JPH01200453A - Internal data control method - Google Patents

Abstract

PURPOSE:To make a cluster having high possibility of application remain in an internal memory as much as possible by sending out the cluster having low mapping priority order to an external memory at and after the head cluster of a matrix in case the internal memory has no idle cluster. CONSTITUTION:The input/output of data is carried out between an internal memory INM and an external memory ETM for each cluster having the fixed length. The information which controls each cluster of the memory INM contains the mapping priority order showing the degree of difficulty for send-out of cluster to the memory ETM and a matrix showing the degree of difficulty for send-out cluster to the memory ETM for each mapping priority order. When no idle cluster is included in the memory INM, the cluster are sent out in order to the memory ETM at and after the head cluster of a matrix having a low mapping priority order. Thus, the file send-out order is controlled regardless of the file defining order and the used cluster is set again at the end of the matrix. As a result, the cluster having high possibility of application can be stored in the memory INM as much as possible.

Description

DETAILED DESCRIPTION OF THE INVENTION <Field of Industrial Application> The present invention relates to an internal data management method in a system capable of memory management using a virtual memory method, such as an automatic programming device.

<Prior art> Application cage of automatic programming device that automatically creates NC programs? An application program (simply called an application) needs to handle a large amount of data that exceeds the capacity of internal memory (real memory). For this reason,
In automatic programming devices with randomly accessible external memory (e.g. hard disk),
With the virtual memory method, a part of the external memory is used as an extension of the internal memory (virtual memory area), and data on the internal memory can be temporarily moved to the virtual memory area, or data on the virtual memory area can be used as needed. Recovery on internal memory has been done accordingly.

Then, data is purged and recovered in file units between the internal memory and external memory.

For example, if you run out of free space while adding data to a file in internal memory, the file will be removed using the following procedure. That is, (i) Closed files are ejected in order from the first file, and each time a file is evicted, it is checked whether the size is OK. At this time, your own files are excluded.

fii) - If the above is not enough, remove the open files from the first file around 1, and
Checks whether the size is OK every time a file is removed. At this time, your own files are excluded.

(iii), (i), and (ii) if the size is not sufficient, then the own file is ejected. If the size is not obtained even with 6ψ, (i) to G11), an overbuff 4- error is assumed.

<Problem to be solved by the invention> However, in the conventional method of evicting files in units of files, the order in which files are evicted follows the originally defined order, so it is not possible to specify the priority when executing the evicting, and furthermore, it is only possible to evict files in units of files. Therefore, used data is often evicted, external memory access increases, and processing efficiency decreases.Also, even when using multiple files at the same time, if you focus on a particular moment, you will not be able to see what is actually needed. Only a small amount of data in a file is used, and there is a high possibility that data that has been used once will be used again in the near future, but with conventional methods, it is possible to leave part of the file in internal memory. This results in an increase in external memory accesses.

From the above, the purpose of the present invention is to introduce the concept of clusters, to perform data eviction and recovery in cluster units rather than file units, and to give each cluster in the internal memory a priority that indicates how difficult it is to be evicted. With this, you can create a highly usable cluster! An object of the present invention is to provide an internal data management method that allows data to be stored in an internal memory.

Means for Solving the Problems> FIG. 1 is an explanatory diagram of an internal data management method according to the present invention.

ETM is external memory, INM is internal memory, PGA is page area in internal memory, vCT is management information area, VM
A is a virtual memory area.

Function> Data is input/output between the internal memory TNM and the external memory ETM in fixed-length cluster units, and management information (PMAP) that manages each page P of the page area PGA (page length = cluster length) .

MPQ) is prepared in the management information area VCT, and each page P
The external memory ET of the cluster mapped for ,
A mapping priority of O to 15 representing the difficulty of being evicted to M, and a chain saw of the mapping priority having a matrix indicating the difficulty of being evicted to the external memory ETM of the cluster,
When there are no free pages in the page area PGA, data is externally memorized in cluster units starting from the first page of the matrix with the lowest mapping priority! Kick it out to J (ETM).

<Embodiment> FIG. 1 is an explanatory diagram of an internal data management method according to the present invention.

