The first version of the storage server with protocol access has been released, designed to solve the problem of a large number of small files on file systems, including cluster ones.
Some possibilities:
- multithreading;
- multiserver, providing fault tolerance and load balancing;
- maximum transparency for the user or developer;
- supported HTTP methods: GET, HEAD, PUT and DELETE;
- controlling read and write behavior through client-side headers;
- support for flexibly configured virtual hosts;
- support for data integrity CRC when writing / reading;
- semi-dynamic buffers for minimal memory consumption and optimal network performance tuning;
- delayed data compaction;
- as an addition - a multi-threaded archiver wZA to migrate files without stopping the service.
The product is designed for mixed use, including support for working with large files without sacrificing performance.
The main recommended use on origin servers and large storages to significantly reduce the amount of metadata in clustered file systems and expand their capabilities.
Server spreads under the BSD-3 license.
Source: linux.org.ru