EP3732560A1 - Procede de stockage de donnees et procede d'execution d'application avec reduction du temps d'acces aux donnees stockees - Google Patents
Procede de stockage de donnees et procede d'execution d'application avec reduction du temps d'acces aux donnees stockeesInfo
- Publication number
- EP3732560A1 EP3732560A1 EP18842433.7A EP18842433A EP3732560A1 EP 3732560 A1 EP3732560 A1 EP 3732560A1 EP 18842433 A EP18842433 A EP 18842433A EP 3732560 A1 EP3732560 A1 EP 3732560A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- data
- processes
- servers
- slices
- application
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 170
- 238000013500 data storage Methods 0.000 title claims description 15
- 230000008569 process Effects 0.000 claims abstract description 117
- 238000003860 storage Methods 0.000 claims abstract description 84
- 238000009826 distribution Methods 0.000 claims abstract description 33
- 238000004364 calculation method Methods 0.000 claims description 23
- 230000003252 repetitive effect Effects 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 3
- 230000006399 behavior Effects 0.000 description 17
- 238000005457 optimization Methods 0.000 description 7
- 230000009467 reduction Effects 0.000 description 6
- 230000015572 biosynthetic process Effects 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 238000003786 synthesis reaction Methods 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 3
- 238000011144 upstream manufacturing Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 241001522296 Erithacus rubecula Species 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000001684 chronic effect Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000002932 luster Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0604—Improving or facilitating administration, e.g. storage management
- G06F3/0605—Improving or facilitating administration, e.g. storage management by facilitating the interaction with a user or administrator
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/17—Details of further file system functions
- G06F16/172—Caching, prefetching or hoarding of files
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
- G06F16/1824—Distributed file systems implemented using Network-attached Storage [NAS] architecture
- G06F16/183—Provision of network file services by network file servers, e.g. by using NFS, CIFS
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0629—Configuration or reconfiguration of storage systems
- G06F3/0631—Configuration or reconfiguration of storage systems by allocating resources to storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/064—Management of blocks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
Definitions
- the invention relates to the field of data storage methods with reduced access time to stored data, as well as the field of corresponding application execution methods using these data storage methods.
- a first disadvantage of this first prior art is that this more complex and more expensive technology, especially for several applications each managing a large amount of data, makes the overall system complex and expensive.
- a second disadvantage of this first prior art is that, according to the invention, even after each read operation and each write operation has been made very fast, if there is a large number to perform, and this is particularly the case for several applications each managing a large number of data, the overall access time to stored files remains important, or very important, the scale of the overall time of running applications. This time of access to the stored files is particularly noticeable at the level of the progress of the applications, during the these applications are carried out during a very large process of data processing, such as a large calculation, because then the many periodic phases of data backup ("checkpoint" in English) inherently occupy a proportion important overall time of progress of this process of treatment, for example of the order of 10 to 20 minutes every hour.
- the optimization of the unwinding time is mainly centered on the reduction of the calculation time of the applications.
- this second prior art often only deals with reducing the calculation time of the application because it is considered better controlled and more important.
- decreasing the input / output time that is to say reading and / or writing each partial result backup, will be able to pay first because this backup time can take from 10% to 30% of the total execution time of an application, and then can be achieved regardless of the internal logic of the application.
- the management of the input / output time could be made upstream of the execution of the application, that is to say at the level of a optimization of the distribution of data which would then be carried out by an upstream design phase which of course would then save the time of the observation phase.
- this would be achieved at the cost of two major disadvantages which would be on the one hand the total dependence of the application with respect to a single type of storage system and on the other hand the increase in difficulty development of the application to incorporate this additional constraint.
- the object of the present invention is to provide a storage method at least partially overcomes the aforementioned drawbacks.
