Although s3 has high availability within a region, it is not inconceivable that an entire region could go down.
This requires us to bring up another cluster in another region using another bucket. Is it a possible configuration under EON?
Eon Mode (v11.0) uses a single communal storage location for all data and the catalog metadata.
You can bring up another primary cluster in another region using another bucket only to serve as independent separated cluster,
Or maintain as DR cluster with fork loading.
Thanks, Mosheg, how fast is the sync between s3s in different regions, in your experience?
If one region did go down, should we expect some level of data loss?
No S3 sync is required if you fork load the same data into two separated databases in two separated independent clusters.
If one region goes down, there will be no impact on the other cluster in the other region.
After the region will be available again, either load the relevant data again or transfer the missing data from the other cluster.