JP2010165004A - Data storage system and data management method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a data storage system, etc., for streamlining storage data in a physical disk and virtually generating latest data at a specified time point in the past or tabulation data during a period, based on the storage data. <P>SOLUTION: Not the whole of past data are held and only the new or updated data are held concerning the data to be stored in an analysis data storage database server. Consequently, the physical disk to be used is streamlined. Besides, the latest state data on any designated days in the past is displayed and also the number of cases within a designated period is tabulated by devising designation in retrieving the data. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for making the accumulated data in a physical disk more efficient or virtually creating the latest data at a specific point in the past from the accumulated data when using the concept of DWH (data warehouse).

  In general, in the accumulation of data for analysis, a certain period such as daily, weekly, monthly, quarterly, or annual is defined, and all data necessary for analysis is accumulated for each period, and the change of those changes And use it for free analysis such as obtaining statistics.

  However, it is essential to ensure that the physical disk capacity of the server with the database that stores data is much larger than that of the response history database (online data storage / update database). Met.

  In addition, there is a concern that restrictions on the latest data at a certain point in the past (X days, X months, X years, etc.) may occur due to disk limitations due to the above.

As a related technique, a technique has been proposed in which a DB is copied from a CD-ROM to a hard disk in advance, then difference information about the DB is distributed, and the contents of the hard disk are rewritten based on the difference information (for example, patents). Reference 1).
Also, when searching for a product DB from a user terminal, a technique has been proposed in which a CD-ROM distributed in advance is searched, and then difference information sent via the NW is searched to obtain desired data (for example, , See Patent Document 2).

JP 2003-140931 A JP 2004-240562 A

  However, the above-mentioned related technology uses the DWH (data warehouse) concept to analyze customer data and consultation / complaint content at a contact center, etc., and use it in marketing etc. Since the disk capacity of the equipment procured in the past was equal to the response history database server, the accumulated data in the physical disk can be made more efficient, the latest data at a specific point in the past than this accumulated data, or the total during the period It was difficult to create data virtually.

  The present invention has been made in order to solve the above-described problems, and improves the efficiency of the accumulated data in the physical disk, and the latest data at a specific point in the past or the aggregated data during the period from the accumulated data. An object of the present invention is to provide a data storage system and a data management method capable of virtually creating a data.

The first data storage system of the present invention does not perform data copy of all cases, but determines only the case data that has been newly added or changed since the previous copy, based on the last registration update date of the record. To extract the target,
If it is update data, the data before update is also kept on the DWH-DB side,
At the time of extraction, in the data storage system that allocates a dedicated unique code separately on the DWH-DB side,
All records that match the last update date before the specified target date are searched, and if there are duplicated items, name identification is performed with the item code, and only the latest data is extracted and displayed.

The second data storage system of the present invention does not perform data copy of all cases, but determines only the case data that has been newly added or changed since the previous copy, based on the record last registration update date. To extract the target,
If it is update data, the data before update is also kept on the DWH-DB side,
At the time of extraction, in the data storage system that allocates a dedicated unique code separately on the DWH-DB side,
All records of the initial reception date that match the specified period are searched. If there are duplicate cases, name identification is performed based on the case code and the initial reception date, and only the latest data is extracted and displayed.

In the first data management method of the present invention, the data of all cases is not copied, and only the case data that has been newly added or changed since the previous copy is copied based on the last registration update date of the record. To extract the target,
If it is update data, the data before update is also kept on the DWH-DB side,
In the data management method of allocating a dedicated unique code separately on the DWH-DB side at the time of extraction,
All records that match the last update date before the specified target date are searched, and if there are duplicated items, name identification is performed with the item code, and only the latest data is extracted and displayed.

In the second data management method of the present invention, the data of all cases is not copied, and only the case data that has been newly added or changed since the previous copy is copied based on the last registration update date of the record. To extract the target,
If it is update data, the data before update is also kept on the DWH-DB side,
In the data management method of allocating a dedicated unique code separately on the DWH-DB side at the time of extraction,
All records of the initial reception date that match the specified period are searched. If there are duplicate cases, name identification is performed based on the case code and the initial reception date, and only the latest data is extracted and displayed.

  According to the present invention, the accumulated data in the physical disk can be made more efficient, and the latest data at a specific point in the past or the total data during the period can be virtually created from the accumulated data.

It is a figure which shows the system configuration | structure which concerns on embodiment of this invention. It is a flowchart which shows the operation | movement of the whole system which concerns on embodiment of this invention.

  Hereinafter, a first embodiment of the present invention will be described in detail with reference to the drawings. The data storage system in the present embodiment shown in FIG. 1 includes a response history data input terminal 10, a response history database server 20, an analysis data storage database server 30, and an analysis terminal 40. These response history data input terminal 10, response history database server 20, analysis data storage database server 30 and analysis terminal 40 operate under program control and are connected to each other via local area network 100. Yes.