- the invention aims at providing a data storage method and an application execution method, which, rather than being limited to reducing the time of each file access or only to optimize the calculation time applications, and in particular to seek to over-optimize this computing time applications, considers that it is particularly interesting on the one hand to seek to streamline and organize the access time to files stored by applications, and on the other hand to carry out this rationalization and this organization in an efficient way, first by basing itself on the actual progress of the applications and on their effective interaction, by observing them in their actual functioning, and then by proposing principles of rationalization and the stored file access strategy, which will be more independent of both the type of server and the rs storing data and the type of applications taking place, thus offering increased robustness to changes.
- the present invention proposes a storage method that can be adapted to either data file slices, or even groups of data, or slices of data objects, this storage being performed on different data servers, or even on different storage spaces of data servers, these different storage spaces of data servers, or these different data servers, being accessible separately and independently of each other by applications outside their data servers.
- the present invention also proposes a method for executing application processes corresponding to one or other of these storage methods, and more particularly to the method of storing file slices of data on different data servers.
- a method of storing, on data servers, slices of data files resulting from the execution of several processes of one or more applications comprising: a division of the slices data files stored on different data servers, characterized in that: this distribution is performed so that slices of data files that can be accessed later simultaneously by different application processes are stored on different data servers. data, so as to reduce subsequent access, at each of all or part of these data servers, simultaneously by too many application processes, and in that: the determination of the slices of data files that can be accessed simultaneously by different application processes was carried out, during a preliminary phase of implementation n of these application processes, by observing the behavior of these application processes to access, over time, these slices of stored data files.
- a method of storing, on storage spaces of data servers, slices of data files resulting from the execution of several processes of one or more applications comprising: a distribution slices of data files stored on different storage spaces of different data servers, characterized in that: this distribution is carried out so that slices of data files that can be accessed later simultaneously by different processes of application are stored on different storage spaces of different data servers, so as to reduce subsequent access to each of all or part of these storage spaces simultaneously by too many application processes, and in that: Determine which data file slices can be accessed simultaneously by different app processes During a preliminary phase of execution of these application processes, the application of the behavior of these application processes to access, over time, these slices of stored data files was carried out.
- a method for executing several processes of one or more applications comprising: a first phase of observation of the progress of said processes and of their way of accessing data over time; stored during said execution, during which the determination of the slices of data files that can be accessed simultaneously by different processes application is performed, a second phase of setting data storage by said processes on data servers and on their storage spaces, associating the data file slots with storage spaces on the data servers, a third phase distribution of the data file slices stored on different data servers and their storage spaces, this distribution being performed so that the data file slices that can be accessed later simultaneously by different application processes are stored on different data servers, if necessary on different storage spaces of these data servers, so as to reduce subsequent access to each of all or part of these data servers, if any of these storage spaces at the same time by too many application processes.
- a method of storing, on data servers, data resulting from the execution of several processes comprising: a distribution of the data stored on different data servers, characterized in that: distribution is performed so that groups of data that can be accessed later simultaneously by different processes are stored on different data servers, so as to reduce subsequent access to each of all or part of these data servers , simultaneously by too many processes, and in that: the determination of the groups of data that can be accessed simultaneously by different processes has been performed, during a preliminary phase of execution of these processes, by the observation of the behavior of these processes to access, over time, these groups of stored data.
- a method of storing, on data servers, slices of data objects resulting from the execution of several processes of one or more applications comprising: a distribution of the slices of data objects stored on different data servers, characterized in that: this distribution is performed so as to that slices of data objects that can later be accessed simultaneously by different application processes are stored on different data servers, so as to reduce subsequent access to each of all or part of these data servers , at the same time by too many application processes, and in that: the determination of the slices of data objects that can be accessed simultaneously by different application processes has been performed, during a preliminary execution phase of these application processes, by observing the behavior of these application processes to access, over time, these slices of stored data objects.
- the invention comprises one or more of the following features which can be used separately or in partial combination with one another or in total combination with one another, or with one or the other of the aforementioned objects of the invention. .
- said determination of the slices of data files that can be accessed simultaneously by different application processes has been performed, during a preliminary phase of execution of these processes for a single application or application by application in case plurality of applications, by observing the behavior of these processes of a single application at a time to access, over time, these slices of stored data files.
