Milvus runs on machines with or without GPU. To use Milvus on machines without a GPU, install CPU-only Milvus. Otherwise, if you have CUDA-enabled GPUs in your computer, it is recommended that you use GPU-enabled Milvus to achieve better search performance for larger queries.
In Milvus, the vector search includes 2 separate processes: index building and search. For GPU-enabled Milvus, these two processes can run concurrently, which facilitates more efficient query, especially for incremental data. For CPU-only Milvus, search computation can only be executed after index building is completed, which makes it more suitable for static data.
Docker is the easiest and recommended way to install and run Milvus.
If you cannot use your host to acquire Docker images and configuration files online because of network restrictions, please acquire them online with another available host and transfer the files offline. Refer to installation steps to learn how to acquire a Docker image. The following steps use
milvusdb/milvus as the Docker image tag.
Save the Docker image to a tar file and use appropriate methods to transfer the tar file.
$ docker save milvusdb/milvus > milvus_image.tar
After transferring the tar file, use the following command to load the tar file to a Docker image.
$ docker load < milvus_image.tar
You can download configuration files and use appropriate methods to transfer the configuration files. Refer to installation steps for details.
Refer to Build Milvus from source to learn how to build Milvus from source.