CN109271249B - Cloud container pre-copy online migration method based on P.haul framework - Google Patents
Cloud container pre-copy online migration method based on P.haul framework Download PDFInfo
- Publication number
- CN109271249B CN109271249B CN201810840047.7A CN201810840047A CN109271249B CN 109271249 B CN109271249 B CN 109271249B CN 201810840047 A CN201810840047 A CN 201810840047A CN 109271249 B CN109271249 B CN 109271249B
- Authority
- CN
- China
- Prior art keywords
- container
- migration
- host
- migrated
- memory
- 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.)
- Active
Links
Images
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]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
- G06F9/5088—Techniques for rebalancing the load in a distributed system involving task migration
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
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 discloses a cloud container pre-copying online migration method based on a P.haul framework, aiming at the field of container online migration, under the condition of ensuring uninterrupted access, containers among different physical machines or cloud platforms are moved so as to ensure that the containers can continue the working state of a source host after migration, and meanwhile, the container shutdown time caused in the migration process is reduced by adopting a pre-copying and history-based transmission delay mechanism. More flexible application is provided for the container, and basic work is made for load balancing and resource scheduling of the container.
Description
Technical Field
The invention relates to the technical field of online migration of cloud computing containers, in particular to a Docker container pre-copying online migration method based on a P.haul framework.
Background
The container technology provides an increasingly wide application environment for cloud computing, and in order to provide more flexible application for a container life cycle and adapt to richer use scenes, an online migration technology to the container is required. Online migration of containers, also called live migration, refers to the process of moving containers between different physical machines or cloud platforms while ensuring uninterrupted access. In the migration process, state information such as a memory, a file system, network connection and the like when a container on a source host runs needs to be migrated to a target host, and meanwhile, the container can be ensured to continue the working state of the source host after the container is migrated.
The online migration technology can transfer the ongoing work of the source host to the destination host in real time in the migration process, so that a user can not even feel that the migration occurs. Today, many online migration schemes have mature applications in virtual machines, and the online migration field of containers, especially the online migration based on Docker, does not have a set of very mature solutions at present.
In recent years, academic circles propose migration strategies of 'stop-copy' and 'pre-copy' of containers according to the idea of virtual machine migration, and realize online container migration of Docker. There have also been studies attempting migration strategies that use logging-and-playback, but they rely on the logging of container events to be migrated and have many limitations, so they are not yet mature in container migration.
Disclosure of Invention
The invention aims to overcome the defects and shortcomings of the prior art, and provides a cloud container pre-copying online migration method based on a P.haul framework, which can realize online migration of a Docker container between a source host and a destination host, and can effectively reduce data transmission amount and total migration time in the migration process through pre-copying and a history-based transmission delay mechanism.
The purpose of the invention is realized by the following technical scheme: a cloud container pre-copy online migration method based on a P.haul framework comprises the following steps:
t1, container configuration acquisition and migration network establishment steps: acquiring a configuration file of a container to be migrated, and establishing network connection from an origin host to a destination host so as to facilitate related migration work in the container migration step;
t2, a container transferring step, comprising:
a container assembly migration substep: the system comprises a storage unit, a storage unit and a server, wherein the storage unit is used for storing container component related files and working state files;
memory migration substep: the system comprises a target host, a migration module and a migration module, wherein the migration module is used for iteratively sending a current working state file and memory original data of a container to be migrated to the target host;
t3, container rebuilding and cleaning steps: the method is used for reconstructing a new container in the target host according to the state file and the memory data obtained in the step T2, and cleaning the container and the configuration file in the source host.
Preferably, the step T1 of obtaining configuration information of the container to be migrated and generating a checkpoint directory specifically includes:
step T1.1, according to the id of the container to be migrated, using a docker insert command to read the Tty condition of the current container, if the container is opened, the container cannot be migrated, and returning error information;
step T1.2, establishing a network link between the source host and the target host; the network link needs to establish two network links: a command transfer link and a data transfer link; the command transmission link adopts an RPC remote call mode, namely, a command is sent in a source host and runs in a target host; the data link adopts an Rsync data synchronization command to transmit related data from the source host to the destination host;
and step T1.3, acquiring a state file of the container to be migrated by using a command of docker checkpoint create-checkpoint-dir ═ checkpoint purpose } { container id } { container serial number }, and storing the state file in a checkpoint directory for migrating to a destination host.
Further, the state file of the container to be migrated, which needs to be acquired in step T1.3, includes: the system comprises a container description information file, an opened file descriptor information file, a CPU information file and a memory mapping table, wherein the information is stored in corresponding files in a JSON data format; meanwhile, the original data of the memory page during the operation of the container is acquired and stored in a binary file by taking the size of 4Kbyte as a unit.
Further, the two network links in the step T1.2 are constructed in a p.haul framework based on a C/S structure, and in order to ensure that data transmission and control operations can be performed between the source host and the destination host, the two network links are respectively transmitted for control commands and data by establishing two communication sockets; the control command socket adopts an RPC remote procedure control mode, and a source host sends a related control command to a target host; the data socket adopts a TCP protocol and is used for sending the running state file acquired from the source host computer to the destination host computer.
