[Parent task for all related tasks]
# Current status
An object storage design was [described](https://wiki.softwareheritage.org/wiki/A_practical_approach_to_efficiently_store_100_billions_small_objects_in_Ceph) and benchmarks were written T3149 to verify it is efficient (space and speed) for the intended use cases.
* July 16th: run a benchmark during 2 days T3421
* August: run a benchmark during 9 days T3422
* August: publish the benchmark results ([draft](https://hedgedoc.softwareheritage.org/NMh_kgmlR6eWdR2QpgcQfA#))
# Design
* [Design document for the object storage](https://wiki.softwareheritage.org/wiki/A_practical_approach_to_efficiently_store_100_billions_small_objects_in_Ceph)
* [Implementation notes](https://hedgedoc.softwareheritage.org/NK1sq6dmS2uvtmtHP4wFKg?view)
# Discussions
* [Object storage: benchmark updates & preliminary hardware architecture](https://sympa.inria.fr/sympa/arc/swh-devel/2021-05/msg00034.html)
* [Object storage: horizon 2022](https://sympa.inria.fr/sympa/arc/swh-devel/2021-04/msg00012.html)
* [A good name for the object storage](https://sympa.inria.fr/sympa/arc/swh-devel/2021-03/msg00055.html)
* [A practical approach to efficiently store 100 billions small objects in Ceph](https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/DRZ6XP45PEXKMYJSSYSDXC2RGJ3BW75H/)
* [Ceph Quincy CDS & immutable objects](https://lists.ceph.io/hyperkitty/list/dev@ceph.io/message/4MLOSBIIDZV6Z2PIDHJR4YGX3TSSIHEW/)