Abstract: A high-performance cache is disclosed. The cache is designed for time- and space-efficiency for a diverse range of information objects. Information objects are stored in portions of a non-volatile storage device called arenas, which are contiguous regions from which space is allocated in parallel. Objects are substantially contiguously allocated within an arena and are mapped by name keys and content-based object keys to a tag table, an open directory, and a directory table. The tag table is indexed by the name keys, and stores references to sets in the directory table. The tag table is compact and therefore can be stored in fast main memory, facilitating rapid lookups. The directory table is organized so that at least a frequently-accessed portion of it also usually resides in fast main memory, which further speeds lookups. The tag and directory tables are organized to quickly determine non-presence of objects.

Abstract: A method for consistently storing cached objects in the presence of failures is provided. This method ensures atomic object consistency--in the event of failure and restart, an object will either be completely present or completely absent from the cache, never truncated or corrupted. Furthermore, this consistency comes without any time-consuming data structure reconstruction on restart. In this scheme, objects are indexed by a directory table that is stored in main memory and mapped to non-volatile storage, and changes to the directory table are buffered into an open directory that is stored in main memory. Cache objects are either stored in volatile aggregation buffers or in segments of non-volatile disk storage called arenas. Objects are first coalesced into memory-based aggregation buffers, and later committed to disk. Locking is used to control parallel storage to aggregation buffers.