Skip to content

diskann-benchmark refactor pt2 - shared logic overhaul #1169

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
JordanMaples wants to merge 7 commits into
base: main
Choose a base branch
from
Open

diskann-benchmark refactor pt2 - shared logic overhaul #1169

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
29b35f1
Refactor streaming benchmarks and unify bftree inputs
JordanMaples Jun 15, 2026
a0f481c
undo name rename
JordanMaples Jun 18, 2026
b37c21f
enum
JordanMaples Jun 18, 2026
634b25c
QuantConfig implements display
JordanMaples Jun 18, 2026
b306e94
Validate search_l >= search_n in StreamingSearchParams
JordanMaples Jun 22, 2026
f9a9213
Address review: restore InlineFilterSearchPhase name, add recall_k <=…
JordanMaples Jun 22, 2026
ff5cff6
Make Managed opt-in: bftree streams bypass ID management layer
JordanMaples Jun 25, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions diskann-benchmark/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -269,9 +269,9 @@ First, set up the runbook and ground truth for the desired workload. Refer to th

Benchmarks are run with
```sh
cargo run --release --package diskann-benchmark -- run --input-file ./diskann-benchmark/example/graph-index-dynamic.json --output-file dynamic-output.json
cargo run --release --package diskann-benchmark -- run --input-file ./diskann-benchmark/example/graph-index-stream.json --output-file stream-output.json
```
Note in the example json that the benchmark is registered under `graph-index-dynamic-run`,
Note in the example json that the benchmark is registered under `graph-index-stream-run`,
instead of `graph-index-build` etc..

