Skip to content

Latest commit

 

History

History
121 lines (79 loc) · 6.39 KB

DifferentTestMode-EN.md

File metadata and controls

121 lines (79 loc) · 6.39 KB
Raw

1. Different operation modes of iot-benchmark

Again, before you start any new test case, you need to confirm whether the configuration information in the configuration file config.properties meets your expectations.

1.1. General test mode: write (single database)

Assume that the workload parameters are:

+--------------------------+------------+--------------+-------------+------------+--------------------+--------+
| Measurement/StorageGroup | tag/device | field/sensor | concurrency | batch size | point interval(ms) | loop |
+--------------------------+------------+--------------+-------------+------------+--------------------+--------+
| 10 | 50 | 500 | 20 | 100 | 200 | 10000 |
+--------------------------+------------+--------------+-------------+------------+--------------------+--------+

Note: The total number of time series in this configuration is: device * sensor = 25,000, and the number of points in each time series is batch size * loop = 20,000, The total number of data points is deivce * sensor * batch size * loop = 500,000,000. The space occupied by each data point can be estimated as 16 bytes, so the total size of the raw data is 8G.

Then, the config.properties file needs to be modified as shown below. Please remove the # before the corresponding configuration item after completing the modification to ensure that the changes take effect:

HOST=127.0.0.1
PORT=6667
DB_SWITCH=IoTDB-100-SESSION_BY_TABLET
BENCHMARK_WORK_MODE=testWithDefaultPath
OPERATION_PROPORTION=1:0:0:0:0:0:0:0:0:0:0
GROUP_NUMBER=10
DEVICE_NUMBER=50
SENSOR_NUMBER=500
CLIENT_NUMBER=20
BATCH_SIZE_PER_WRITE=100
POINT_STEP=200
LOOP=10000

If you want to write data in a disorderly manner, you need to modify the following properties of the config.properties file:

# Whether to write out of order
IS_OUT_OF_ORDER=true
# Out of order write mode, currently there are 2 types
# POISSON Out of order mode according to Poisson distribution
# BATCH Batch insert out of order mode
OUT_OF_ORDER_MODE=BATCH
# The proportion of data written out of order
OUT_OF_ORDER_RATIO=0.5
# Is it equal time stamp
IS_REGULAR_FREQUENCY=true
# Expectation and variance of Poisson distribution
LAMBDA=2200.0
# Maximum value of random number of Poisson distribution model
MAX_K=170000

Some basic modification operations of parameters have been introduced here, and the relevant settings will be omitted in the following part. If necessary, please enter the following page to view. Testconfigurations

2. General test mode: query (single database, no system records)

In addition to writing data, the general test mode can also query data.

3. General test mode: read-write mixed mode (single database)

General test mode can support users to perform read-write mixed tests. It should be noted that the timestamps of read-write mixed in this scenario all start from the write start time.

4. General test mode: read-write mixed mode (single database, query the most recently written data)

General test mode can support users to perform read-write mixed tests (query the most recently written data). It should be noted that the query time range in this scenario is the data adjacent to the left of the current maximum write timestamp.

5. General test mode: use system records (single database)

IoTDB Benchmark supports you to use the database to store system data during the test. Currently, it supports the use of CSV records.

6. Conventional test mode: test process persistence (single database)

For subsequent analysis, iot-benchmark can store test information in the database (if you do not want to store test data, set TEST_DATA_PERSISTENCE=None)

7. Generate data mode

In order to generate reusable data sets, iot-benchmark provides a mode for generating data sets, generating data sets to FILE_PATH for subsequent use in correctness write mode and correctness query mode.

8. Correctness write mode (single database, external data set)

In order to verify the correctness of data set writing, you can use this mode to write the data set generated in the generate data mode. Currently, this mode only supports IoTDB v1.0 and later versions and InfluxDB v1.x

9. Correctness single-point query mode (single database, external data set)

Before running this mode, you need to use the correctness write mode to write data to the database. To verify the correctness of the data set written, you can use this mode to query the data set written to the database. Currently, this mode only supports IoTDB v1.0 and InfluxDB v1.x.

10. Dual database mode

To complete the correctness verification more conveniently and quickly, iot-benchmark also supports dual database mode.

  1. For all the test scenarios mentioned above, unless otherwise specified, dual databases are supported. Please start the test in the verification project.

  2. For the relevant test scenarios for correctness verification below, they must be run in dual database mode, and currently only IoTDB v1.0 and newer versions and timescaledb are supported.

11. General test mode: write (dual database)

In order to perform the correctness verification below, you first need to write the data to two databases.

12. Correctness single point query mode (dual database comparison)

In order to more efficiently verify the correctness of database data, iot-benchmark provides correctness verification by comparing the data between two databases. Note: Before performing this test, please use the general test mode of writing (dual databases) to complete the database writing. Currently, it is recommended to use the JDBC method.

13. Correctness function query mode (dual database comparison)

In order to more efficiently verify the correctness of database queries, iot-benchmark provides correctness verification by comparing the differences in data query results between two databases.

Note:

  1. Before performing this test, please use the general test mode of writing (dual databases) to complete the database writing.

  2. The value of LOOP cannot be too large, satisfying: LOOP(query) * QUERY_INTERVAL(query) * DEVICE_NUMBER(write) <= LOOP(write) * POINT_STEP(write)

14. Further explanation of correctness verification

  1. Currently, the correctness verification part only supports IoTDB v1.0 and later versions and TimeScaleDB

  2. Quick Guide