ETM is an external memory such as a hard disk or floppy disk, and a part of it is the virtual memory area V.
It is MA. The virtual memory area VMA can exchange data with the internal memory INM in cluster units, and has multiple storage areas C3-CN (cluster areas O to CN) of the cluster length.
It has a cluster area N).

INM is a real memory (internal memory) provided inside the system main body, PGA is a page area, and VCT is a management information area. The page area PGA contains many pages P0~
It consists of PM (page θ to page M), and each page stores data in units of clusters. Note that the page length=cluster length.

The management information area VCT includes (a bitmap FAT indicating the free status of each page P, ~PM of the ill page area PGA, and (11) free space of the storage area (cluster area) C0~CN in cluster units of the virtual memory area VMA. A pit map CAT indicating the status, and 611) each page P0 to P1 of the page area PGA. ．． A page map PMAP having management information PC, (i=o~M) corresponding to (each cluster area C0~ of the ivl virtual memory area VMA)
When setting a cluster map CMAP having management information CC1 to CCN corresponding to CN, and (v) mapping priorities (0 to 15) representing the difficulty of expelling clusters on each page, each of the mapping priorities ( 0 to 15), a matrix indicating the difficulty of eviction of the cluster is further formed, and a mapping priority queue MPQ ( (see Figure 2).

Page map P, each page P of MAP, (i=1 to M)
As shown in FIG. 3, the management information PC provided in correspondence with (a) the mapping priority level MPR indicating the difficulty of expelling the clusters on page P, and (b) the mapping priority level MPR indicating the difficulty of expelling the clusters on page P1. virtual memory area V
When expelling to MA, the cluster number CLN instructs which cluster area to expel, and the matrix having the same mapping priority as (,1 page P), the page P
, a forward link NXP indicating the next page to be connected to, and (d) a reverse link PRP indicating the previous page to which page P1 is connected in the matrix having the same mapping priority as page P1. It is.

Also, each cluster area C(j
The management information CC provided corresponding to Cluster Area C (=1 to N) includes a page number that instructs to which page of the page area PGA the data in units of clusters stored in the cluster area C is to be recovered. There is.

Note that each management information CC is a page P on which the cluster is recovered,
This shows the location of the page map management information PC provided corresponding to the page map. Similarly, mapping priority queue MPQ (
Both the first page pointer PHDK' (=1 to 15) and the last page pointer PTLK in FIG.
A forward link pointer NXP and a reverse link pointer PRP also indicate the positions of the page map management information of the previous and previous pages of the matrix.

Therefore, as shown in FIG.
The first page pointer PHD of the mapping priority i in PQ indicates the page map management information PC2 corresponding to the first page Pp of the matrix of the priority i, and the forward link pointer NXP of the page map management information PCI is the priority Indicates the next page map management information PC of the matrix of i, and forward link pointer N of the page map management information PC.
XP is the next page map management information pc of the matrix with priority i
, instruct.

Also, the last page pointer PTL, l of mapping priority i in MPQ, the last page P of the matrix of priority i,
Page map management information PC corresponding to.

The reverse link pointer PRP of the page map management information PCr points to the previous page map management information PC in the matrix with priority 1, and the reverse link pointer PRP of the page map management information PC similarly priority i
Indicates the page map management information PC immediately before the matrix in .

The mapping priority (1 to 15) of each page P is designated by the application program for each cluster, and is used to determine from which page data is to be ejected in cluster units when free space is exhausted. The page P1 in use has a forward link pointer NXP and a reverse link pointer PRP in the page map management information PC1.
is always linked to some mapping priority matrix. Note that within the same matrix, the first page is evicted, and when there are no more free pages, the first page indicated by the first page pointer P HDK of the matrix with the lowest priority is deleted from that matrix, and the cluster of the first page is deleted. to the virtual memory area VNA.

Thereafter, the clusters are removed by sequentially following the links, and when the end of the link is detected, the process moves to the next lowest priority matrix and the clusters on the first page are removed in the same way.

On the other hand, if the application specifies the mapping priority of clusters in a predetermined page, the page is linked to the end of the matrix of the specified mapping priority. Also, when a cluster of a predetermined page is accessed, or when a change in mapping priority is instructed, the link of the page is deleted from the corresponding matrix, and the link of the page is deleted from the corresponding matrix (or the same matrix if accessed). Link to the end of. Therefore, in the matrix, previously accessed pages are linked to the front of the matrix, and recently accessed pages are linked to the back of the matrix.

