Current Dashboard exhibits some "potential" scaling bottlenecks with:
- OSDs: some requests' payloads increase linearly with OSD count. As these requests (/api/osd) are not paginated, they may potentially generate 1 MB/request with 500 OSDs. This endpoint is refreshed every 5 s per dashboard instance (a background Chrome tab running Ceph dashboard would consume 17 GB/day).
- Other endpoints could also retrieve large/unbounded amounts of data (config options, RBD images, etc, RGW buckets).

This BZ should serve as a tracking one for other BZs covering remediation/fixes of this scalability issues:
- Network compression.
- Caching
- Pagination
- Endpoint Refactoring