CN106201719A - The method and apparatus of management distributed task scheduling RapidIO shared drive - Google Patents

The method and apparatus of management distributed task scheduling RapidIO shared drive Download PDF

Info

Publication number
CN106201719A
CN106201719A CN201610529101.7A CN201610529101A CN106201719A CN 106201719 A CN106201719 A CN 106201719A CN 201610529101 A CN201610529101 A CN 201610529101A CN 106201719 A CN106201719 A CN 106201719A
Authority
CN
China
Prior art keywords
shared drive
rapidio
task scheduling
distributed task
subregion
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201610529101.7A
Other languages
Chinese (zh)
Inventor
杨刚
周兴社
高荣
吴晓
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northwestern Polytechnical University
Original Assignee
Northwestern Polytechnical University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Northwestern Polytechnical University filed Critical Northwestern Polytechnical University
Priority to CN201610529101.7A priority Critical patent/CN106201719A/en
Publication of CN106201719A publication Critical patent/CN106201719A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/54—Interprogram communication
    • G06F9/544—Buffers; Shared memory; Pipes

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention belongs to computer realm, be specifically related to a kind of method and apparatus managing distributed task scheduling RapidIO shared drive, a kind of method managing distributed task scheduling RapidIO shared drive comprises the following steps: construct shared drive subregion in interconnection datatron；By the way of shared drive subregion, distributed task scheduling RapidIO shared drive is managed, solve distributed task scheduling RapidIO shared drive creates, open, in closing process, the maintenance issues of internal memory use status information.A kind of equipment realizing said method includes constructing module；Creation module；Operation module.The distributed task scheduling RapidIO shared drive subregion of the inventive method and device build, solve distributed task scheduling on RapidIO interconnection datatron and, for the regulatory requirement of RapidIO shared drive state, provide support for communication based on RapidIO shared drive between distributed task scheduling.

