Rabata·Nov 3Inside Rabata S3: how our S3-compatible storage works, from metadata to dead disksThree services, one atomic script per S3 operation, range sharding, Reed–Solomon 5+2, and failures that never reach users.
Rabata·Oct 29Dead disks, no errors: how we keep failures away from usersRabata is S3-compatible object storage running on our own hardware in the EU and the US. In this four-part series our engineers explain how…
Rabata·Oct 29Same durability, less than half the disk: how we pick redundancy per fileRabata is S3-compatible object storage running on our own hardware in the EU and the US. In this four-part series our engineers explain how…
Rabata·Oct 27No hashing, no fan-out: how we shard S3 metadata for ListObjectsRabata is S3-compatible object storage running on our own hardware in the EU and the US. In this four-part series our engineers explain how…
Rabata·Oct 27Zero bytes of your data: how our S3 layer keeps only metadataRabata is S3-compatible object storage running on our own hardware in the EU and the US. We built the stack ourselves, from the S3 API down…