Information processing apparatus using data correlation for data storage, computer-readable recording medium recording storage control program using data correlation for data storage, and storage system using data correlation for data storage
AssigneeFujitsu Limited
InventorTaketoshi Yoshida
Life of this patent
Estimating the term…
Abstract
An information processing apparatus includes: a memory; and a processor coupled to the memory and configured to: perform control to store data on the memory; determine a correlation between a plurality of pieces of data, based on pieces of attribute information of the plurality of pieces of data, wherein dispose pieces of data having a high correlation among the plurality of pieces of data at neighboring locations on the memory.
Claims9
An information processing apparatus comprising:
- a memory;
- and
- a processor coupled to the memory and configured to:
- perform control to store data on the memory;
- determine a correlation between a plurality of pieces of data, based on pieces of attribute information of the plurality of pieces of data;
- and
- dispose pieces of data having a high correlation among the plurality of pieces of data at neighboring locations on the memory, the memory is a sequential access medium, the processor is configured to:
- determine the correlation between the plurality of pieces of data, using at least one of a user-related weight indicating a correlation which is assigned to each pair of users included in a plurality of users and a process-related weight indicating a correlation which is assigned to each pair of processes included in a plurality of processes executed in response to requests made by the plurality of users;
- determine an order such that the plurality of pieces of data are disposed on the memory in descending order of the correlation;
- correct the user-related weight and generate a corrected user-related weight by calculating, for all pairs of the users, an absolute value of a difference between values of identifiers of the users of the respective pairs of the users, summing differences of the all pairs of the users to obtain a first summed difference and multiplying the user-related weight by a value obtained by dividing the user-related weight by the first summed difference and adding a specific value to a first divided result;
- and
- correct the process-related weight and generate a corrected process-related weight by calculating, for all pairs of the processes, an absolute value of a difference between values of identifiers of the processes of the respective pairs of the processes, summing differences of the all pairs of the processes to obtain a second summed difference and multiplying the process-related weight by a value obtained by dividing the process-related weight by the second summed difference and adding the specific value to a second divided result.
A non-transitory computer-readable recording medium having stored therein a storage control program for causing a computer to execute a process, the process comprising:
- determining a correlation between a plurality of pieces of data, based on pieces of attribute information of the plurality of pieces of data;
- and
- disposing pieces of data having a high correlation among the plurality of pieces of data at neighboring locations on a storage medium, the memory is a sequential access medium, the process further includes:
- determining the correlation between the plurality of pieces of data, using at least one of a user-related weight indicating a correlation which is assigned to each pair of users included in a plurality of users and a process-related weight indicating a correlation which is assigned to each pair of processes included in a plurality of processes executed in response to requests made by the plurality of users;
- determining an order such that the plurality of pieces of data are disposed on the memory in descending order of the correlation;
- correcting the user-related weight and generating a corrected user-related weight by calculating, for all pairs of the users, an absolute value of a difference between values of identifiers of the users of the respective pairs of the users, summing differences of the all pairs of the users to obtain a first summed difference and multiplying the user-related weight by a value obtained by dividing the user-related weight by the first summed difference and adding a specific value to a first divided result;
- and
- correcting the process-related weight and generating a corrected process-related weight by calculating, for all pairs of the processes, an absolute value of a difference between values of identifiers of the processes of the respective pairs of the processes, summing differences of the all pairs of the processes to obtain a second summed difference and multiplying the process-related weight by a value obtained by dividing the process-related weight by the second summed difference and adding the specific value to a second divided result.
A storage system comprising:
- an application server configured to acquire data from a user;
- a proxy server configured to store pieces of attribute information of a plurality of pieces of data;
- and
- a storage control apparatus configured to perform control o store data on a memory, wherein the application server is configured to transmit the data which is acquired from the user to the storage control apparatus via the proxy server, and the storage control apparatus is configured to:
- determine, based on attribute information of the received data, with reference to the pieces of attribute information stored in the proxy server, correlations between the received data and the plurality of pieces of data;
- and
- dispose pieces of data having a high correlation at neighboring locations on the memory,
- the memory is a sequential access medium,
- the storage control apparatus is configured to:
- determine the correlation between the plurality of pieces of data, using at least one of a user-related weight indicating a correlation which is assigned to each pair of users included in a plurality of users and a process-related weight indicating a correlation which is assigned to each pair of processes included in a plurality of processes executed in response to requests made by the plurality of users;
- determine an order such that the plurality of pieces of data are disposed on the memory in descending order of the correlation;
- correct the user-related weight and generate a corrected user-related weight by calculating, for all pairs of the users, an absolute value of a difference between values of identifiers of the users of the respective pairs of the users, summing differences of the all pairs of the users to obtain a first summed difference and multiplying the user-related weight by a value obtained by dividing the user-related weight by the first summed difference and adding a specific value to a first divided result;
- and
- correct the process-related weight and generate a corrected process-related weight by calculating, for all pairs of the processes, an absolute value of a difference between values of identifiers of the processes of the respective pairs of the processes, summing differences of the all pairs of the processes to obtain a second summed difference and multiplying the process-related weight by a value obtained by dividing the process-related weight by the second summed difference and adding the specific value to a second divided result.
Dependencies are inferred from the claim text.