- the application or applications are portable on other types of data storage servers.
- the application is therefore independent of a data storage system, and this is made possible because the optimization of the distribution of the data is performed downstream, by a phase of observation of the behavior of the application, at instead of being carried out by a design phase in upstream which certainly would save the time of the observation phase but at the cost of two major disadvantages which are on the one hand the total dependence of the application with respect to a single type of storage system and on the other hand the increase the difficulty of developing the application incorporating this additional constraint.
- the processes of the application include repetitive calculations.
- observation phase will offer an excellent compromise, namely to be simple and short and allow a high optimization of the distribution of the data storage areas resulting in a significant decrease in the time of entry / exit during the course of the proceedings. execution of the application.
- said repetitive calculations include weather forecast calculations.
- weather forecasting calculations are a particularly critical example of very repetitive and highly complex calculations, that is to say, requiring a lot of resources but allowing a high optimization of the distribution of the data storage areas that can lead to a significant reduction.
- the input / output time during the execution of the execution of the application is to say, requiring a lot of resources but allowing a high optimization of the distribution of the data storage areas that can lead to a significant reduction.
- the application or applications are executed within a network comprising several thousand compute nodes, preferably at least 5000 compute nodes, preferably at least 10000 compute nodes.
- said distribution of the slices of data files stored on different data servers or on different storage spaces of different data servers is performed by a library of functions which: on the one hand intercepts the creation of data files, on the other hand realizes the storage of the slices of these data files on the storage spaces on the data servers that were associated with them during said preliminary phase of execution of the application processes.
- the data file slices are immediately and directly stored in the right places which will then reduce the access time to these slices of data files.
- FIG. 1 schematically represents an example of a storage system to which the storage method according to one embodiment of the invention can be applied.
- FIG. 2 diagrammatically represents an example of a random storage method that can generate the problem solved by the solution proposed by the storage method according to one embodiment of the invention.
- Figure 3 schematically shows an example of storing data of several application files on several servers generating the problem solved by the solution proposed by the storage method according to one embodiment of the invention.
- FIG. 4 diagrammatically represents an example of a first phase of several application processes generating the problem solved by the solution proposed by the storage method according to one embodiment of the invention.
- FIG. 5 schematically represents an example of a second phase of the unfolding of several application processes generating the problem solved by the solution proposed by the storage method according to one embodiment of the invention.
- FIG. 6 schematically represents an example of a first phase of several application processes using the solution proposed by the storage method according to one embodiment of the invention.
- FIG. 7 schematically represents an example of a second phase of several application processes using the solution proposed by the storage method according to one embodiment of the invention.
- FIG. 8 schematically represents an example of synthesis of the observation phase of the storage method according to one embodiment of the invention.
- FIG. 9 schematically represents an example of synthesis of the parameterization phase of the storage method according to one embodiment of the invention.
- FIG. 1 schematically represents an example of a storage system to which the storage method according to one embodiment of the invention can be applied.
- a calculation node 1 performs a calculation and from time to time needs to read and / or write data to the data servers 3 and 4 using a metadata server 2.
- the data server 3 (OSS for "object storage server” in English) comprises several disk spaces 31 to 33 of data (OST for "object storage target” in English).
- the data server 4 comprises several disk spaces 41 to 43 of data.
- the metadata server 2 (MDS for "meta data server” in English) comprises several disk spaces 41 to 43 of metadata (MDT for "meta data target” in English).
- the language used by the compute node to communicate with the servers 2 to 4 is for example the language Luster (free software language or "open source" in English).
- the computation node 1 sends a file opening request 11 to the metadata server 2.
- the metadata server 2 returns a response 12 containing attributes and file identifiers. Among these attributes are a chunk block size of the data file to be stored as well as a list of data servers or even a list of data disk spaces.
- the computation node 1 sends the data server 3 a request 13 for reading data or a request 14 for writing data.
- the data server 3 reads or writes the data on one of the disk spaces 31 to 33 of data.
- the computation node 1 sends the data server 4 a data read request 15 or a data write request 16.