A streaming run happens in several phases.
Expand Down
35 changes: 19 additions & 16 deletions diskann-benchmark/example/graph-index-bftree-spherical.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
],
"jobs": [
{
"type": "graph-index-build-bftree-spherical-quantization",
"type": "graph-index-bftree",
"content": {
"build": {
"data_type": "float32",
Expand Down Expand Up @@ -40,10 +40,24 @@
}
]
},
"seed": 42,
"transform_kind": "null",
"num_bits": 2,
"pre_scale": "reciprocal_mean_norm",
"quantization": {
"kind": "spherical",
"seed": 42,
"transform_kind": "null",
"num_bits": 2,
"pre_scale": "reciprocal_mean_norm",
"quant_store_config": {
"cb_size_byte": 67108864,
"leaf_page_size": 4096,
"cb_max_record_size": null,
"cb_min_record_size": null,
"read_promotion_rate": null,
"scan_promotion_rate": null,
"cb_copy_on_access_ratio": null,
"read_record_cache": null,
"cache_only": null
}
},
"vector_store_config": {
"cb_size_byte": 67108864,
"leaf_page_size": 4096,
Expand All @@ -65,17 +79,6 @@
"cb_copy_on_access_ratio": null,
"read_record_cache": null,
"cache_only": null
},
"quant_store_config": {
"cb_size_byte": 67108864,
"leaf_page_size": 4096,
"cb_max_record_size": null,
"cb_min_record_size": null,
"read_promotion_rate": null,
"scan_promotion_rate": null,
"cb_copy_on_access_ratio": null,
"read_record_cache": null,
"cache_only": null
}
}
}
Expand Down
84 changes: 84 additions & 0 deletions diskann-benchmark/example/graph-index-bftree-stream-spherical.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,84 @@
{
"search_directories": [
"test_data/disk_index_search"
],
"jobs": [
{
"type": "graph-index-stream-bftree",
"content": {
"build": {
"data_type": "float32",
"data": "disk_index_siftsmall_learn_256pts_data.fbin",
"distance": "squared_l2",
"max_degree": 32,
"l_build": 50,
"start_point_strategy": "medoid",
"alpha": 1.2,
"backedge_ratio": 1.0,
"num_threads": 1
},
"search": {
"queries": "disk_index_sample_query_10pts.fbin",
"reps": 1,
"num_threads": 1,
"search_l": 40,
"search_n": 10,
"recall_k": 10
},
"runbook_params": {
"runbook_path": "example_runbook.yaml",
"dataset_name": "sift-small-256",
"gt_directory": "example_runbook_gt",
"ip_delete_method": {
"method": "visited_and_top_k",
"params": {
"k_value": 10,
"l_value": 40
}
},
"ip_delete_num_to_replace": 3
},
"quantization": {
"kind": "spherical",
"seed": 42,
"transform_kind": "null",
"num_bits": 2,
"pre_scale": "reciprocal_mean_norm",
"quant_store_config": {
"cb_size_byte": 67108864,
"leaf_page_size": 4096,
"cb_max_record_size": null,
"cb_min_record_size": null,
"read_promotion_rate": null,
"scan_promotion_rate": null,
"cb_copy_on_access_ratio": null,
"read_record_cache": null,
"cache_only": null
}
},
"vector_store_config": {
"cb_size_byte": 67108864,
"leaf_page_size": 4096,
"cb_max_record_size": null,
"cb_min_record_size": null,
"read_promotion_rate": null,
"scan_promotion_rate": null,
"cb_copy_on_access_ratio": null,
"read_record_cache": null,
"cache_only": null
},
"neighbor_store_config": {
"cb_size_byte": 67108864,
"leaf_page_size": 4096,
"cb_max_record_size": null,
"cb_min_record_size": null,
"read_promotion_rate": null,
"scan_promotion_rate": null,
"cb_copy_on_access_ratio": null,
"read_record_cache": null,
"cache_only": null
}
}
}
]
}
26 changes: 9 additions & 17 deletions diskann-benchmark/example/graph-index-bftree-stream.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
],
"jobs": [
{
"type": "graph-index-stream-bftree-full-precision",
"type": "graph-index-stream-bftree",
"content": {
"build": {
"data_type": "float32",
Expand All @@ -17,24 +17,13 @@
"backedge_ratio": 1.0,
"num_threads": 1
},
"search_phase": {
"search-type": "topk",
"search": {
"queries": "disk_index_sample_query_10pts.fbin",
"groundtruth": "disk_index_10pts_idx_uint32_truth_search_res.bin",
"reps": 1,
"num_threads": [
1
],
"runs": [
{
"search_n": 10,
"search_l": [
20,
40
],
"recall_k": 10
}
]
"num_threads": 1,
"search_l": 40,
"search_n": 10,
"recall_k": 10
},
"runbook_params": {
"runbook_path": "example_runbook.yaml",
Expand All @@ -49,6 +38,9 @@
},
"ip_delete_num_to_replace": 3
},
"quantization": {
"kind": "none"
},
"vector_store_config": {
"cb_size_byte": 67108864,
"leaf_page_size": 4096,
Expand Down
5 changes: 4 additions & 1 deletion diskann-benchmark/example/graph-index-bftree.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
],
"jobs": [
{
"type": "graph-index-build-bftree-full-precision",
"type": "graph-index-bftree",
"content": {
"build": {
"data_type": "float32",
Expand Down Expand Up @@ -37,6 +37,9 @@
}
]
},
"quantization": {
"kind": "none"
},
"vector_store_config": {
"cb_size_byte": 67108864,
"leaf_page_size": 4096,
Expand Down
23 changes: 6 additions & 17 deletions ...enchmark/example/graph-index-dynamic.json → ...benchmark/example/graph-index-stream.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
],
"jobs": [
{
"type": "graph-index-dynamic-run",
"type": "graph-index-stream-run",
"content": {
"build": {
"data_type": "float32",
Expand All @@ -17,24 +17,13 @@
"num_threads": 4,
"start_point_strategy": "medoid"
},
"search_phase": {
"search-type": "topk",
"search": {
"queries": "disk_index_sample_query_10pts.fbin",
"groundtruth": "disk_index_10pts_idx_uint32_truth_search_res.bin",
"reps": 1,
"num_threads": [
2
],
"runs": [
{
"search_n": 10,
"search_l": [
20,
40
],
"recall_k": 10
}
]
"num_threads": 2,
"search_l": 40,
"search_n": 10,
"recall_k": 10
},
"runbook_params": {
"runbook_path": "example_runbook.yaml",
Expand Down
Loading
Loading