  The response history data input terminal 10 is an information processing device such as a personal computer installed in the contact center, and inputs contents such as consultations and complaints received by telephone to the contact center, and receives the contents via the network. It has a function of registering in the history database server 20. Similarly, it is possible to access the response history database server 20 to search and display registered cases or update registered data once.

  The response history database server 20 is an information processing apparatus such as a server installed in a data center, and has a relational database as middleware installed therein. It has a function to store the entire history of responses conducted at the contact center in a relational database. The data has one record per case and is uniquely identified by the case number.

  The analysis data storage database server 30 is an information processing apparatus such as a server installed in a department that performs marketing or the like, and a relational database that is middleware is installed therein. It has a function of extracting and storing data contents of the response history database server 20 outside business hours. For this reason, data is not updated during work, and is used exclusively for search and aggregation.

  The analysis terminal 40 is an information processing apparatus such as a personal computer installed in a department that conducts marketing, etc., and is an OLAP that can access the analysis data storage database server 30 and freely analyze items and time series. It has a function. Based on this analysis, for example, there is a tendency that a certain age in a certain region or season has a lot of interest in a certain product, and the data is used for marketing or customer service depending on the tendency of data.

Next, the operation of the present embodiment will be described in detail with reference to FIGS.
When receiving a phone call for consultation / complaint from a customer at a contact center or the like, the response history data input terminal 10 accesses the response history database server 20 and receives information related to the customer that matches the number of the incoming call (name, name, Address, telephone number, age, etc.) are extracted and displayed on the screen (steps A1, A2, A3). In addition, since it is not displayed in the case of a new customer, manual input is performed as a new customer.

  Thereafter, the contents of the consultation / complaint over the telephone and the contents of the answer are input (step A4) and registered in the response history database server 20. At the time of registration, the system automatically assigns the case number, the first reception date and time, and the last registration date and time (steps A5 and A6).

  The data stored in the response history database server 20 registered and updated as described above is copied to the data storage database server 30 for analysis at a time determined every day (time period when the business is not operated). The data is extracted and stored according to the following rules (steps A7, A8, A9).

  (1) The data is not copied for all items, and the record is determined based on the last registration update date of the record so as to copy only the item data that has been newly added or changed since the previous copy, and the target is extracted.

  (2) If it is update data, the data in the pre-update state is also retained on the DWH-DB side.

  (3) At the time of extraction, a dedicated unique code is allocated on the DWH-DB side (if the case is update data, the case number is duplicated and cannot be uniquely identified).

  The data stored in the analysis data storage database server 30 in this way is accessed from the analysis terminal 40 for analysis or totalization. The analysis target is extracted by the following method (steps A10 and A11). , A12, A13).

1) Displaying the latest all data on the specified date Search for all records that match the last update date before the specified target date (including the specified date). Extract and display data only. Since this extracted data becomes the latest data as of the designated date, various analyzes are performed based on this.

2) Aggregation of data within the specified period All records with the initial reception date that match the specified period are searched. If there are duplicates, the name is sorted by the case code and the initial reception date, and only the latest data is extracted and displayed. . Search or tabulate based on this.
If name identification is performed within the specified period on the last update date, it will be separately recorded if the same case is duplicated in another period, and the total number of cases does not match the number of cases, so the initial reception date is used.

  As described above, it is possible to virtually hold the latest state of the entire day from the introduction of the system without limiting the data retention period, and make it the target of analysis and aggregation.

  When data was accumulated by the above method, a trial calculation of the disk size to be used was performed as an example compared with the case where all data was retained.

The prerequisites are as follows.
・ The number of cases per month is 277,500 (cases), the number of cases per year is 2,918,000 (cases) ・ Capacity assumption per case is 5KB ・ The first registered cases are updated 30%

(1) Amount of accumulated data when all data is stored The above conditions and the following conditions are assumed.
・ For past data, date / time: 60 days, monthly: 24 months, annual: 5 years are retained ・ Delete expired data a) Response history DB daily increase 1 month projects Number ÷ 30 (days) × Capacity assumption per project
2,918,000 (cases) ÷ 360 (days) × 5 (KB) = 40527 (KB) = 39.6 (MB)
B Estimated value of past data volume (daily)
Data stored daily is accumulated for 60 days once a day, including the latest data as of the previous day.
Day 1: [39.6 (MB) x 1 (day)] = 39.6 (MB)
Day 2: [39.6 (MB) x 2 (Sun)] = 79.2 (MB)
...
Day 1741: {39.6 (MB) x 1741 (Sunday)} = 68943.6 (MB)
...
Day 1800: {39.6 (MB) x 1800 (Sun)} = 71280 (MB)
Of these, the maximum expected data is the total value from the 1741st day to the 1800th day,
68943.6 (MB) +... +71280 (MB) = 4108.1 (GB).
C Estimated value of past data volume (monthly)
As for past data to be stored monthly, data is accumulated once a month for two years (24 months).
First month: [39.6 (MB) x 30 (days)] = 1188 (MB)
Month 2: [39.6 (MB) x 60 (Sun)] = 2376 (MB)
...
Month 37: {39.6 (MB) x 1410 (Sun)} = 55836 (MB)
...
60th month: {39.6 (MB) x 1800 (Sun)} = 71280 (MB)
Of these, the maximum expected data is the total for the 37th to 60th months,
55836 (MB) + ... + 71280 (MB) = 1489.6 (GB).
D Expected value of past data volume (annual)
The past data to be saved annually is accumulated once a year for five years. First year: [39.6 (MB) x 360 (days)] = 14256 (MB)
Second year: {39.6 (MB) × 720 (Sun)} = 28512 (MB)
...
5th year: {39.6 (MB) x 1800 (Sun)} = 71280 (MB)
The maximum expected data is the total value for this 1-5 year,
14256 (MB) + 28512 (MB) + ... + 71280 (MB) = 208.8 (GB)
It becomes.
E Estimated value (maximum value) of DWH data volume
The estimated DWH data amount is obtained from the total value of the above (i) to (d).
4108.1 (GB) + 1489.6 (GB) + 208.8 (GB) = 5806.5 (GB)

