Hypertable is an open source database modeled after Bigtable, Google's massively scalable database. It goes real-time at Baidu. Hypertable runs on top of a distributed file system. It supports HDFS, MapR, Ceph, KFS, and local. It is developed by C++.
Hypertable was released by the company Zvents at 2008, and Doug Judd promoted to the Hypertable. Baidu, the search engine in China, became project sponsor in January 2009. The development ended in March 2016.
Hypertable provides the Hypertable Query Language (HQL) to create, modify, and query tables. HQL can also be used to invoke administrative commands. HQL can be interpreted by hypertable command line interface (ht shell), Thrift API methods, and Hypertable::HqlInterpreter C++ class.
Hypertable provides snapshot isolation for queries with 8-byte timestamps.
Hypertable is able to run on top of any filesystem. A File System (FS) broker process all filesystem requests. FS brokers currently support HDFS, MapR, Ceph, KFS, and local.
Hypertable backup by outputting table data in random order. Hypertable will always be back into a consistent and operational state at the checkpoint.
The architecture of the Hypertable is shown below.
Hyperspace - Hyperspace is a lock manager. It contains a filesystem to store some metadata. Equivalent to Google's Chubby service.
Master - The Master handles all the meta operations. It also detects range server failures. It can re-assign ranges if necessary.
Range Server - Range servers handle the reading and writing of data.
FS Broker - File System (FS) broker process all filesystem requests. FS brokers currently support HDFS, MapR, Ceph, KFS, and local.
ThriftBroker - ThriftBroker provides the application interface. It is implemented with Apache Thrift. It currently supports Java, PHP, Ruby, Python, Perl, and C++.