IntenseFinance/nosqlbench
3
0
Fork 0
mirror of https://github.com/nosqlbench/nosqlbench.git synced 2025-02-25 18:55:28 -06:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #719 from nosqlbench/feature/mongodb_baselines2_workloads

Browse Source 
Initial Draft for designing baselines2 workloads for MongoDB
This commit is contained in:
Jonathan Shook 2022-09-22 15:56:53 -05:00 committed by GitHub
commit adb218e57e
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 576 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

125
mongodb-keyvalue2.yaml Normal file
View File

@ -0,0 +1,125 @@
# Connection Guide: https://www.mongodb.com/docs/drivers/java/sync/current/fundamentals/connection/
# nb5 run driver=mongodb workload=/path/to/mongodb-keyvalue2.yaml tags=block:schema connection='mongodb+srv://user:pass@sample-db.host.mongodb.net/?retryWrites=true&w=majority' database=baselines -vv --show-stacktraces
# nb5 run driver=mongodb workload=/path/to/mongodb-keyvalue2.yaml tags=block:rampup cycles=25 connection='mongodb+srv://user:pass@sample-db.host.mongodb.net/?retryWrites=true&w=majority' database=baselines -vv --show-stacktraces
# nb5 run driver=mongodb workload=/path/to/mongodb-keyvalue2.yaml tags='block:main-.*' cycles=25 connection='mongodb+srv://user:pass@sample-db.host.mongodb.net/?retryWrites=true&w=majority' database=baselines -vv --show-stacktraces
min_version: "4.17.24"
description: |
This workload is analogous to the cql-keyvalue2 workload, just implemented for MongoDB.
scenarios:
default:
schema: run driver=mongodb tags==block:schema threads==1 cycles==UNDEF
rampup: run driver=mongodb tags==block:rampup cycles===TEMPLATE(rampup-cycles,10000000) threads=auto
main: run driver=mongodb tags==block:"main-.*" cycles===TEMPLATE(main-cycles,10000000) threads=auto
drop: run driver=mongodb tags==block:drop-collection threads==1 cycles==UNDEF
params:
instrument: true
bindings:
seq_key: Mod(TEMPLATE(keycount,1000000000)); ToString();
seq_value: >-
Hash();
Mod(TEMPLATE(valuecount,1000000000));
CharBufImage('A-Za-z0-9 _|/',16000000,HashRange(TEMPLATE(mintext,50000)TEMPLATE(addzeroes,),TEMPLATE(maxtext,150000)TEMPLATE(addzeroes,)));
ToString();
rw_key: TEMPLATE(keydist,Uniform(0,1000000000)); ToString() -> String
rw_value: Hash(); TEMPLATE(valdist,Uniform(0,1000000000)); CharBufImage('A-Za-z0-9 _|/',16000000,HashRange(TEMPLATE(mintext,50000)TEMPLATE(addzeros,),TEMPLATE(maxtext,150000)TEMPLATE(addzeros,))); ToString();
blocks:
schema:
params:
prepared: false
ops:
# https://www.mongodb.com/docs/manual/reference/method/db.createCollection/
# https://www.mongodb.com/docs/manual/core/schema-validation/specify-json-schema/
# `clusteredIndex` only support creation of an index on `_id` field (as shown below) so its optional
create-collection: |
{
create: "TEMPLATE(collection,keyvalue)",
clusteredIndex: {
key: { "_id": 1 },
unique: true,
name: "_id_idx"
},
writeConcern: { w: "majority" },
validator: {
$jsonSchema: {
bsonType: "object",
title: "Key/Value collection schema validation",
required: [ "key" ],
properties: {
key: {
bsonType: "string",
description: "'key' must be a string and is required"
}
}
}
},
validationLevel: "strict",
validationAction: "error",
comment: "keyvalue collection creation with strict types and required 'key' field."
}
create-key-index: |
{
createIndexes: "TEMPLATE(collection,keyvalue)",
indexes: [
{
key: {
key: 1,
},
name: "kv_key_idx",
unique: true
}
],
writeConcern: { w: "majority" },
comment: "'key' index creation for keyvalue collection. Values should be unique.",
commitQuorum: "majority"
}
rampup:
ops:
rampup-insert: |
{
insert: "TEMPLATE(collection,keyvalue)",
documents: [
{
key: "{rw_key}",
value: "{rw_value}"
}
],
comment: "Insert documents into keyvalue collection."