- the data server 4 reads or writes the data on one of the disk spaces 41 to 43 of data.
- the largest calculators also called supercomputers, are today composed of several thousand independent compute nodes, such as compute node 1, collectively running one or more parallel applications, that is, an application. executes on a large number of calculation nodes 1 simultaneously.
- FIG. 2 diagrammatically represents an example of a random storage method that can generate the problem solved by the solution proposed by the storage method according to one embodiment of the invention.
- the compute node 1 has a data file 5 to store.
- the file 5 is divided into eight slices 51 to 58.
- the data storage strategy is a random strategy, that is to say that the file slices are stored randomly on all the data disk spaces that have been assigned to the data file 5, here the data spaces 31 and 32 of the data server 3 and the data spaces 41 and 42 of the data server 4.
- the data server 3 stores the slices 51 and 55 on the data disk space 31 and the slices 52 and 56 on the data disk space 32.
- the data server 4 stores the slices 53 and 57 on the data disk space 41 and the slices 54 and 58 on the disk space 42 of data.
- the file system When creating a file 5, the file system must choose which servers 3 or 4 data and which disk space 31 to 33 and / or 41 to 43 will be used to store its contents.
- the algorithms used today are based on a method of random type ("round robin" in English) in order to distribute the data randomly on all data servers 3 and 4 and promote a seamless filling.
- the principle is that we set in advance a size of elementary slice of data ("stripe" in English) and a number of disk spaces 31 to 43 for each file 5 to create.
- FIG. 3 schematically represents an example of data storage of several application files on several servers generating the problem solved by the solution proposed by the storage method according to one embodiment of the invention.
- the first slice 611 is stored on the server 100
- the second slice 612 is stored on the server 101
- the third slice 613 is stored on the server 102
- the fourth slice 614 is stored on the server 103.
- the first slice 621 is stored on the server 103
- the second slice 622 is stored on the server 100
- the third slice 623 is stored on the server 101
- the fourth slice 624 is stored on the server 102.
- the first slice 631 is stored on the server
- the second slice 632 is stored on the server 103
- the third slice 633 is stored on the server 100
- the fourth slice 634 is stored on the server 101.
- the first slice 641 is stored on the server 101
- the second slice 642 is stored on the server 102
- the third slice 643 is stored on the server 103
- the fourth slice 644 is stored on the server 100.
- FIG. 4 diagrammatically represents an example of a first phase of several application processes generating the problem solved by the solution proposed by the storage method according to one embodiment of the invention.
- Four processes 65 to 68 of applications will have to access over time, whether for a reading of data or for writing data, to their respective files stored on the servers 100 to 103 as detailed in Figure 3.
- the data represents a local version of a four-column matrix data model. If the size of a column is of the same order of magnitude as that of a slice of file, or even just a multiple of a slice of file, we find a correspondence between the columns of the matrix and the storage slices files 61 to 64.
- FIG. 5 schematically represents an example of a second phase of the unfolding of several application processes generating the problem solved by the solution proposed by the storage method according to one embodiment of the invention.
- the four processes 65 to 68 of applications will have to access over time, whether for a reading of data or for a writing of data, to their respective files stored on the servers 100 to 103 as detailed in FIG. Chance here doing things wrong, at the same time, the process 65 wants to access the slot 611 stored on the server 100, the process 66 wants to access 76 to the slice 622 still stored on the server 100, the process 67 wants to make an access 77 to the slot 633 still stored on the server 100, the process 68 wants to make an access 78 to the slot 644 still stored on the server 100.
- the server 100 to which the four processes 65 to 68 want to make four simultaneous access 75 to 78, will only be able to respond to these requests with a speed four times less than its nominal speed, which will multiply by four the overall time of completion of the four accesses 75 to 78.
- a speed four times less than its nominal speed which will multiply by four the overall time of completion of the four accesses 75 to 78.
