workload.json

{% import "benchmark.helpers" as benchmark with context %}
{
    "version": 2,
    "description": "Benchmark vector search engine performance for different engine types like faiss, lucene and nmslib",
    "indices": [
        {
            "name": "{{ target_index_name }}",
            "body": "{{ target_index_body }}"
        },
        {
            "name": "{{ train_index_name }}",
            "body": "{{ train_index_body }}"
        }
    ],
    "corpora": [
    {
      "name": "cohere",
      "base-url": "https://dbyiw3u3rf9yr.cloudfront.net/corpora/vectorsearch/cohere-wikipedia-22-12-en-embeddings",
      "target-index": "{{ target_index_name }}",
      "documents": [
        {
          "source-file": "documents-1k.hdf5.bz2",
          "source-format": "hdf5",
          "document-count": 1000
          
        }
      ]
    },
    {
      "name": "cohere-100k",
      "base-url": "https://dbyiw3u3rf9yr.cloudfront.net/corpora/vectorsearch/cohere-wikipedia-22-12-en-embeddings",
      "target-index": "{{ target_index_name }}",
      "documents": [
        {
          "source-file": "documents-100k.hdf5.bz2",
          "source-format": "hdf5",
          "document-count": 100000
        }
      ]
    },
    {
      "name": "cohere-1m",
      "base-url": "https://dbyiw3u3rf9yr.cloudfront.net/corpora/vectorsearch/cohere-wikipedia-22-12-en-embeddings",
      "target-index": "{{ target_index_name }}",
      "documents": [
        {
          "source-file": "documents-1m.hdf5.bz2",
          "source-format": "hdf5",
          "document-count": 1000000
        }
      ]
    },
    {
      "name": "cohere-10m",
      "base-url": "https://dbyiw3u3rf9yr.cloudfront.net/corpora/vectorsearch/cohere-wikipedia-22-12-en-embeddings",
      "target-index": "{{ target_index_name }}",
      "documents": [
        {
          "source-file": "documents-10m.hdf5.bz2",
          "source-format": "hdf5",
          "document-count": 10000000
        }
      ]
    }
  ],
    "operations": [
      {{ benchmark.collect(parts="operations/*.json") }}
    ],
      "test_procedures": [
        {{ benchmark.collect(parts="test_procedures/*.json") }}
      ]
}