Hi all! Is it possible to organize "one-way" replication inside cluster? My idea is to speed-up search by dividing my data to "fast" group (for example in that group all document will be cleaned up after one day by timestamp) and "long-term" storage (for example 1 year) . I want to store the same document with the same _id in two replicas, but clean one of replic every day, still having same document in "long-term" replica without cleaning?
Thanks in advance!
No, that is not possible. If you can describe your data and use case someone might be able to give other suggestions though.