- server on which the data were stored with a random strategy For reasons of dependence between the data, at the end of the calculation, each of the processes 65 to 68 begins to write in its result file the data of the column corresponding to its rank in the parallel computing application. The result is that all the processes "attack" at the same time the same data server, here the server 100.
- FIG. 6 schematically represents an example of a first phase of several application processes using the solution proposed by the storage method according to one embodiment of the invention.
- the table of FIG. 6 represents the access requirements for the four files 61 to 64 arranged in columns during five periods of time. These access requirements are identified during the observation phase of the execution of the applications and the corresponding storage of their data. During the first, fourth and fifth periods of time, no file needs to be accessed. On the other hand, during the second period of time, as well as during the third period of time, the first slice 611 of the file 61 must be accessed simultaneously, the second slice 622 of the file 62, the third slice 633 of the file 63, the fourth slice 644 of the file 64.
- the solution proposed by this embodiment to solve this problem consists here in directing the creation of the files 61 to 64 on a set of data servers 100 to 103 from previous observations of executions of the processes 65 to 68 of the application or applications to determine the future behavior in terms of simultaneous access to the files 61 to 64.
- the solution proposed by this embodiment of the invention is therefore based on the possibility of observing the behavior of an application or the application processes 65 to 68 during multiple executions and to register it in a database. knowledge to derive an ideal distribution profile of files 61 to 64 created on data servers 100 to 103.
- an analysis can be performed to detect simultaneous accesses performed by the application processes 65 to 68 or applications.
- the accesses are classified by access periods and by file regions accessed as a slot number, as shown in the table of Figure 6.
- FIG. 7 schematically represents an example of a second phase of several application processes using the solution proposed by the storage method according to one embodiment of the invention.
- the data storage performed is the same for all slices of all files. This storage is the following.
- the first slice 611 is stored on the server 100
- the second slice 612 is stored on the server 101
- the third slice 613 is stored on the server 102
- the fourth slice 614 is stored on the server 103.
- the first slice 621 is stored on the server 100
- the second slice 622 is stored on the server 101
- the third slice 623 is stored on the server 102
- the fourth slice 624 is stored on the server 103.
- the first slice 631 is stored on the server 100
- the second slice 632 is stored on the server 101
- the third slice 633 is stored on the server 102
- the fourth slice 634 is stored on the server 103.
- the first slice 641 is stored on the server
- the second slice 642 is stored on the server 101
- the third slice 643 is stored on the server 102
- the fourth slice 644 is stored on the server 103.
- the strategy will consist here in deducing from the table of FIG. 6 that the slices of one and the same rank of these four files 61 to 64 would ideally be placed on different data servers 100 to 103, in order to make them all work simultaneously and in parallel, during the second and third periods of time. This corresponds to the ideal distribution of files 61 to 64 on data servers 100 to 103 as shown in FIG.
- This mechanism can notably be implemented in the form of a function library intercepting the file creations of the application processes and performing this operation with predetermined parameters. This library will have access to the ideal placement information of files 61 to 64 that have been developed from the analysis of previous executions.
- FIG. 8 schematically represents an example of synthesis of the observation phase of the storage method according to one embodiment of the invention.
- Different applications 83 or different application process (s), are instrumented so that their behavior is observed during the execution of their execution.
- an observation software 82 of the behavior of these applications 83 observes their behavior to determine the profile of each of the applications 83. Once determined, the profile of each application 83 is stored on an archival space 81 application profiles 83.
- FIG. 8 schematically represents an example of synthesis of the parameterization phase of the storage method according to one embodiment of the invention.
- Optimal application parameterization software 84 reads on the archiving space 81 of the profiles of the applications 83, the profiles of the applications. With the aid of these application profiles, this optimal application programming software 84 will parameterize each of the applications so that it becomes an instrumented application 85 for the acceleration of its behavior, and more specifically for the reduction of its time. data exchange with the data storage servers.