}
main-read:
params:
ratio: 5
ops:
main-select: |
{
find: "TEMPLATE(collection,keyvalue)",
filter: { key: { $eq: "{rw_key}" } },
readConcern: { level: "majority" },
comment: "Find the value for the given 'key'."
}
main-write:
params:
ratio: 5
ops:
main-insert: |
{
insert: "TEMPLATE(collection,keyvalue)",
documents: [
{
key: "{rw_key}",
value: "{rw_value}"
}
],
writeConcern: { w: "majority" },
comment: "Insert documents into keyvalue collection."
}
drop-collection:
ops:
drop-collection: |
{
drop: "TEMPLATE(collection,keyvalue)",
comment: "Drop keyvalue collection to start afresh."
}

280
mongodb-tabular2.yaml Normal file
View File

@ -0,0 +1,280 @@
# Connection Guide: https://www.mongodb.com/docs/drivers/java/sync/current/fundamentals/connection/
# nb5 run driver=mongodb workload=/path/to/mongodb-tabular2.yaml tags=block:schema connection='mongodb+srv://user:pass@sample-db.host.mongodb.net/?retryWrites=true&w=majority' database=baselines -vv --show-stacktraces
# nb5 run driver=mongodb workload=/path/to/mongodb-tabular2.yaml tags=block:rampup cycles=25 connection='mongodb+srv://user:pass@sample-db.host.mongodb.net/?retryWrites=true&w=majority' database=baselines -vv --show-stacktraces
# nb5 run driver=mongodb workload=/path/to/mongodb-tabular2.yaml tags='block:main.*' cycles=25 connection='mongodb+srv://user:pass@sample-db.host.mongodb.net/?retryWrites=true&w=majority' database=baselines -vv --show-stacktraces
min_version: "4.17.24"
description: |
This workload is analogous to the cql-tabular2 workload, just implemented for MongoDB.
scenarios:
default:
schema: run driver=mongodb tags==block:schema threads==1 cycles==UNDEF
rampup: run driver=mongodb tags==block:rampup cycles===TEMPLATE(rampup-cycles,10000000) threads=auto
main: run driver=mongodb tags==block:"main-.*" cycles===TEMPLATE(main-cycles,10000000) threads=auto
drop: run driver=mongodb tags==block:drop-collection threads==1 cycles==UNDEF
params:
instrument: true
bindings:
# for ramp-up and verify phases
#
part_layout: Div(TEMPLATE(partsize,1000)); ToString() -> String
clust_layout: Mod(TEMPLATE(partsize,1000)); ToString() -> String
# todo: update these definitions to use the simpler 10,0.1, 20, 0.2, ...
data0: Add(10); HashedFileExtractToString('data/lorem_ipsum_full.txt',9,11)
data1: Add(20); HashedFileExtractToString('data/lorem_ipsum_full.txt',18,22)
data2: Add(30); HashedFileExtractToString('data/lorem_ipsum_full.txt',27,33)
data3: Add(40); HashedFileExtractToString('data/lorem_ipsum_full.txt',45,55)
data4: Add(50); HashedFileExtractToString('data/lorem_ipsum_full.txt',72,88)
data5: Add(60); HashedFileExtractToString('data/lorem_ipsum_full.txt',107,143)
data6: Add(70); HashedFileExtractToString('data/lorem_ipsum_full.txt',189,231)
data7: Add(80); HashedFileExtractToString('data/lorem_ipsum_full.txt',306,374)
# for main phase
# for write
part_write: Hash(); Uniform(0,TEMPLATE(partcount,100))->int; ToString() -> String
clust_write: Hash(); Add(1); Uniform(0,TEMPLATE(partsize,1000000))->int; ToString() -> String
data_write: Hash(); HashedFileExtractToString('data/lorem_ipsum_full.txt',50,150) -> String
# for read
limit: Uniform(1,10) -> int
part_read: Uniform(0,TEMPLATE(partcount,100))->int; ToString() -> String
clust_read: Add(1); Uniform(0,TEMPLATE(partsize,1000000))->int; ToString() -> String
blocks:
schema:
params:
prepared: false
ops:
