Buffertars

Buffertars is a term used in computing to describe an adaptation of tar archives that employs an in-memory buffering layer to improve write performance. In practice, buffertars refer to archives created by software that buffers data before writing to the tar file, with the goal of increasing throughput on slower storage devices or high-latency networks. The concept is not standardized and appears mainly in niche discussions or experimental tooling.

Origin and usage: The term is a portmanteau of buffer and tar and has no single formal specification. Some projects describe buffertars as streaming archives that accumulate data in memory before flushing in large blocks to disk or over a network. Because implementations vary, behavior and compatibility with traditional tar tools can differ.

Technical characteristics: Common elements include a user-configurable buffer size, in-memory accumulation of input data, optional compression of the tar stream, and integrity checks such as checksums or hashes. Support for multi-volume archives and streaming from multiple sources is sometimes provided. The approach may require careful handling of memory usage and flush semantics to avoid data loss on crashes or power failures.

Applications and considerations: Buffertars are discussed in contexts such as backup pipelines and large-scale data transfers where I/O latency dominates. Advantages include potentially higher throughput and more efficient disk writes; drawbacks include higher RAM usage, added complexity, and potential compatibility issues with tools that expect standard tar streams.

Related topics include tar, buffering, data backup, and archive formats.