0

I run GridDB on GridDB Cloud (Free plan) with a single cluster and a TimeSeries workload.

Environment

  • GridDB Cloud (Free), single DB

  • TimeSeries containers (e.g., TSDB with ts TIMESTAMP PRIMARY KEY)

    Continuous ingest (~tens of writes/sec) + concurrent reads

    Late-arriving inserts are possible

Requirement

  • Take a consistent backup while writes continue (no long maintenance window)
  • RPO ≤ 15 minutes, RTO ≤ 1 hour
  • Acceptable write pause ≤ ~30 seconds if necessary

I couldn’t find a clear, Cloud-specific procedure that guarantees point-in-time consistency across containers while the system is under write load.

Question: What is the supported way on GridDB Cloud to take a consistent online backup (or snapshot) of the whole database while writes continue—so that a restore yields a self-consistent state at a specific point in time?

If the answer is “no true online snapshot,” I’d appreciate the recommended operational sequence that meets the above RPO/RTO—for example, whether to:

  • quiesce writers briefly, force a checkpoint, then run a logical export (and how to do that safely), or
  • rely on a Cloud-provided snapshot mechanism (if any) that’s transactionally consistent, or
  • another supported approach.

I’m looking for a concrete, documented sequence (or official reference) I can run in production on GridDB Cloud.

Improve this question

0

Reset to default

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.