# https://www.mongodb.com/docs/manual/reference/method/db.createCollection/
# https://www.mongodb.com/docs/manual/core/schema-validation/specify-json-schema/
# `clusteredIndex` only support creation of an index on `_id` field (as shown below) so its optional
create-collection: |
{
create: "TEMPLATE(collection,tabular)",
writeConcern: { w: "majority" },
validator: {
$jsonSchema: {
bsonType: "object",
title: "Tabular collection schema validation",
required: [ "part", "clust" ],
properties: {
part: {
bsonType: "string",
description: "'part' must be a string, unique and is required"
},
clust: {
bsonType: "string",
description: "'clust' must be a string, unique and is required"
},
data0: {
bsonType: "string",
description: "'data0' must be a string"
},
data1: {
bsonType: "string",
description: "'data1' must be a string"
},
data2: {
bsonType: "string",
description: "'data2' must be a string"
},
data3: {
bsonType: "string",
description: "'data3' must be a string"
},
data4: {
bsonType: "string",
description: "'data4' must be a string"
},
data5: {
bsonType: "string",
description: "'data5' must be a string"
},
data6: {
bsonType: "string",
description: "'data6' must be a string"
},
data7: {
bsonType: "string",
description: "'data7' must be a string"
}
}
}
},
validationLevel: "strict",
validationAction: "error",
comment: "tabular-like collection creation with strict types and required 'part' field."
}
create-part-clust-index: |
{
createIndexes: "TEMPLATE(collection,tabular)",
indexes: [
{
key: {
key: 1,
},
name: "tab_part_idx",
unique: true
},
{
key: {
clust: 1,
},
name: "tab_clust_idx",
unique: true
}
],
writeConcern: { w: "majority" },
comment: "'part' & 'clust' index creation for tabular collection. Values should be unique.",
commitQuorum: "majority"
}
rampup:
ops:
rampup-insert: |
{
insert: "TEMPLATE(collection,tabular)",
documents: [
{
part: "{part_write}",
clust: "{clust_write}",
data0: "{data0}",
data1: "{data1}",
data2: "{data2}",
data3: "{data3}",
data4: "{data4}",
data5: "{data5}",
data6: "{data6}",
data7: "{data7}"
}
],
writeConcern: { w: "majority" },
comment: "Insert documents into tabular collection."
}
main-read:
params:
ratio: TEMPLATE(read_ratio,1)
ops:
main-select-all:
statement: |
{
find: "TEMPLATE(collection,tabular)",
filter: { part: { $eq: "{part_read}" } },
limit: {limit},
readConcern: { level: "majority" },
comment: "Find the value for the given 'part'."
}
ratio: TEMPLATE(read_ratio,1)
main-select-01:
statement: |
{
find: "TEMPLATE(collection,tabular)",
filter: { part: { $eq: "{part_read}" } },
projection: { data0: 1, data1: 1 },
limit: {limit},
readConcern: { level: "majority" },
comment: "Find the data01 value for the given 'part'."
}
ratio: TEMPLATE(read_ratio,1)
main-select-0246:
statement: |
{
find: "TEMPLATE(collection,tabular)",
filter: { part: { $eq: "{part_read}" } },
projection: { data0: 1, data2: 1, data4: 1, data6: 1 },
limit: {limit},
readConcern: { level: "majority" },
comment: "Find the data0246 value for the given 'part'."
}
ratio: TEMPLATE(read_ratio,1)
main-select-1357:
statement: |
{
find: "TEMPLATE(collection,tabular)",
filter: { part: { $eq: "{part_read}" } },
projection: { data1: 1, data3: 1, data5: 1, data7: 1 },
limit: {limit},
readConcern: { level: "majority" },
comment: "Find the data1357 value for the given 'part'."
}
ratio: TEMPLATE(read_ratio,1)
main-select-0123:
statement: |
{
find: "TEMPLATE(collection,tabular)",
filter: { part: { $eq: "{part_read}" } },
projection: { data0: 1, data1: 1, data2: 1, data3: 1 },
limit: {limit},
readConcern: { level: "majority" },
comment: "Find the data0123 value for the given 'part'."