Description

The method and apparatus of management distributed task scheduling RapidIO shared drive
Technical field
The invention belongs to computer realm, be specifically related to a kind of method managing distributed task scheduling RapidIO shared drive And equipment.
Background technology
On RapidIO interconnection datatron, distributed task scheduling can be by RapidIO internal memory mapping mechanism application memory mapping area Between as the shared drive between task, but the realization of this distributed task scheduling shared drive based on RapidIO, it is right to need RapidIO shared drive uses status information to be managed.
In existing technology, do not propose distributed task scheduling on a kind of RapidIO interconnection datatron RapidIO is total to Enjoy the method that internal memory uses condition managing.The method mentioned in patent of the present invention, to distributed by the way of memory partitioning Business RapidIO shared drive is managed, solve distributed task scheduling RapidIO shared drive create, open, in closing process, Internal memory uses the maintenance issues of status information.
Summary of the invention
For the deficiency of the problems referred to above, the present invention proposes a kind of method managing distributed task scheduling RapidIO shared drive And equipment.It is an object of the invention to solve at least one the problems referred to above or defect, and provide at least one will be described later Advantage.
It is a still further object of the present invention to provide a kind of method managing distributed task scheduling RapidIO shared drive, its By the way of structure shared drive subregion, it is achieved RapidIO interconnection multiprocessor RapidIO shared drive is effectively managed Reason, solves to be positioned at distributed task scheduling on processor and RapidIO shared drive is used to the maintenance of state.
It is a still further object of the present invention to provide a kind of method managing distributed task scheduling RapidIO shared drive, its Effectively solve distributed task scheduling on RapidIO interconnection datatron and the mutual of status information is used for shared drive, for The establishment of RapidIO shared drive, open and shutoff operation is laid a good foundation.
It is a still further object of the present invention to provide a kind of equipment managing distributed task scheduling RapidIO shared drive, its The method that have effectively achieved management distributed task scheduling RapidIO shared drive.
In order to realize according to object of the present invention and further advantage, the invention provides a kind of management distributed task scheduling The method of RapidIO shared drive, comprises the following steps:
Shared drive subregion is constructed in interconnection datatron；
Obtain the available information of described shared drive subregion, at least one interconnection datatron creates new distributed Business RapidIO shared drive；
When opening described new distributed task scheduling RapidIO shared drive, read described shared drive subregion, obtain other Memory block collection identical in distributed task scheduling RapidIO shared drive on interconnection datatron, determines and uses unappropriated interior Counterfoil collection；
When closing new distributed task scheduling RapidIO shared drive, revise the state of described shared drive subregion.
The distributed task scheduling RapidIO shared drive subregion of the inventive method structure, solves RapidIO and interconnects datatron On distributed task scheduling for the regulatory requirement of RapidIO shared drive state, for sharing based on RapidIO between distributed task scheduling The communication of internal memory provides support.
Preferably, described shared drive subregion includes that shared drive descriptor uses state table, memory block distribution state Table, query specification symbol and memory block attribute area.
The use state of described shared drive descriptor stores with byte, and the use state of described memory block is deposited with binary system Storage, described query specification symbol is used for sharing of same shared drive descriptor, and described memory block attribute is used for same shared drive Representated by descriptor, shared drive set of blocks shares
Preferably, in interconnection datatron, construct shared drive subregion to include:
Calculating described shared drive descriptor respectively uses state table, described memory block distribution state table, described inquiry to retouch State the memory headroom shared by symbol and described memory block attribute；
Calculate for managing the memory headroom needed for described shared drive subregion, and pre-in interconnection datatron shared drive Reserve the memory headroom needed for described shared drive subregion, described shared drive is described simultaneously and use state table, described internal memory Block distributes state table, described query specification symbol and the described memory block attribute internal memory successively needed for described shared drive subregion Space stores, it is thus achieved that described shared drive subregion.
Described shared drive descriptor uses the calculating of memory headroom needed for state table, described memory block distribution state table Shared by the calculating of the memory headroom shared by the required calculating of memory headroom, described query specification symbol and described memory block attribute The calculating of memory headroom be separate, the memory headroom of four is added that to obtain the space of described shared drive subregion big Little.
Preferably, obtain the available information of described shared drive subregion, at least one interconnection datatron creates new The specifically comprising the following steps that of distributed task scheduling RapidIO shared drive
A, obtain described shared drive descriptor use state table use information, it may be judged whether have available；It is then to perform Step B, no, then perform step C；
B, obtain the use information of described shared drive block distribution state table, it may be judged whether have and meet application and create quantity Free memory block；It is then to create new distributed task scheduling RapidIO shared drive, no, then perform step C；