(2) Accumulated data amount when the present invention is used The above conditions are assumed.
Increase in daily response history DB Number of projects per month ÷ 30 (days) × Capacity assumption per project
2,918,000 (cases) ÷ 360 (days) × 5 (KB) = 40527 (KB) = 39.6 (MB)
Expected value of data volume (initial registration data)
From Ki, the expected data to be newly added for 5 years (1800 days) will be the same result as i,
(39.6 (MB) x 1800 (day)) = 70.0 (GB).
Estimated value of data volume (update data for existing projects)
When existing response history data is updated, the data storage database server side needs to store it together with the data before update.
Multiply the result of i by the rate at which the records created are updated (30%)
70.0 (GB) x 30% = 21.0 (GB).
D Expected value of DWH data volume (maximum value)
The estimated DWH data amount is obtained from the total value of the keys.
70.0 (GB) + 21.0 (GB) = 91.0 (GB)
Therefore, in the above example, the result is 91.0 (GB) compared to the conventional method 5806.5 (GB), and the efficiency can be improved to about 1 / 63.8, and the designated date can be obtained in all days. .

  Each of the above-described embodiments is a preferred embodiment of the present invention, and various modifications can be made without departing from the scope of the present invention. For example, a process for realizing the function of each device may be performed by causing each device to read and execute a program for realizing the function of each device. Further, the program is transmitted to another computer system by a transmission wave via a computer-readable recording medium such as a CD-ROM or a magneto-optical disk, or via a transmission medium such as the Internet or a telephone line. Also good. In addition, it is also within the scope of the present invention that the functions of each device are realized by other devices collectively or the functions are distributed by additional devices.

  The present invention is effective when data analysis is performed using a data warehouse concept in a system that accumulates customer information and data that can be used for new product development, etc. It becomes.

10 Response History Data Input Terminal 20 Response History Database Server 30 Analysis Data Storage Database Server 40 Analysis Terminal 100 Local Area Network

Claims (6)

Do not copy the data of all the matters, but make a judgment based on the last registration update date of the record so as to copy only the matter data that has been newly added or changed since the previous copy, extract the target,
If it is update data, the data before update is also kept on the DWH-DB side,
At the time of extraction, in the data storage system that allocates a dedicated unique code separately on the DWH-DB side,
A data storage system that searches for all records that match the last update date before the specified target date, and if there are duplicates of items, performs a name identification with the item code and extracts and displays only the latest data. Do not copy the data of all the matters, but make a judgment based on the last registration update date of the record so as to copy only the matter data that has been newly added or changed since the previous copy, extract the target,
If it is update data, the data before update is also kept on the DWH-DB side,
At the time of extraction, in the data storage system that allocates a dedicated unique code separately on the DWH-DB side,
A data storage system that searches for all records of the initial reception date that match the specified period, and if there are duplicates of items, performs a name identification with the case code and the initial reception date, and extracts and displays only the latest data.   3. The data storage system according to claim 1, wherein a search or a total is performed based on the latest data extracted and displayed. Do not copy the data of all the matters, but make a judgment based on the last registration update date of the record so as to copy only the matter data that has been newly added or changed since the previous copy, extract the target,
If it is update data, the data before update is also kept on the DWH-DB side,
In the data management method of allocating a dedicated unique code separately on the DWH-DB side at the time of extraction,
A data management method characterized by searching for all records that match the last update date before the specified target date, and extracting and displaying only the latest data by performing name identification when the items are duplicated. Do not copy the data of all the matters, but make a judgment based on the last registration update date of the record so as to copy only the matter data that has been newly added or changed since the previous copy, extract the target,
If it is update data, the data before update is also kept on the DWH-DB side,
In the data management method of allocating a dedicated unique code separately on the DWH-DB side at the time of extraction,
A data management method characterized in that all records of the initial reception date that match the specified period are searched, and if a case is duplicated, name identification is performed with the case code and the initial reception date, and only the latest data is extracted and displayed.   6. The data management method according to claim 4, wherein a search or a total is performed based on the latest data extracted and displayed.

Images