Technical articles about high-performance object storage, system design, and engineering insights from the FractalBits team.
As databases, streaming systems, and AI pipelines move their live data onto object storage, GB/s no longer tells you enough. The number that matters is objects per second at a small, stated size.
How we used pre-allocation, O_DIRECT, and an SSD-aware journal to keep our local storage engine's writes crash-consistent without fsync.
How FractalBits uses cloud object storage as a bootstrap-time coordination plane to solve the cluster startup chicken-and-egg problem.
Why we replaced RocksDB with Fractal ART, an on-disk adaptive radix tree for object storage metadata.
High-performance object storage exists, but the economics make it unusable at scale. FractalBits breaks out of the high-performance trap.