We currently utilize Hazelcast in a non-clustered environment alongside Spring Cache Abstraction. However, we now intend to transition into a clustered environment and require our cache to be replicated across multiple machines. The challenge we're facing is that our cached entities involve one-to-many relationships, resulting in slow performance.
For instance, consider the following example:
class Student {During the serialization process, all lazy fields are being serialized. Since our entities have deep relationships and sometimes even circular references, this leads to a significant amount of data being transmitted over the network in the clustering setup.
Is there a configuration setting in Hazelcast that allows us to bypass the serialization of lazy fields?