hypertable-0.9.2.7-alpha-src.tar

Hypertable 是一个正在进行中的开源项目，以google的bigtable论文为基础指导，使用c++语言实现。目的是为了解决大并发，大数据量的数据库需求。目前只支持最基本的查询功能，对于事物，关联查询等都不支持。对单条查询的响应时间可能也不如传统数据库（要看数据量，量越大，对hypertable越有力）。好处是，可以处理大量并发请求，和管理大量数据。可扩缩性好，扩容只需要增加集群中的机器就ok了。任何节点失效，既不会造成系统瘫痪也不会丢失数据。在集群节点足够的情况下，并发量和数据量对性能基本没有影响。

-Hypertable stores data in a table, sorted by a primary key. There is no typing for data in the cells, all data is stored as uninterpreted byte strings. Scaling is achieved by breaking tables in contiguous ranges and splitting them up to different physical machines. There are two types of servers in a Hypertable cluster, Range Servers which hold a piece (or range) of the data and Master Servers which handle administrative tasks and oversee the Range Servers. A single physical machine can run both a Range Server and Master Server process. A single Range Server may hold many discontinuous ranges, the Master Server is responsible for farming them out in an intelligent way. If a single range fills up, the range is split in half and reallocated. The top half of the range remains and the lower half is reassined to a new Range Server by the master. The default maximum range size is 200MB. If an entire Range Server fills up, the Master migrates ranges from the filled server to less full ones. T