Hi, can you tell me more about your use case? We can enable you to run sync on a sharded cluster, but it is more likely than not that if you are trying to sync 4TB of data then sync might not be able to handle that load. Generally, our customers that have that much data either:
Use a multi-tenant approach (they have customers and each customer is its own cluster / app)
Use a app-per-region approach
Are only syncing on a subset of their actual data (in this case 4TB)
Happy to chat more about the specifics about what you are trying to build with sync,
Thanks for the info. We’re currently a offline productivity app that includes note taking and scribble, but we’re trying to add collaboration support to it, therefore it’s almost certain that the total data size will exceed 4TB based on our current calculation, and this is exclusive of the realm ops history data size.
Each user should only receive a subset of the actual data though, at most a few hundreds MBs based on our current data. However, as we’re building collaboration support where one user needs to share notes with other users, so it doesn’t sound like the multi-tenant or app-per-region approach could work, since realm data cannot sync across different apps.