}
ratio: TEMPLATE(read_ratio,1)
main-select-4567:
statement: |
{
find: "TEMPLATE(collection,tabular)",
filter: { part: { $eq: "{part_read}" } },
projection: { data4: 1, data5: 1, data6: 1, data7: 1 },
limit: {limit},
readConcern: { level: "majority" },
comment: "Find the data4567 value for the given 'part'."
}
ratio: TEMPLATE(read_ratio,1)
main-select-67:
statement: |
{
find: "TEMPLATE(collection,tabular)",
filter: { part: { $eq: "{part_read}" } },
projection: { data6: 1, data7: 1 },
limit: {limit},
readConcern: { level: "majority" },
comment: "Find the data67 value for the given 'part'."
}
ratio: TEMPLATE(read_ratio,1)
main-select:
statement: |
{
find: "TEMPLATE(collection,tabular)",
filter: { part: { $eq: "{part_read}" } },
projection: { data0: 1, data1: 1, data2: 1, data3: 1, data4: 1, data5: 1, data6: 1, data7: 1 },
limit: {limit},
readConcern: { level: "majority" },
comment: "Find the data01234567 value for the given 'part'."
}
ratio: TEMPLATE(read_ratio,1)
main-write:
params:
ratio: TEMPLATE(write_ratio,8)
ops:
main-insert:
statement: |
{
insert: "TEMPLATE(collection,tabular)",
documents: [
{
part: "{part_write}",
clust: "{clust_write}",
data0: "{data0}",
data1: "{data1}",
data2: "{data2}",
data3: "{data3}",
data4: "{data4}",
data5: "{data5}",
data6: "{data6}",
data7: "{data7}"
}
],
comment: "Insert documents into tabular collection."
}
ratio: TEMPLATE(write_ratio,8)
# The below drop-collection blocks expects the collection to exist or else this will fail
drop-collection:
ops:
drop-collection: |
{
drop: "TEMPLATE(collection,tabular)",
comment: "Drop tabular collection to start afresh."
}

171
mongodb-timeseries2.yaml Normal file
View File

@ -0,0 +1,171 @@
# Connection Guide: https://www.mongodb.com/docs/drivers/java/sync/current/fundamentals/connection/
# nb5 run driver=mongodb workload=/path/to/mongodb-timeseries2.yaml tags=block:schema connection='mongodb+srv://user:pass@sample-db.host.mongodb.net/?retryWrites=true&w=majority' database=baselines -vv --show-stacktraces
# nb5 run driver=mongodb workload=/path/to/mongodb-timeseries2.yaml tags=block:rampup cycles=25 connection='mongodb+srv://user:pass@sample-db.host.mongodb.net/?retryWrites=true&w=majority' database=baselines -vv --show-stacktraces
# nb5 run driver=mongodb workload=/path/to/mongodb-timeseries2.yaml tags='block:main.*' cycles=25 connection='mongodb+srv://user:pass@sample-db.host.mongodb.net/?retryWrites=true&w=majority' database=baselines -vv --show-stacktraces
min_version: "4.17.24"
description: |
This workload is analogous to the cql-timeseries2 workload, just implemented for MongoDB.