C, do not create new distributed task scheduling RapidIO shared drive.
The use information of use state table is described, if use information comprises available letter through obtaining described shared drive Breath, then obtain the use information of described shared drive block distribution state table further, see if there is free memory block available information, If having, then create new distributed task scheduling RapidIO shared drive, thus avoid create with being not added with any investigation new Distributed task scheduling, but when opening and using, have been found that the drawback not having free memory block to use, save the time, improve efficiency.
Preferably, when opening described new distributed task scheduling RapidIO shared drive, by reading described shared drive Described query specification symbol in subregion and described memory block attribute, obtain the distributed task scheduling on other interconnection datatron Memory block collection identical in RapidIO shared drive, determines and uses unappropriated memory block collection.
When opening described new distributed task scheduling RapidIO shared drive, by reading in described shared drive subregion Which memory block described query specification symbol and described memory block attribute, check already be allocated on other interconnection datatron and be distributed Formula task RapidIO shared drive uses, and which is not previously allocated, and determines and uses unappropriated memory block collection, passing through information Share, improve work efficiency.
Preferably, when closing new distributed task scheduling RapidIO shared drive, the institute of described shared drive subregion is revised State shared drive descriptor and use state table.
When on RapidIO interconnection datatron, distributed task scheduling performs shutoff operation by shared drive descriptor, to altogether The reference count enjoying descriptor memory symbol performs the operation that subtracts 1, if reference count is kept to 0, then revises this shared drive descriptor and exists The memory block set represented in memory block distribution state table uses state.
A kind of equipment managing distributed task scheduling RapidIO shared drive, described equipment includes:
Constructing module, for constructing the shared drive subregion on interconnection datatron；
Creation module, for obtaining the available information of described shared drive subregion, creates at least one interconnection datatron Build new distributed task scheduling RapidIO shared drive；
Operation module, for opening and closing described new distributed task scheduling RapidIO shared drive.
The described constructing module of present device, the distributed task scheduling shared drive subregion of its structure, solve RapidIO On interconnection datatron, distributed task scheduling is for the regulatory requirement of RapidIO shared drive state, between distributed task scheduling based on The communication of RapidIO shared drive provides support.
Preferably, described constructing module specifically for:
Calculate shared drive descriptor respectively and use state table, memory block distribution state table, query specification symbol and memory block Memory headroom shared by attribute；
Calculate for managing the memory headroom needed for described shared drive subregion, and pre-in interconnection datatron shared drive Reserve the memory headroom needed for described shared drive subregion, described shared drive is described simultaneously and use state table, described internal memory Block distributes state table, described query specification symbol and the described memory block attribute internal memory successively needed for described shared drive subregion Space stores, it is thus achieved that described shared drive subregion.
Preferably, described creation module specifically for:
Obtain described shared drive descriptor and use the use information of state table, it may be judged whether have available；It is then to obtain institute State the use information of shared drive block distribution state table, no, the most do not create new distributed task scheduling RapidIO shared drive；
Obtain the use information of described shared drive block distribution state table, it may be judged whether have the sky meeting application establishment quantity Not busy memory block；It is then to create new distributed task scheduling RapidIO shared drive, no, the most do not create new distributed task scheduling RapidIO shared drive.
Preferably, described operation module specifically for:
When opening described new distributed task scheduling RapidIO shared drive, that reads in described shared drive subregion is described Query specification symbol and described memory block attribute, obtain in the distributed task scheduling RapidIO shared drive on other interconnection datatron Identical memory block collection, determines and uses unappropriated memory block collection；
When closing new distributed task scheduling RapidIO shared drive, revise the described shared interior of described shared drive subregion Deposit descriptor and use state table.
Beneficial effects of the present invention:
The method of the management distributed task scheduling RapidIO shared drive that 1, the present invention provides, it is by structure shared drive The mode of subregion, it is achieved RapidIO interconnection multiprocessor RapidIO shared drive is effectively managed, solves to be positioned at everywhere On reason machine, distributed task scheduling uses the maintenance of state for RapidIO shared drive, improves work efficiency.
The method of the management distributed task scheduling RapidIO shared drive that 2, the present invention provides, it effectively solves On RapidIO interconnection datatron, distributed task scheduling uses the mutual of status information for shared drive, in sharing for RapidIO The establishment deposited, open and shutoff operation is laid a good foundation.
The equipment of the management distributed task scheduling RapidIO shared drive that 3, the present invention provides, it have effectively achieved management The method of distributed task scheduling RapidIO shared drive.
Accompanying drawing explanation
Fig. 1 is the structural representation of shared drive subregion of the present invention；
