IntenseFinance/nosqlbench
3
0
Fork 0
mirror of https://github.com/nosqlbench/nosqlbench.git synced 2025-01-11 00:12:04 -06:00
Code Issues Packages Projects Releases Wiki Activity

enable knn

Browse Source
This commit is contained in:
Jonathan Shook 2024-02-08 16:18:22 -06:00
parent 351e8e333b
commit 8e6a14ee43
2 changed files with 58 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
adapter-aws-opensearch/src/main/java/io/nosqlbench/adapter/opensearch/dispensers/CreateIndexOpDispenser.java
View File

@ -23,6 +23,7 @@ import org.opensearch.client.json.JsonData;
import org.opensearch.client.opensearch.OpenSearchClient;
import org.opensearch.client.opensearch._types.mapping.*;
import org.opensearch.client.opensearch.indices.CreateIndexRequest;
import org.opensearch.client.opensearch.indices.IndexSettings;
import java.util.Map;
import java.util.function.LongFunction;
@ -41,6 +42,7 @@ public class CreateIndexOpDispenser extends BaseOpenSearchOpDispenser {
CreateIndexRequest.Builder eb = new CreateIndexRequest.Builder();
LongFunction<CreateIndexRequest.Builder> bfunc =
l -> new CreateIndexRequest.Builder()
.settings(b -> b.knn(true))
.index(targetF.apply(1));
bfunc = op.enhanceFunc(bfunc, "mappings", Map.class, this::resolveTypeMapping);

60
adapter-aws-opensearch/src/main/resources/activities/osvectors.yaml
View File

@ -1,16 +1,68 @@
description: |
basic test for vectors in open search
https://www.elastic.co/guide/en/elasticsearch/reference/current/rest-apis.html
template vars:
TEMPLATE(indexname,vectors_index)
TEMPLATE(dimensions,25)
params:
driver: opensearch
instrument: true
scenarios:
schema: run driver=opensearch tags=block:create threads===1 cycles===undef
# schema: run tags=block:schema threads===1 cycles===undef
# rampup: run tags=block:rampup threads=auto cycles=TEMPLATE(rampup_cycles,10)
vectors:
drop: run tags='block:drop' labels='target:opensearch' threads===1 cycles===UNDEF
schema: run tags='block:schema' labels='target:astra' threads===1 cycles===UNDEF
rampup: run tags='block:rampup' labels='target:astra' threads=1 cycles=TEMPLATE(trainsize)
# threads=auto
# errors=counter,warn,log
# search_and_index: >-
# run alias=search_and_index tags='block:search_and_index,optype=select' labels='target:astra'
# cycles=TEMPLATE(testsize) errors=count,retry rate=100
# errors=counter,warn threads=auto
bindings:
id: ToString()
test_floatlist: HdfFileToFloatList("testdata/TEMPLATE(dataset).hdf5", "/test");
relevant_indices: HdfFileToIntArray("testdata/TEMPLATE(dataset).hdf5", "/neighbors")
distance_floatlist: HdfFileToFloatList("testdata/TEMPLATE(dataset).hdf5", "/distance")
train_floatlist: HdfFileToFloatList("testdata/TEMPLATE(dataset).hdf5", "/train");
synthetic_vectors: HashedFloatVectors(TEMPLATE(dimensions));
blocks:
create:
drop:
ops:
index:
create_index: TEMPLATE(indexname,vectors_index)
drop_index:
delete_index: TEMPLATE(indexname,vectors_index)
schema:
ops:
create_index:
create_index: TEMPLATE(indexname)
mappings:
m1: v1
bulkrampup:
ops:
bulk_index:
bulk_index:
- '{"key":"{id}","value":"{train_floatlist}"'
-
rampup:
ops:
index:
index: TEMPLATE(indexname)
document:
key: "{id}"
value: "{train_floatlist}"
# document: |
# {
# "key": "{id}",
# "value": {train_floatlist}
# }