Distributed storage system and storage control method
A distributed storage system includes one or a plurality of storage units including a plurality of physical storage devices, and a plurality of computers connected to the one or plurality of storage units via a communication network. When receiving a write request for a logical volume, the computer writes write data corresponding to the write request and redundant data for making the write data redundant in a plurality of physical storage devices of the storage unit in a distributed manner, and collectively controls writing of a journal of write data for managing a write history of the write data and a journal of redundant data for managing a write history of the redundant data.
1. A distributed storage system comprising:
a plurality of storage units that each include a plurality of physical storage devices; and
a computer connected to the plurality of storage units via a communication network, wherein
when receiving a write request for a logical volume, the computer writes write data corresponding to the write request and redundant data for making the write data redundant in the plurality of storage units in a distributed manner, and collectively controls writing of a journal of write data for managing a write history of the write data and a journal of redundant data for managing a write history of the redundant data, and
when receiving the write request, the computer generates a guarantee code for confirming that the write data does not change until the write data is read, and writes the write data and a corresponding guarantee code in a continuous area in a respective physical storage device.
2. The distributed storage system according to claim 1 , wherein
when holding the write data in a temporary storage area, the computer provides a free area for holding the guarantee code, generates the corresponding guarantee code from the write data held in the temporary storage area and holds the same in the free area, and writes the write data and the guarantee code held in the temporary storage area in the respective physical storage device.
3. The distributed storage system according to claim 1 , wherein
a response of write completion to a request source of the write request is made after completion of writing of the journal of the write data and the journal of the redundant data, and
when a failure occurs in the computer, a second computer takes over a role of the computer in which the failure occurs using the write data, the redundant data, the journal of the write data, and the journal of the redundant data.
4. The distributed storage system according to claim 1 , wherein
the computer further sets, as the redundant data, a same duplicated data as the write data.
5. The distributed storage system according to claim 1 , wherein
the computer uses erasure coding of applying a mathematical function to the write data to generate the redundant data.
6. A method for operating a plurality of storage units, the method comprising:
writing, when receiving a write request for a logical volume, write data corresponding to the write request and redundant data for making the write data redundant in the plurality of storage units in a distributed manner;
generating, when receiving the write request, a guarantee code for confirming that the write data does not change until the write data is read, and writes the write data and a corresponding guarantee code in a continuous area in a respective physical storage device; and
collectively controlling writing of a journal of write data for managing a write history of the write data and a journal of redundant data for managing a write history of the redundant data.