Fig. 2 is the flow chart of shared drive subregion of the present invention structure；
Fig. 3 is the flow chart of the method for management distributed task scheduling RapidIO shared drive of the present invention；
Fig. 4 is the structural representation of the equipment of management distributed task scheduling RapidIO shared drive of the present invention.
Detailed description of the invention
The present invention will be further described with detailed description of the invention below in conjunction with the accompanying drawings.
The flow chart of the present invention as shown in Figure 3, comprises the following steps:
The invention provides a kind of method managing distributed task scheduling RapidIO shared drive, comprise the following steps:
Shared drive subregion is constructed in interconnection datatron；
Obtain the available information of described shared drive subregion, at least one interconnection datatron creates new distributed Business RapidIO shared drive；
When opening described new distributed task scheduling RapidIO shared drive, read described shared drive subregion, obtain other Memory block collection identical in distributed task scheduling RapidIO shared drive on interconnection datatron, determines and uses unappropriated interior Counterfoil collection；
When closing new distributed task scheduling RapidIO shared drive, revise the state of described shared drive subregion.
The distributed task scheduling RapidIO shared drive subregion of the inventive method structure, solves RapidIO and interconnects datatron On distributed task scheduling for the regulatory requirement of RapidIO shared drive state, for sharing based on RapidIO between distributed task scheduling The communication of internal memory provides support.
On the basis of above-described embodiment, another embodiment, described shared drive subregion includes shared drive descriptor Use state table, memory block distribution state table, query specification symbol and memory block attribute area.
The use state of described shared drive descriptor stores with byte, and the use state of described memory block is deposited with binary system Storage, described query specification symbol is used for sharing of same shared drive descriptor, and described memory block attribute is used for same shared drive Representated by descriptor, shared drive set of blocks shares
On the basis of above-described embodiment, another embodiment, in interconnection datatron, construct shared drive subregion include:
Calculating described shared drive descriptor respectively uses state table, described memory block distribution state table, described inquiry to retouch State the memory headroom shared by symbol and described memory block attribute；
Calculate for managing the memory headroom needed for described shared drive subregion, and pre-in interconnection datatron shared drive Reserve the memory headroom needed for described shared drive subregion, described shared drive is described simultaneously and use state table, described internal memory Block distributes state table, described query specification symbol and the described memory block attribute internal memory successively needed for described shared drive subregion Space stores, it is thus achieved that described shared drive subregion.
Described shared drive descriptor uses the calculating of memory headroom needed for state table, described memory block distribution state table Shared by the calculating of the memory headroom shared by the required calculating of memory headroom, described query specification symbol and described memory block attribute The calculating of memory headroom be separate, the memory headroom of four is added that to obtain the space of described shared drive subregion big Little.
Being S byte for space size, storage allocation block size is block_size byte, it is provided that N number of shared drive is retouched State the distributed task scheduling RapidIO shared drive of symbol (shared drive descriptor is for one memory block collection of mark), in it is shared The building method depositing management subregion is as follows:
Store the use state of each shared drive descriptor with byte, 0 be the free time, value more than 0 be distributed The number of times that shared drive is quoted by task, and by formula 1 calculate record shared drive descriptor use state needed in Deposit space size_1 byte, the number of times quoted of task, referred to how many tasks carried out shared drive quoting (or It is referred to as using), the memory headroom that formula 1 calculates is only intended to store shared drive and " quotes " state value counted.Formula 1 is such as Following formula:
N*8
With the use state of a binary bit stored memory block, 0 be idle, 1 for, and calculated note by formula 2 Memory headroom size_2 byte needed for record memory block distribution state.Formula 2 such as following formula:
[S/(8*block_size)]
Memory headroom size_3 byte as shared by formula 3 calculates query specification symbol；Query specification symbol is used for Between RapidIO interconnection datatron distributed task scheduling, same shared drive descriptor is shared.Formula 3 such as following formula:
[[log2 N]/8]
Memory headroom size_4 byte as shared by formula 4 calculates memory block attribute；Memory block attribute is used for For shared drive set of blocks representated by same shared drive descriptor between RapidIO interconnecting heterogeneous datatron distributed task scheduling Share.Formula 4 such as following formula:
[[(log2[S/block_size])]/8]*[S/block_size]
Calculate for managing the memory headroom size_5 byte needed for RapidIO shared drive by formula 5, and throughout In reason machine RapidIO shared drive, the space of reserved size_5 byte, uses state table, internal memory by shared drive descriptor simultaneously Block distribution state table, query specification symbol and memory block attribute area store successively in this size_5 byte, and formula 5 is as follows Formula:
size_1+size_2+size_3+size_4
The calculating of size_1, size_2, size_3 and size_4 tetra-value is separate, but four additions draw jointly size_5.Used here as size_1, size_2, size_3, size_4 and size_5 have referred to result of calculation, simplify step The expression of middle formula.