- the applications 85 will be launched with the acceleration library which will be parameterized by analyzing the previous behaviors recorded in the knowledge base stored on the repository space 81
- the acceleration library will then tell the file system how to distribute the file slots on the disk space so as to avoid future conflicts of access and thus optimize the performance in terms of execution time in particular.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Environmental & Geological Engineering (AREA)
- Life Sciences & Earth Sciences (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR1763236A FR3076001B1 (fr) | 2017-12-27 | 2017-12-27 | Procede de stockage de donnees et procede d'execution d'application avec reduction du temps d'acces aux donnees stockees |
PCT/FR2018/053447 WO2019129958A1 (fr) | 2017-12-27 | 2018-12-20 | Procede de stockage de donnees et procede d'execution d'application avec reduction du temps d'acces aux donnees stockees |
Publications (1)
Publication Number | Publication Date |
---|---|
EP3732560A1 true EP3732560A1 (fr) | 2020-11-04 |
Family
ID=62017429
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP18842433.7A Pending EP3732560A1 (fr) | 2017-12-27 | 2018-12-20 | Procede de stockage de donnees et procede d'execution d'application avec reduction du temps d'acces aux donnees stockees |
Country Status (4)
Country | Link |
---|---|
US (1) | US11561934B2 (fr) |
EP (1) | EP3732560A1 (fr) |
FR (1) | FR3076001B1 (fr) |
WO (1) | WO2019129958A1 (fr) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR3076001B1 (fr) * | 2017-12-27 | 2021-05-28 | Bull Sas | Procede de stockage de donnees et procede d'execution d'application avec reduction du temps d'acces aux donnees stockees |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7010532B1 (en) * | 1997-12-31 | 2006-03-07 | International Business Machines Corporation | Low overhead methods and apparatus for shared access storage devices |
US8296812B1 (en) * | 2006-09-01 | 2012-10-23 | Vudu, Inc. | Streaming video using erasure encoding |
US8281181B2 (en) * | 2009-09-30 | 2012-10-02 | Cleversafe, Inc. | Method and apparatus for selectively active dispersed storage memory device utilization |
US8539197B1 (en) * | 2010-06-29 | 2013-09-17 | Amazon Technologies, Inc. | Load rebalancing for shared resource |
CN102142006B (zh) * | 2010-10-27 | 2013-10-02 | 华为技术有限公司 | 分布式文件系统的文件处理方法及装置 |
US10410137B2 (en) * | 2013-08-23 | 2019-09-10 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and system for analyzing accesses to a data storage type and recommending a change of storage type |
US10037340B2 (en) * | 2014-01-21 | 2018-07-31 | Red Hat, Inc. | Tiered distributed storage policies |
FR3076001B1 (fr) * | 2017-12-27 | 2021-05-28 | Bull Sas | Procede de stockage de donnees et procede d'execution d'application avec reduction du temps d'acces aux donnees stockees |
US10594773B2 (en) * | 2018-01-22 | 2020-03-17 | Spredfast, Inc. | Temporal optimization of data operations using distributed search and server management |
US11144502B2 (en) * | 2019-03-08 | 2021-10-12 | Netapp Inc. | Object store file system format for representing, storing, and retrieving data in an object store according to a structured format |
US11893064B2 (en) * | 2020-02-05 | 2024-02-06 | EMC IP Holding Company LLC | Reliably maintaining strict consistency in cluster wide state of opened files in a distributed file system cluster exposing a global namespace |
US20210390080A1 (en) * | 2020-06-15 | 2021-12-16 | Nutanix, Inc. | Actions based on file tagging in a distributed file server virtual machine (fsvm) environment |
-
2017
- 2017-12-27 FR FR1763236A patent/FR3076001B1/fr active Active
-
2018
- 2018-12-20 EP EP18842433.7A patent/EP3732560A1/fr active Pending
- 2018-12-20 US US16/958,314 patent/US11561934B2/en active Active
- 2018-12-20 WO PCT/FR2018/053447 patent/WO2019129958A1/fr unknown
Also Published As
Publication number | Publication date |
---|---|
FR3076001A1 (fr) | 2019-06-28 |
US20210064575A1 (en) | 2021-03-04 |