Next, an outline of the operation during expulsion and recovery will be explained.

(5) Internal memo in fixed length cluster units! Data is input/output between JINM and external memory ETM, and each page P of page area PGA (page length = cluster length)
A page map PMAP and a mapping priority queue MPQ are prepared in the management information area VCT in correspondence with the above. Then, the page map management information P, C of each page P is set to 0 to 1 indicating the difficulty of being evicted to the virtual memory area VMA.
5, and pointers PHD, , NXP, etc. that specify a matrix (FIG. 4) indicating the difficulty of being evicted to the virtual memory area VMA for each mapping priority.

Then, when the page area PGAI and free pages are exhausted, the matrix with the lowest mapping priority m (m=15
) first page (if there is no first page, then the next lowest matrix (
The first page of m=14) (the same applies hereafter) is deleted from the matrix, and the data of the page is ejected in units of clusters to a predetermined cluster area of the virtual memory area VMA indicated by the cluster number CLN of the page map management information.

If the predetermined record is not stored in the page area PGA and is evicted to the virtual memory area VMA,
Since information indicating the cluster area C4 in which the record is stored is given, data is recovered from the cluster area C4 to the page P indicated by the cluster map management information CC1 corresponding to the cluster area C4. In addition, at this time, page Pl
By generating page map management information P1C according to , page P is linked to the end of the mapping priority matrix specified by the application manager.

<Effects of the Invention> According to the present invention, data is input/output between internal memory and external memory in fixed-length cluster units, and
In the information that manages each cluster on the internal memory, there is a mapping priority that indicates how difficult the cluster is to be evicted to the external memory, and a matrix that indicates how difficult the cluster is to be evicted to the external memory for each mapping priority. In addition, when there are no free clusters in the internal memory, the clusters in the mapping matrix with the lowest mapping priority are flushed out to the external memory in order, starting from the first cluster, so it is possible to specify a separate priority for each file. The order in which files are evicted can be controlled without depending on the definition order, and clusters that have been accessed (used) are relocated to the end of the matrix, so clusters with high usability can be kept in internal memory as much as possible. Can be done.

Furthermore, since data is flushed out in cluster units and rearranged at the end of the matrix each time it is accessed, it is possible to retain frequently used data in the internal memory as much as possible. can.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram of the internal data management method according to the present invention, Fig. 2 is an explanatory diagram of a mapping priority queue, Fig. 3 is an explanatory diagram of page map management information, and Fig. 4 is a matrix of mapping priority l. It is an explanatory diagram. ETM...External memory, INM...Internal memory, PGA...Page area, P...Page, VCT...Management information area, VMA...Virtual memory area, C...Cluster area, MP
Q... Mapping priority queue, PMAP... Page map, PC... Page map management information, CMAP... Cluster map patent applicant Fanuc Corporation Figure 2 Figure 3

Claims (4)

[Claims] (1) In an internal data management method in a system that has a randomly accessible external memory and uses the external memory as an extension of the internal memory, data is input between the internal memory and the external memory in fixed-length cluster units. In addition to outputting data, the information that manages each cluster on the internal memory includes a mapping priority that represents the difficulty of each cluster being evicted to external memory, and a mapping priority that represents the difficulty of evicting each cluster to external memory for each mapping priority. An internal data management method characterized by having a matrix indicating the number of clusters, and when there are no free clusters in the internal memory, the clusters of the matrix with a low mapping priority are flushed out to the external memory in order from the top cluster. (2) When a cluster on the internal memory is accessed and when a cluster on the external memory is read into the internal memory, the cluster is linked to the end of a predetermined matrix. Internal data management methods described. (3) Management information is prepared corresponding to each cluster on the internal memory, and each management information includes the mapping priority of the corresponding cluster and link data for forming a matrix of the priority. An internal data management method according to claim 2. (4) Internal data according to claim 3, characterized in that the management information includes data indicating where in the external memory the cluster is to be ejected, and the data is referred to to eject the cluster to the external memory. Management method.