On the basis of above-described embodiment, another embodiment, obtain the available information of described shared drive subregion, extremely Specifically comprising the following steps that of new distributed task scheduling RapidIO shared drive is created on a few interconnection datatron
B, obtain described shared drive descriptor use state table use information, it may be judged whether have available；It is then to perform Step B, no, then perform step C；
B, obtain the use information of described shared drive block distribution state table, it may be judged whether have and meet application and create quantity Free memory block；It is then to create new distributed task scheduling RapidIO shared drive, no, then perform step C；
C, do not create new distributed task scheduling RapidIO shared drive.
The use information of use state table is described, if use information comprises available letter through obtaining described shared drive Breath, i.e. value is the shared drive descriptor of 0, then obtain the use information of described shared drive block distribution state table further, See if there is free memory block available information (state value be 0 expression memory block idle), when have value be 0 (idle) shared in When depositing descriptor and sufficient amount of free memory block, i.e. create new distributed task scheduling RapidIO shared drive, thus Avoid create new distributed task scheduling with being not added with any investigation, but have been found that when opening and using and do not have free memory block to use Drawback, has saved the time, improves efficiency.
On the basis of above-described embodiment, further embodiment, open described new distributed task scheduling RapidIO and share interior When depositing, by reading the described query specification symbol in described shared drive subregion and described memory block attribute, obtain other interconnection Memory block collection identical in distributed task scheduling RapidIO shared drive on datatron, determines and uses unappropriated memory block Collection.
When opening described new distributed task scheduling RapidIO shared drive, by reading in described shared drive subregion Which memory block described query specification symbol and described memory block attribute, check already be allocated on other interconnection datatron and be distributed Formula task RapidIO shared drive uses, and which is not previously allocated, and determines and uses unappropriated memory block collection, passing through information Share, improve work efficiency.
On the basis of above-described embodiment, further embodiment, close new distributed task scheduling RapidIO shared drive Time, the described shared drive descriptor revising described shared drive subregion uses state table.
When on RapidIO interconnection datatron, distributed task scheduling performs shutoff operation by shared drive descriptor, to altogether The reference count enjoying descriptor memory symbol performs the operation that subtracts 1, if reference count is kept to 0, then revises this shared drive descriptor and exists The memory block set represented in memory block distribution state table uses state.
A kind of equipment managing distributed task scheduling RapidIO shared drive, described equipment includes:
Constructing module, for constructing the shared drive subregion on interconnection datatron；
Creation module, for obtaining the available information of described shared drive subregion, creates at least one interconnection datatron Build new distributed task scheduling RapidIO shared drive；
Operation module, for opening and closing described new distributed task scheduling RapidIO shared drive.
The described constructing module of present device, the distributed task scheduling shared drive subregion of its structure, solve RapidIO On interconnection datatron, distributed task scheduling is for the regulatory requirement of RapidIO shared drive state, between distributed task scheduling based on The communication of RapidIO shared drive provides support.
On the basis of above-mentioned situation, another embodiment, described constructing module specifically for:
Calculate shared drive descriptor respectively and use state table, memory block distribution state table, query specification symbol and memory block Memory headroom shared by attribute；
Calculate for managing the memory headroom needed for described shared drive subregion, and pre-in interconnection datatron shared drive Reserve the memory headroom needed for described shared drive subregion, described shared drive is described simultaneously and use state table, described internal memory Block distributes state table, described query specification symbol and the described memory block attribute internal memory successively needed for described shared drive subregion Space stores, it is thus achieved that described shared drive subregion.
Being S byte for space size, storage allocation block size is block_size byte, it is provided that N number of shared drive is retouched State the distributed task scheduling RapidIO shared drive of symbol (shared drive descriptor is for one memory block collection of mark), in it is shared The building method depositing management subregion is as follows:
Store the use state of each shared drive descriptor with byte, 0 be the free time, value more than 0 be distributed The number of times that shared drive is quoted by task, and by formula 1 calculate record shared drive descriptor use state needed in Deposit space size_1 byte.The number of times quoted of task, referred to how many tasks carried out shared drive quoting (or It is referred to as using), the memory headroom that formula 1 calculates is only intended to store shared drive and " quotes " state value counted.Formula 1 is such as Under:
N*8
With the use state of a binary bit stored memory block, 0 be idle, 1 for, and calculated note by formula 2 Memory headroom size_2 byte needed for record memory block distribution state.Formula 2 is as follows:
[S/(8*block_size)]
Memory headroom size_3 byte as shared by formula 3 calculates query specification symbol；Query specification symbol is used for Between RapidIO interconnection datatron distributed task scheduling, same shared drive descriptor is shared.Public
[[log2 N]/8]
Formula 3 is as follows:
Memory headroom size_4 byte as shared by formula 4 calculates memory block attribute；Memory block attribute is used for For shared drive set of blocks representated by same shared drive descriptor between RapidIO interconnecting heterogeneous datatron distributed task scheduling Share.Formula 4 is as follows:
[[(log2[S/block_size])]/8]*[S/block_size]
Calculate for managing the memory headroom size_5 byte needed for RapidIO shared drive by formula 5, and throughout In reason machine RapidIO shared drive, the space of reserved size_5 byte, uses state table, internal memory by shared drive descriptor simultaneously Block distribution state table, query specification symbol and memory block attribute area store successively in this size_5 byte, and formula 5 is such as Under:
size_1+size_2+size_3+size_4
The calculating of size_1, size_2, size_3 and size_4 tetra-value is separate, but four additions draw jointly size_5.Used here as size_1, size_2, size_3, size_4 and size_5 have referred to result of calculation, simplify step The expression of middle formula.
On the basis of above-described embodiment, another embodiment, described creation module specifically for:
Obtain described shared drive descriptor and use the use information of state table, it may be judged whether have available, i.e. value is 0 Shared drive descriptor；It is then to obtain the use information of described shared drive block distribution state table, no, the most do not create new dividing Cloth task RapidIO shared drive；
Obtain the use information of described shared drive block distribution state table, it may be judged whether have the sky meeting application establishment quantity Not busy memory block (state value is that 0 expression memory block is idle).When the shared drive descriptor having value to be 0 (idle) and q.s Free memory block time；It is then to create new distributed task scheduling RapidIO shared drive, no, the most do not create new distributed Business RapidIO shared drive.
On the basis of above-described embodiment, another embodiment, described operation module specifically for:
When opening described new distributed task scheduling RapidIO shared drive, that reads in described shared drive subregion is described Query specification symbol and described memory block attribute, obtain in the distributed task scheduling RapidIO shared drive on other interconnection datatron Identical memory block collection, determines and uses unappropriated memory block collection.
When closing new distributed task scheduling RapidIO shared drive, revise the described shared interior of described shared drive subregion Deposit descriptor and use state table.On RapidIO interconnection datatron, distributed task scheduling performs closedown by shared drive descriptor During operation, the reference count of shared drive descriptor is performed to subtract 1 operation, if reference count is kept to 0, then revise in this shares Deposit the memory block set use state that descriptor represents in memory block distribution state table.
The RapidIO shared drive subregion of structure, can complete on RapidIO interconnection datatron distributed task scheduling for altogether Enjoy internal memory establishment, open, the operation such as closedown.
The when of establishment, it is S's that task A1 on RapidIO bus interconnection datatron A uses descriptor D to create size RapidIO shared drive, step is:
Shared drive retouch symbol use state table is searched descriptor D the most by other task in creating and sharing Deposit, i.e. the value of descriptor is more than 0；Create unsuccessfully more than 0, end operation；Otherwise the value of descriptor D is assigned to 1:
Then scanning Memory Allocation state table is performed, according to the size of application space S, it may be judged whether there are enough memory blocks It is allocated, joins if memory headroom fills, then return the corresponding memory block number obtained collection, and in memory block allocation table state table Registration, will binary location 1 corresponding to memory block；Otherwise distribute failure, end operation.
The when of opening, task B1 on RapidIO bus interconnection datatron B or the RapidIO bus interconnection of establishment Task A1 request on datatron A opens the RapidIO shared drive representated by descriptor D, through following steps:
1, shared drive retouch symbol use state table is searched descriptor D the most by other task in creating and sharing Deposit, i.e. the value of descriptor is more than 0；Can open more than 0, and the value of D is added 1, then perform step 2)；Otherwise opening operation without Effect, end operation.
2, the value that query specification accords with is assigned to D, query specification is accorded with the shared drive block collection representated by D and is stored in memory block Attribute area, then performs step 3).
3, task B1 and task A1 read memory block attribute area, can obtain the memory block collection corresponding to descriptor D.
The when of closedown, task C1 or B1 or A1 request on the datatron C of RapidIO bus interconnection are closed and are described Symbol RapidIO shared drive representated by D, through following steps, (premise performing shutoff operation is that descriptor D is used by task In creating RapidIO shared drive):
Retouch symbol at shared drive and use the value searching descriptor D in the state table operation that carries out subtracting 1, if after the value of D subtracts 1 Become 0, the memory block collection having used descriptor D to refer to without other task is described, then belonged to by the memory block representated by descriptor D Property collection value in memory block distribution state table be assigned to 0, release shared drive block collection representated by descriptor D；Otherwise close behaviour Work terminates.
Invention also has the data of other embodiments at this, does not enumerates.
Although embodiment of the present invention are disclosed as above, but it is not restricted in description and embodiment listed Using, it can be applied to various applicable the field of the invention completely, for those skilled in the art, and can be easily Realizing other amendment, therefore under the general concept limited without departing substantially from claim and equivalency range, the present invention does not limit In specific details with shown here as the legend with description.