US11561934B2 (en) | 2023-01-24 |
FR3076001B1 (fr) | 2021-05-28 |
WO2019129958A1 (fr) | 2019-07-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1483673B1 (fr) | Methode de stockage de blocs de donnees dans une memoire | |
FR3000578A1 (fr) | Systeme et procede de calcul partage utilisant une fourniture automatisee de ressources informatiques heterogenes | |
EP2366147A1 (fr) | Gestionnaire physique de barriere de synchronisation entre processus multiples | |
EP2776927A1 (fr) | Procédé, programme d'ordinateur et dispositif d'allocation de ressources informatiques d'un cluster pour l'exécution d'un travail soumis audit cluster | |
US20050165823A1 (en) | Binary dependency database | |
WO2015004207A1 (fr) | Procede d'optimisation de traitement parallele de donnees sur une plateforme materielle | |
Gonçalves Junior et al. | A multi-criteria approach for assessing cloud deployment options based on non-functional requirements | |
WO2013107819A1 (fr) | Procédé d'optimisation de traitement parallèle de données sur une plateforme matérielle. | |
EP3732560A1 (fr) | Procede de stockage de donnees et procede d'execution d'application avec reduction du temps d'acces aux donnees stockees | |
EP3579108A1 (fr) | Contrôleur de partage de ressources d'une plate-forme informatique et procédé associé de partage des ressources | |
EP2856323B1 (fr) | Procédé, dispositif et programme d'ordinateur de contrôle dynamique de distances d'accès mémoire dans un système de type numa | |
FR2906955A1 (fr) | Systeme de gestion de droits numeriques. | |
EP2709008B1 (fr) | Procédé et dispositif de décompte du temps déporté pour unité de traitement dans un système de traitement de l'information | |
EP3724752B1 (fr) | Procede de gestion du systeme de fichiers d'un terminal informatique | |
EP2704010A1 (fr) | Procédé et dispositif de traitement de commandes dans un ensemble d'éléments informatiques | |
FR2897961A1 (fr) | Procede de gestion de l'execution d'un jeu video pour la diffusion en temps reel de publicites dynamiques | |
EP2953029B1 (fr) | Methodes et systemes de test de performances a debit configurable | |
EP3239851B1 (fr) | Gestion de l'accès a des données dans un système de stockage | |
FR2980007A1 (fr) | Procede, dispositif et programme d'ordinateur pour allouer dynamiquement des ressources d'un cluster a l'execution de processus d'une application | |
Keita | Big Data et Technologies de Stockage et de Traitement des Données Massives: Comprendre les bases de l’écosystème HADOOP (HDFS, MAPREDUCE, YARN, HIVE, HBASE, KAFKA et SPARK) | |
FR3073298A1 (fr) | Procede et dispositif de recherche de marge d’optimisation de ressources d’une chaine applicative | |
FR3094509A1 (fr) | Système de stockage redondant de données, procédé et programme d’ordinateur correspondants. | |
EP4113297A1 (fr) | Procédé de gestion des travaux dans un système informatique et système associé | |
EP2572253B1 (fr) | Procede d'optimisation de gestion de veille d'un microprocesseur permettant la mise en oeuvre de plusieurs coeurs logiques et programme d'ordinateur mettant en oeuvre un tel procede | |
CN117076696A (zh) | 数据处理方法、装置、电子设备及存储介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: UNKNOWN |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
17P | Request for examination filed |
Effective date: 20200625 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
AX | Request for extension of the european patent |
Extension state: BA ME |
|
DAV | Request for validation of the european patent (deleted) | ||
DAX | Request for extension of the european patent (deleted) | ||
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
17Q | First examination report despatched |
Effective date: 20220307 |
|
RAP1 | Party data changed (applicant data changed or rights of an application transferred) |
Owner name: COMMISSARIAT A L'ENERGIE ATOMIQUE ET AUX ENERGIES ALTERNATIVES Owner name: BULL SAS |
|
P01 | Opt-out of the competence of the unified patent court (upc) registered |
Effective date: 20230330 |