摘要:
The present invention provides high-speed copying of a compressed data volume.The control unit of the storage apparatus divides the pool into a plurality of chunks comprising a plurality of pages storing data, compresses data which is written to the logical volume by the host and assigns one of the plurality of chunks to a compressed data logical volume which stores the compressed data, and, when the compressed data logical volume is copied, the control unit makes the page length of the chunk which is assigned to the compressed data logical volume which is the copy source the same as the page length of the chunk which is assigned to the compressed data logical volume which is the copy destination.
摘要:
Storage apparatus and data management methods provide for high-speed copying of a compressed data volume. A control unit of the storage apparatus divides a pool into a plurality of chunks including a plurality of pages storing data, compresses data which is written to a logical volume by a host, and assigns one of the plurality of chunks to a compressed data logical volume which stores the compressed data. When the compressed data logical volume is copied, the control unit makes a page length of the chunk assigned to the compressed data logical volume that is a copy source equivalent to a page length of the chunk assigned to the compressed data logical volume that is the copy destination.
摘要:
The present invention creates a pool comprising tiers corresponding to the performance of logical volumes.A first storage apparatus either acquires or measures the performance of multiple external pool volumes of an external storage (S1711 through S1714), and creates a pool volume group based on the performance of the external pool volumes (S1715, S1716). The rank order of a tier is decided in accordance with the performance of each pool volume group (S1717, S1718). This makes it possible to relatively easily create a pool comprising multiple appropriate tiers that are suited to the current situation.
摘要:
A storage system in an embodiment of this invention comprises a non-volatile storage area for storing write data from a host, a cache area capable of temporarily storing the write data before storing the write data in the non-volatile storage area, and a controller that determines whether to store the write data in the cache area or to store the write data in the non-volatile storage area without storing the write data in the cache area, and stores the write data in the determined area.
摘要:
The present invention creates a pool comprising tiers corresponding to the performance of logical volumes.A first storage apparatus either acquires or measures the performance of multiple external pool volumes of an external storage (S1711 through S1714), and creates a pool volume group based on the performance of the external pool volumes (S1715, S1716). The rank order of a tier is decided in accordance with the performance of each pool volume group (S1717, S1718). This makes it possible to relatively easily create a pool comprising multiple appropriate tiers that are suited to the current situation.
摘要:
A storage system in an embodiment of this invention comprises a non-volatile storage area for storing write data from a host, a cache area capable of temporarily storing the write data before storing the write data in the non-volatile storage area, and a controller that determines whether to store the write data in the cache area or to store the write data in the non-volatile storage area without storing the write data in the cache area, and stores the write data in the determined area.
摘要:
When a snapshot virtual volume is provided to the host as an OS image of a virtual machine in a system where a single V-VOL is used by a single VM and all the VMs are started concurrently, burdensome Copy-on-Write (CoW) accesses placing heavy I/O loads on the storage occur in concentrated manner, and the starting time is elongated. The present invention solves the problem by measuring the I/O pattern (number of IO per unit time for reading/writing) in page units during starting of the system prior to having the VMs started concurrently, and based on the measurement results, performs saving and copying of the target pages of the write access to a snapshot pool prior to starting the VM. This preliminary saving enables to reduce the CoW accesses having a high access load, and to enable reduction of the VM starting time and efficient use of the pool capacity.
摘要:
A host I/F unit has a management table for managing an MPPK which is in-charge of the control of input/output processing for a storage area of an LDEV, and if a host computer transmits an input/output request for the LDEV, the host I/F unit transfers the input/output request to the MPPK which is in-charge of the input/output processing for the LDEV based on the management table, an MP of the MPPK performs the input/output processing based on the input/output request, and the MP of the MPPK also judges whether the MPPK that is in-charge of the input/output processing for the LDEV is to be changed, and sets the management table so that an MPPK which is different from the MPPK that is in-charge is to be in-charge of the input/output processing for the LDEV.
摘要:
A storage system 200 has a storage device 240 providing a plurality of logical volumes 250 used as data storage areas of a host 100 and a plurality of MPPKs 210 executing data IO processes for the logical volumes 250 from the host 100 and, when it is determined whether the MPPKs 210 correlated with the logical volumes 250 must be changed based on a storage function such as local copy set for the logical volumes 250, if it is determined that the change is necessary, the usage rates of processors 212 of the MPPKs 210 are acquired and compared with each other and if it is determined that the comparison result satisfies a predetermined condition, the process cost for the logical volumes 250 is calculated for each of the responsible MPPKs 210 to determine the destination MPPK 210 such that the costs of the MPPKs 210 are equalized.
摘要:
A copy source storage controller received write data added with a time and issued from a host computer transfers the write data with the time to a copy destination storage controller. If there are a plurality of copy destination storage controllers, a representative copy destination storage controller compares times of write data copied to the plurality of copy destination storage controllers, and writes the write data in copy destination logical volumes in the sequential order of time. The representative copy destination storage controller judges that integrity of the write data is established, if a communication procedure is established with the copy destination storage controller and if the statuses of the copy source/destination logical volumes are coincident. In remote copy which guarantees integrity of write data and traverses a plurality of storage controllers, it is possible to judge at an optional time point whether integrity of write data can be guaranteed.