From fa39f85a005b33929304e0410deb045158be5070 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Erik=20S=C3=B8e=20S=C3=B8rensen?= Date: Sun, 8 Jan 2012 23:53:53 +0800 Subject: [PATCH] README: Explain 'LSM' and provide link to paper. --- README.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index 7bf8189..767e591 100644 --- a/README.md +++ b/README.md @@ -2,7 +2,9 @@ # LSM B-Tree Storage -This Erlang-based storage engine may eventually provide an alternative to Basho Bitcask and Google's LevelDB. Of those two, `lsm_btree` is closer to LevelDB in operational characteristics, except it uses fewer file descriptors than LevelDB, is not as performant, but it is implemented in just ~1000 lines of Erlang. For some the benefit of having a clean and simple Erlang implementation make it worth it. +This Erlang-based storage engine implements “Log-structured merge-trees” — a storage structure described in [this paper](http://staff.ustc.edu.cn/~jpq/paper/flash/1996-The%20Log-Structured%20Merge-Tree%20%28LSM-Tree%29.pdf). + +The storage engine may eventually provide an alternative to Basho Bitcask and Google's LevelDB. Of those two, `lsm_btree` is closer to LevelDB in operational characteristics, except it uses fewer file descriptors than LevelDB, is not as performant, but it is implemented in just ~1000 lines of Erlang. For some the benefit of having a clean and simple Erlang implementation make it worth it. Here's the bullet list: