records: 100,000,000	100
  32 vCPUs | 64 GB	Records: 10,000,000	Records: 100,000,000	200
  32 vCPUs | 128 GB	Records: 10,000,000	Records: 100,000,000	200
  64 vCPUs | 256 GB	Records: 10,000,000	Records: 100,000,000	400

• Data consistency model

  Weak consistency: For the write concern setting of {w: 1, j: false}, an acknowledgment is returned after data is written to the disk on a single node. Data is persisted on disks in asynchronous mode with the default write concern setting.