Can milvus+haystack be used for cluster deployment？

[mc112611]

I would like to ask how much text storage Haystack supports, while ensuring retrieval accuracy and speed.
So, if it is in commercial applications, it must be inseparable from cluster deployment.
I saw the recently updated official documentation for Milvus, which seems to support cluster deployment and is very simple to deploy.

Install Milvus Cluster

Or perhaps I misunderstood before.

Based on this code.

stores/milvus-documentstore/milvus_documentstore/milvus.py

The cluster deployment of Milvus is still a Docker deployment. So this parameter named uri is the main entry point of the vector database. Just configure the milvus cluster in advance.

So, I would like to inquire about deploying a cluster vector database using Milvus. Is it compatible with Haystack

[vblagoje]

Hey @mc112611 yes it should be compatible. Did you run into some issues?

[mc112611]

That's really great.
Unfortunately, I currently do not have any additional clustered computers for me to test Milvus+Haystack.
However, Haystack's compatibility with Milvus cluster deployment is really great.
By the way, I would like to ask a question about Haystack copyright.
Can I conduct secondary development based on the official Haystack code. For example, I modified some functions. And publish it on github. Is this behavior allowed.
Of course, I will indicate the address of the original project, which is Haystack's Github homepage.