Claims (10)

1. the method managing distributed task scheduling RapidIO shared drive, it is characterised in that comprise the following steps:
Shared drive subregion is constructed in interconnection datatron；
Obtain the available information of described shared drive subregion, at least one interconnection datatron creates new distributed task scheduling RapidIO shared drive；
When opening described new distributed task scheduling RapidIO shared drive, read described shared drive subregion, obtain other interconnection Memory block collection identical in distributed task scheduling RapidIO shared drive on datatron, determines and uses unappropriated memory block Collection；
When closing new distributed task scheduling RapidIO shared drive, revise the state of described shared drive subregion.
2. the method managing distributed task scheduling RapidIO shared drive as claimed in claim 1, it is characterised in that described common Enjoy memory partitioning and include that shared drive descriptor uses state table, memory block distribution state table, query specification symbol and memory block to belong to Property district.
3. the method managing distributed task scheduling RapidIO shared drive as claimed in claim 2, it is characterised in that in interconnection Datatron constructs shared drive subregion include:
Calculate described shared drive descriptor respectively and use state table, described memory block distribution state table, described query specification symbol And the memory headroom shared by described memory block attribute；
Calculate for managing the memory headroom needed for described shared drive subregion, and reserve in interconnection datatron shared drive Memory headroom needed for described shared drive subregion, describes described shared drive simultaneously and uses state table, described memory block to divide Join state table, described query specification symbol and the described memory block attribute memory headroom successively needed for described shared drive subregion Store, it is thus achieved that described shared drive subregion.
4. the method managing distributed task scheduling RapidIO shared drive as claimed in claim 2, it is characterised in that obtain institute State the available information of shared drive subregion, at least one interconnection datatron creates new distributed task scheduling RapidIO and shares Specifically comprising the following steps that of internal memory
A, obtain described shared drive descriptor use state table use information, it may be judged whether have available；It is then to perform step B, no, then perform step C；
B, obtain described shared drive block distribution state table use information, it may be judged whether have meet application create quantity free time Memory block；It is then to create new distributed task scheduling RapidIO shared drive, no, then perform step C；
C, do not create new distributed task scheduling RapidIO shared drive.
5. the method managing distributed task scheduling RapidIO shared drive as claimed in claim 2, it is characterised in that open institute When stating new distributed task scheduling RapidIO shared drive, accord with by reading the described query specification in described shared drive subregion With described memory block attribute, obtain internal memory identical in the distributed task scheduling RapidIO shared drive on other interconnection datatron Block collection, determines and uses unappropriated memory block collection.
6. the method managing distributed task scheduling RapidIO shared drive as claimed in claim 2, it is characterised in that close new Distributed task scheduling RapidIO shared drive time, revise described shared drive subregion described shared drive descriptor use shape State table.
7. the equipment managing distributed task scheduling RapidIO shared drive, it is characterised in that described equipment includes:
Constructing module, for constructing the shared drive subregion on interconnection datatron；
Creation module, for obtaining the available information of described shared drive subregion, creates new at least one interconnection datatron Distributed task scheduling RapidIO shared drive；
Operation module, for opening and closing described new distributed task scheduling RapidIO shared drive.
8. the equipment managing distributed task scheduling RapidIO shared drive as claimed in claim 7, it is characterised in that described structure Modeling block specifically for:
Calculate shared drive descriptor respectively and use state table, memory block distribution state table, query specification symbol and memory block attribute Shared memory headroom；
Calculate for managing the memory headroom needed for described shared drive subregion, and reserve in interconnection datatron shared drive Memory headroom needed for described shared drive subregion, describes described shared drive simultaneously and uses state table, described memory block to divide Join state table, described query specification symbol and the described memory block attribute memory headroom successively needed for described shared drive subregion Store, it is thus achieved that described shared drive subregion.
9. the equipment managing distributed task scheduling RapidIO shared drive as claimed in claim 7, it is characterised in that described wound Modeling block specifically for:
Obtain described shared drive descriptor and use the use information of state table, it may be judged whether have available；It is then to obtain described common Enjoy the use information of memory block distribution state table, no, the most do not create new distributed task scheduling RapidIO shared drive；
Obtain the use information of described shared drive block distribution state table, it may be judged whether have and meet in the free time that application creates quantity Counterfoil；It is then to create new distributed task scheduling RapidIO shared drive, no, the most do not create new distributed task scheduling RapidIO Shared drive.
10. the equipment managing distributed task scheduling RapidIO shared drive as claimed in claim 7, it is characterised in that described behaviour Make module specifically for:
When opening described new distributed task scheduling RapidIO shared drive, described in reading in described shared drive subregion Query specification symbol and described memory block attribute, obtain in the distributed task scheduling RapidIO shared drive on other interconnection datatron Identical memory block collection, determines and uses unappropriated memory block collection；
When closing new distributed task scheduling RapidIO shared drive, the described shared drive revising described shared drive subregion is retouched State symbol and use state table.
CN201610529101.7A 2016-07-05 2016-07-05 The method and apparatus of management distributed task scheduling RapidIO shared drive Pending CN106201719A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610529101.7A CN106201719A (en) 2016-07-05 2016-07-05 The method and apparatus of management distributed task scheduling RapidIO shared drive

