...
- Idle standbys versus homogeneous load
- Parameters - replicas (z) and shards; how do we choose them? On what granularity are they?
- Major failures: How long to bring up if we lose DC power?
- Reconstituting failed backup disks? Don't?
- Adds to the reason for more, smaller disks in a single machine
- More heads = more write bandwidth per node
- If we partition the shards to only log to a specific disk then we only have to replicate 1/4 as many shards when a single disk fails
- The downside, of course, is that disk failures will occur more often, and cost
- All backups must be able to initiate reconstruction. Why?
- Flash may be worth another look, particularly if we can play games in the FTL.
...