scenarios:
default:
schema: run driver=mongodb tags==block:schema threads==1 cycles==UNDEF
rampup: run driver=mongodb tags==block:rampup cycles===TEMPLATE(rampup-cycles,10000000) threads=auto
main: run driver=mongodb tags==block:"main-.*" cycles===TEMPLATE(main-cycles,10000000) threads=auto
drop: run driver=mongodb tags==block:drop-collection threads==1 cycles==UNDEF
params:
instrument: true
bindings:
machine_id: Mod(TEMPLATE(sources,10000)); ToHashedUUID() -> java.util.UUID
sensor_name: HashedLineToString('data/variable_words.txt')
time: Mod(TEMPLATE(maxtimes,1000000000000L)); Mul(TEMPLATE(timespeed,100)L); Div(TEMPLATE(sources,10000)L); ToDate()
cell_timestamp: Mul(TEMPLATE(timespeed,100)L); Div(TEMPLATE(sources,10000)L); Mul(1000L)
sensor_value: Normal(0.0,5.0); Add(100.0) -> double
station_id: Div(TEMPLATE(sources,10000)); Mod(TEMPLATE(stations,1000); ToHashedUUID() -> java.util.UUID
data: HashedFileExtractToString('data/lorem_ipsum_full.txt',800,1200)
blocks:
schema:
params:
prepared: false
ops:
# https://www.mongodb.com/docs/manual/reference/method/db.createCollection/
# https://www.mongodb.com/docs/manual/core/schema-validation/specify-json-schema/
# https://www.mongodb.com/docs/manual/core/timeseries-collections/
# `clusteredIndex` only support creation of an index on `_id` field (as shown below) so its optional
create-collection: |
{
create: "TEMPLATE(collection,timeseries)",
timeseries: {
timeField: "time",
granularity: "seconds"
},
writeConcern: { w: "majority" },
validator: {
$jsonSchema: {
bsonType: "object",
title: "Timeseries collection schema validation",
required: [ "machine_id", "sensor_name", "time" ],
properties: {
machine_id: {
bsonType: "binData",
description: "'machine_id' must be UUID, unique and is required"
},
sensor_name: {
bsonType: "string",
description: "'sensor_name' must be a string, unique and is required"
},
time: {
bsonType: "timestamp",
description: "'time' must be a timestamp, and required"
},
sensor_value: {
bsonType: "double",
description: "'sensor_value' must be a double"
},
station_id: {
bsonType: "binData",
description: "'station_id' must be a UUID"
},
data: {
bsonType: "string",
description: "'data' must be a string"
}
}
}
},
validationLevel: "strict",
validationAction: "error",
comment: "timeseries-like collection creation with strict types and required '(machine_id, sensor_name), time' field."
}
create-part-clust-index: |
{
createIndexes: "TEMPLATE(collection,timeseries)",
indexes: [
{
key: {
machine_id: 1,
sensor_name: 1
},
name: "ts_machine_id_sensor_name_idx",
unique: true
},
{
key: {
time: -1,
},
name: "tab_time_desc_idx",
unique: true
}
],
writeConcern: { w: "majority" },
comment: "'(machine_id, sensor_name), time' index creation for timeseries collection. Values should be unique. 'time' field is sorted in DESC order.",
commitQuorum: "majority"
}
# UUID https://www.mongodb.com/docs/manual/reference/bson-types/#binary-data
# Example: https://www.tutorialspoint.com/query-bindata-by-type-in-mongodb
rampup:
ops:
rampup-insert: |
{
insert: "TEMPLATE(collection,timeseries)",
documents: [
{
machine_id: "{machine_id}",
sensor_name: "{sensor_name}",
time: BinData(4, "{time}"),
sensor_value: "{sensor_value}",
station_id: "{station_id}",
data: "{data}"
}
],
writeConcern: { w: "majority" },
comment: "Insert documents into timeseries collection."
}
main-read:
params:
ratio: TEMPLATE(read_ratio,1)
ops:
select-read:
statement: |
{
find: "TEMPLATE(collection,timeseries)",
filter: { machine_id: { $eq: "{machine_id}" }, sensor_name: { $eq: "{sensor_name}" } },
limit: TEMPLATE(limit,10),
readConcern: { level: "majority" },
comment: "Find the value for the given 'machine_id' and 'sensor_name'."
}
ratio: TEMPLATE(read_ratio,1)
main-write:
params:
ratio: TEMPLATE(write_ratio,9)
ops:
main-insert:
statement: |
{
insert: "TEMPLATE(collection,timeseries)",
documents: [
{
machine_id: "{machine_id}",
sensor_name: "{sensor_name}",
time: BinData(4, "{time}"),
sensor_value: "{sensor_value}",
station_id: "{station_id}",
data: "{data}"
}
],
writeConcern: { w: "majority" },
comment: "Insert documents into timeseries collection."
}
ratio: TEMPLATE(write_ratio,9)
# The below drop-collection blocks expects the collection to exist or else this will fail
drop-collection:
ops:
drop-collection: |
{
drop: "TEMPLATE(collection,timeseries)",
comment: "Drop timeseries collection to start afresh."
}