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610529101.7A CN106201719A (en) 2016-07-05 2016-07-05 The method and apparatus of management distributed task scheduling RapidIO shared drive

Publications (1)

Publication Number Publication Date
CN106201719A true CN106201719A (en) 2016-12-07

Family

ID=57465659

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610529101.7A Pending CN106201719A (en) 2016-07-05 2016-07-05 The method and apparatus of management distributed task scheduling RapidIO shared drive

Country Status (1)

Country Link
CN (1) CN106201719A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106681842A (en) * 2017-01-18 2017-05-17 迈普通信技术股份有限公司 Management method and device for sharing memory in multi-process system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103376772A (en) * 2012-04-27 2013-10-30 沈阳高精数控技术有限公司 Method for monitoring performance of embedded type numerical control system
CN103631936A (en) * 2013-12-06 2014-03-12 北京首钢自动化信息技术有限公司 Method for managing shared memory
CN105045666A (en) * 2015-07-09 2015-11-11 西北工业大学 Method for partition scheduling in Windows operating system environment

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103376772A (en) * 2012-04-27 2013-10-30 沈阳高精数控技术有限公司 Method for monitoring performance of embedded type numerical control system
CN103631936A (en) * 2013-12-06 2014-03-12 北京首钢自动化信息技术有限公司 Method for managing shared memory
CN105045666A (en) * 2015-07-09 2015-11-11 西北工业大学 Method for partition scheduling in Windows operating system environment

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
黄亮等: ""基于RapidIO和存储映射的高速互连网络"", 《计算机工程》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106681842A (en) * 2017-01-18 2017-05-17 迈普通信技术股份有限公司 Management method and device for sharing memory in multi-process system

Similar Documents

Publication Publication Date Title
CN104503826B (en) The virtual machine mapping method and device of a kind of cloud computation data center
CN103891209B (en) Controllers for network control system
CN103425538B (en) Process communication method and system
CN105183561B (en) A kind of resource allocation methods and system
CN104077239B (en) IP hard disk, and memory system and data operation method thereof
CN104731799B (en) Main memory DBM device
CN107733855A (en) A kind of block catenary system and application process that can support publicly-owned chain, alliance's chain and privately owned chain simultaneously
CN102780574B (en) The collocation method of the office data of service-oriented, device and check method, device
CN100594497C (en) System for implementing network search caching and search method
CN104601680B (en) A kind of method for managing resource and device
CN106060176A (en) Cloud computing application architecture and cloud computing service method based on mixed cloud
CN105991478B (en) Server resource distribution method and its system
CN110489407A (en) Data filling mining method, apparatus, computer equipment and storage medium
CN106385329A (en) Processing method and device of resource pool and equipment
CN106557427A (en) The EMS memory management process and device of shared drive data base
CN103559032A (en) Device and method for managing objects of embedded system
CN107423403A (en) A kind of fusion lock management method and system based on distributed file system
CN103001793A (en) Method and device for managing ACL (access control list)
CN107465687A (en) A kind of implementation method of authority configuration, device and terminal
CN107329801A (en) A kind of node administration method and device, many component servers
CN106201719A (en) The method and apparatus of management distributed task scheduling RapidIO shared drive
CN109542863A (en) Mine machine system is dug in artificial intelligence storage
CN104239520B (en) A kind of HDFS data block Placement Strategies based on historical information
CN103514412B (en) Build the method and Cloud Server of access control based roles system
CN112463881A (en) Efficient block chain system and processing method for cloud edge aggregation computing environment

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20161207