2022-09-22 06:17:05 -04:00
|
|
|
# Connection Guide: https://www.mongodb.com/docs/drivers/java/sync/current/fundamentals/connection/
|
2023-01-26 19:02:43 -06:00
|
|
|
# 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
|
2022-09-22 06:17:05 -04:00
|
|
|
# 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
|
2023-01-26 19:02:43 -06:00
|
|
|
min_version: "5.17.1"
|
2022-09-22 06:17:05 -04:00
|
|
|
|
|
|
|
|
description: |
|
|
|
|
|
This workload is analogous to the cql-tabular2 workload, just implemented for MongoDB.
|
|
|
|
|
|
|
|
|
|
scenarios:
|
|
|
|
|
default:
|
2023-01-26 19:02:43 -06:00
|
|
|
schema: run driver=mongodb tags==block:'schema.*' threads==1 cycles==UNDEF
|
2022-09-22 06:17:05 -04:00
|
|
|
rampup: run driver=mongodb tags==block:rampup cycles===TEMPLATE(rampup-cycles,10000000) threads=auto
|
2023-01-26 19:02:43 -06:00
|
|
|
main: run driver=mongodb tags==block:'main-.*' cycles===TEMPLATE(main-cycles,10000000) threads=auto
|
2022-09-22 06:17:05 -04:00
|
|
|
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: {
|
2022-09-27 09:27:43 -04:00
|
|
|
part: 1,
|
2022-09-22 06:17:05 -04:00
|
|
|
},
|
|
|
|
|
name: "tab_part_idx",
|
2022-09-27 09:27:43 -04:00
|
|
|
unique: false
|
2022-09-22 06:17:05 -04:00
|
|
|
},
|
|
|
|
|
{
|
|
|
|
|
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."
|
2022-09-27 09:27:43 -04:00
|
|
|
}
|
