Hello:

I am looking over Shards for the first time and I am confused by a couple things.

1) ShardAccessStrategy. I understand there is sequential and parellel access. If a query returned objects from multiple shards, what does sequential and parallel actually mean? Sequentially querying shards would not bring any results, right? The same with parallel . . . each shard is being queried independently.

2) In the Shards documentation:

http://www.hibernate.org/hib_docs/shards/reference/en/html_single/

in section 3.4, is this example implying that the shardID be stored as an attribute of an entity (Continent)?

3) In section 3.5, if my shards are "production" and "staging" (please don't ask why), my objectIDs won't necessarily be unique. I'll always be querying against one or the other. If the database sequences are identical in both databases, can I just use the normal id generator for sequences in both?

Thank you very much!

Eric

1.
sequential means that the query is applied on all shards one after the other.
parallel means that the query is applied on all shards in // and the the result will be consolidated.

2. no it implies that a Continent service can associate a shardID based on the entity id persisted, but that's a business implementation detail, you can chose your own strategy.

3. you cannot use Hibernate Shards if ids for a given entity type are not unique across all shards. So it seems you don't need Shards altogether, just two SessionFactory pointing to staging and production and pick up the right one depending on what you want to do.

thanks!