US 9,811,465 B2
Computer system and cache control method
Kazuhide Aikoh, Tokyo (JP); and Keisuke Hatasaki, Tokyo (JP)
Assigned to Hitachi, Ltd., Tokyo (JP)
Appl. No. 14/412,335
Filed by Hitachi, Ltd., Chiyoda-ku, Tokyo (JP)
PCT Filed Jul. 2, 2013, PCT No. PCT/JP2013/068135
§ 371(c)(1), (2) Date Dec. 31, 2014,
PCT Pub. No. WO2014/007249, PCT Pub. Date Jan. 9, 2014.
Claims priority of application No. PCT/JP2012/004360 (WO), filed on Jul. 5, 2012.
Prior Publication US 2015/0161051 A1, Jun. 11, 2015
Int. Cl. G06F 12/00 (2006.01); G06F 12/0831 (2016.01); G06F 9/455 (2006.01); G06F 12/0811 (2016.01); G06F 12/0817 (2016.01); G06F 12/084 (2016.01); G06F 12/0871 (2016.01); H04L 29/08 (2006.01)
CPC G06F 12/0833 (2013.01) [G06F 9/45558 (2013.01); G06F 12/084 (2013.01); G06F 12/0811 (2013.01); G06F 12/0817 (2013.01); G06F 12/0871 (2013.01); G06F 2212/1044 (2013.01); G06F 2212/1048 (2013.01); G06F 2212/283 (2013.01); G06F 2212/62 (2013.01); H04L 67/1002 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A computer system comprising:
a plurality of nodes, wherein
the plurality of nodes include an I/O (Input/Output) node and a plurality of computation nodes which are coupled to the I/O node and send an I/O request to the I/O node,
the I/O node includes a first storage device, which stores data written or read according to the I/O request and a first memory device on which a first cache area is based to temporarily store the data written in the first storage device or read from the first storage device,
one of the computation nodes includes a second memory device on which a second cache area is based to temporarily store the data according to the I/O request,
at least one of the I/O node and the one of the computation nodes stores management information containing physical storage area information on a physical storage area in the cache area of the other of the I/O node and the one of the computation nodes, and virtual storage area information on a virtual storage area associated with the physical storage area and having a part of its own cache area, and
a target node in the plurality of nodes controls arrangement of data stored in the second cache area based on the stored management information,
the target node is the computation node,
the computation node includes a second storage device, uses a part of the second storage device as a part of the second cache area, and uses a plurality of storage areas in the second cache area based on a plurality of priorities corresponding to the plurality of the storage areas, and
in the second cache area, an area based on the second memory device of the computation node has the highest priority, an area based on the second memory device of another computation node has the second highest priority, an area based on the second storage device of the computation node has the third highest priority, and an area based on the second storage device of the other computation node has the fourth highest priority.