Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
1998-09-10
2001-10-23
Alam, Hosain T. (Department: 2172)
Data processing: database and file management or data structures
Database design
Data structure types
C711S165000
Reexamination Certificate
active
06308183
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a file management system using a storage device and a method of managing the file system.
BACKGROUND OF THE INVENTION
The prior art file management system is outlined in a block diagram in FIG.
11
.
The prior art file system as illustrated in
FIG. 11
is so configured that the storage area in the storage device
900
is divided into data blocks of a certain size, 8 kilobytes, for example, each data block given a block number and its occupation status managed by an unoccupied block management means
40
. The blocks already allocated for a specific file are registered on a block table which is stored in the storage device
900
, and are used for writing or reading out data as described below.
FIG. 12
is a conceptional diagram showing the configuration of the block table Tb in the aforesaid prior art file management system. On the block table Tb, the block number Bn (B
1
, B
2
, B
3
, B
4
, . . . Bk) is registered in order of the corresponding file offset
0
n.
It should be understood that for purpose of this specification, file offset is a measure of quantity of data or data size from the head of the file The suffix On corresponds to the code number given to each data fragment when data is fragmented by the specific size (one block size, for example), and can indicate the position or address in a file. It also indicates the data size at a specific offset address. File offset
04
, for example, means four times the data size of offset
01
.
When a new file is to be created, a file creating means
10
directs a data block allocation means
30
to acquire block numbers for the new file, indicating the file offset On, and gives data to write to a data writing means
20
at the same time.
The data block allocation means
30
first reads out the block table Tb stored on the storage device
900
and checks to see whether the area (data block) corresponding to the indicated file offset On is already allocated for the file to write or not.
In case a data block is already allocated in the area corresponding to the indicatecd file offset On, that is, a block number Bn is registered, then the said block number bn is taken out. If no data block is allocated yet, the unoccupied block management means
40
is ordered to search for a new data block.
The said unoccupied block management means
40
, which manages the occupation status of the respective data blocks and the current number of unoccupied blocks in the whole file management system as mentioned above, notifies the data block allocation means
30
of the block number Bn read out by the an unoccupied block management means
40
.
The data block allocation means
30
registers on the block table Tb the received block number Bn in such a way that the block number is paired with the file offset On acquired from the file creating means
10
and stores this block table Tb on the storage device
900
. The data block allocation means
30
informs the data writing means
20
of the acquired block number Bn at the same time. The data writing means
20
then writes data into the space at the received block number Bn on the storage device
900
.
When data in a specific file is to be read out at the request from outside in this system (in practice, the request is made by specifying or inputting subheading from outside when the requirement of a subheading from the top is provided, and the file read-out means acquires an offset On on the basis of the inputted data), the file read-out means
50
notifies a block take-out means
60
of the file offset On. The block take-out means
60
in turn reads out the block table Tb stored on the storage device
900
and acquires from the block table Tb the block number Bn corresponding to the given file offset On, and informs the data read-out means
70
of the block number.
Receiving the block number Bn, the data read-out means
70
reads out the object data from the space at that block number Bn On the storage device
900
and forwards the same to the file read-out means
50
.
If the point specified as starting file offset address for writing a new data is one or more data block size down from the end point of last file offset, there occurs a block region Az without real data or dummy blocks on the block table Tb as shown in FIG.
3
.
In the usual file management, even if continuous data are written in data blocks which are apart from each other, each block is arranged consecutively on the block table Tb. As shown in
FIG. 13
, for example, data on
0
s
0
~
0
s
9
were written in block B
10
to block B
19
. When another continuous data on offset
0
s
10
downward are going to be written in blocks on B
20
downward, it is found that somehow the blocks from B
20
to B
29
had already been occupied by other data, so that, the new data are written in vacant blocks from B
60
downward instead. Even in such a case, the block table Tb stores both data up to offset
0
s
9
and data from offset
0
s
10
downward as one consecutive or continuous data as shown in FIG.
13
(b).
It is to be understood that the block region (dummy blocks) without real data which is described here is so purposely created that other data may be written later as shown in FIG.
3
.
There are two methods to protect the block region without real data, one in which data blocks are allocated in that block region without fail and the other in which no data blocks are allocated.
In the prior art file management system in which data blocks are allocated without fail in the block region without real data, arrangements are made this way: dummy data,
0
(or
1
)-cleared data, for example, is written in the allocated data blocks without real data so that no indetified data may come out from the block region without real data in the read-out step.
In the prior art file management system which allocates no data blocks in the block region without real data, the block take-out means
60
checks the corresponding space in the block table Tb to find, on reading out from the block region without real data, that no data block is allocated yet, and informs the data read-out means
70
that no data block is present. Then, the data read-out means
70
prepares and forwards dummy data to the file read-out means
50
. And when real data is written later in the block region without real data, a step to allocate data blocks is executed anew.
In the prior art file management system which allocates data block without fail in the block region without real data, there will take place simultaneously allocation of blocks and writing dummy data in the block region without real data if a file offset address not consecutive but apart from the end point of last file is specified for writing a new data.
As it often happens that real data is written later in the block region without real data, it is a waste of time to write dummy data.
In the prior art file management system which allocates no data blocks in the block region without real data, data blocks should be allocated in the block region without real data when real data is written later in that block region.
Since, however, no data blocks are allocated in the block region without real data, a data in some other file initially written occupies the block region. There is no assurance that the unoccupied blocks are left to be allocated for the intended writing of data. Thus it may fail to write real data in the unoccupied block.
The present invention is to resolve such problems with the prior art file management system as described above. And it is an object of the present invention to provide a file management system and a method of managing the same in which in writing data, no dummy data is written in the block region without real data but it will be ensured that in case real data is written later in the block region without real data, data blocks are available which can be allocated in the block region without real data.
SUMMARY OF THE INVENTION
To achieve the foregoing object, the present invention adopts the means described below. That is to say, the file system th
Kawakami Yasunori
Taniyama Hiroshi
Alam Hosain T.
Matsushita Electric - Industrial Co., Ltd.
McDermott & Will & Emery
LandOfFree
File management system and file management method capable of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with File management system and file management method capable of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and File management system and file management method capable of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2572067