Preferably, the container component migration substep described in T2 is used to transfer the relevant components of the container in the source host to the destination host, and 3 directories need to be migrated during the migration process of Docker: the root file system, the container configuration directory and the container operation metadata directory specifically include:
and 2, sending an RPC command to the target host by the source host, and rebuilding the three corresponding directories in the step 1 at the target host.
Preferably, the memory migration substep in T2 is configured to transmit the memory raw data of the container to be migrated to the destination host in an iterative manner, so as to reduce the downtime during container migration, and specifically includes:
step 2, inputting a command p.haul Docker < id > -fdrpc < fd > -fdmem < fd > -pre-dump, using a p.haul frame, and starting the online pre-copy migration of the Docker container;
and 3, transmitting the newly generated dirty memory page to the destination host in an incremental mode by the p.haul during each iteration.
Specifically, an iteration pre-copy stop condition is set, and one of three stop conditions is determined to be adopted according to the size of the memory data derived from the container to be migrated: when the size of the migration memory data is smaller than 128M, generating no more than 64 memory dirty pages in the last pre-copying by adopting a condition (1); when the size of the migration memory data is larger than 128M and smaller than 512M, the page copy number of the last pre-copy is increased by less than 10% by adopting the condition (2); when the size of the migration memory data is larger than 512M, the iteration number of the condition (3) is not more than 8 times.
Further, the memory dirty page increment transmission adopts a history-based transmission delay mechanism, and specifically includes:
Step 2, the bitmap value in the process of the nth iteration is dirtymapn;
And 3, determining whether the derivation is carried out at this time by comparing the specific values of the corresponding bits of the two bitmaps.
Specifically, the memory dirty page increment transmission adopts a history-based transmission delay mechanism:
(1) when dirtymapnAnd dirtymapn-1When both are 0, it means thatPages are never modified, not copied;
(2) when dirtymapnIs 1, dirtymapn-1A value of 0 indicates that the page is newly generated and needs to be copied;
(3) when dirtymapnIs 0, dirtymapn-1A value of 1 indicates that the page has been modified, and may be a memory page that has been continuously modified without being copied, and may still not be modified next time, so that the page needs to be copied;
(4) when dirtymapnAnd dirtymapn-1All 1's indicate that the page has been modified continuously, which indicates that the page is likely to be modified the next time, and therefore the page is not copied this time.
Preferably, the container rebuilding and cleaning step T3 is configured to rebuild a new container on the destination host according to the obtained container state file and the obtained original memory data, delete the migrated container in the source host, and delete the corresponding checkpoint directory, and specifically includes:
step T3.1, deleting the migrated container in the source host by using a docker rm < id > command;
step T3.2, restarting a docker system of the target host to enable the newly migrated container configuration information and the root file system to take effect;
and step T3.3, reconstructing a new container in the destination host by using a docker start-checkpoint-dir ═ checkpoint directory > - - -checkpoint ═ checkpoint sequence number > < container id >.
Compared with the prior art, the invention has the following advantages and beneficial effects:
(1) the invention can realize the online migration of the Docker container between the source host and the destination host.
(2) The invention is based on a mature P.haul framework and has good stability and expansibility.
(3) The invention can effectively reduce the data transmission amount and the total migration time in the migration process through the pre-copy and history-based transmission delay mechanism.
Drawings
FIG. 1 is a block diagram of an embodiment container online migration system.
FIG. 2 is a network link topology diagram of an embodiment migration system.
FIG. 3 is an embodiment container migration flow diagram.
FIG. 4 is a schematic diagram of an embodiment of iterative pre-copy.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited thereto.
Example 1
The invention relates to a cloud container pre-copy online migration method based on a P.haul framework, which is mainly applied to online migration of Docker containers among different hosts and is the basic work for realizing container load balancing and container resource scheduling in a cloud environment.
The following is specifically described with reference to a container online migration system structure diagram (fig. 1), a migration system network link topology diagram (fig. 2), a container migration flow diagram (fig. 3), and a history-based transmission delay mechanism description table (table 1).
A cloud container pre-copy online migration method based on a p.haul framework, which adopts a system structure diagram shown in fig. 1, and the implementation method specifically includes the following steps, and a flow chart is shown in fig. 3:
t1, container configuration acquisition and migration network establishment steps:
acquiring a configuration file of a container to be migrated, and establishing network connection from an origin host to a destination host so as to facilitate related migration work in the container migration step, wherein the specific steps comprise:
and step T1.1, reading the current Tty condition of the container by using a docker insert command according to the id of the container to be migrated, and if the container is opened, failing to migrate the container and returning error information.
And step T1.2, establishing a network link between the source host and the target host. In the present system, two network links need to be established: a command transfer link and a data transfer link. The command transmission link adopts an RPC remote call mode, namely, a command is sent in a source host and runs in a target host; the data link uses the Rsync data synchronization command to transmit the relevant data from the source host to the destination host.
And step T1.3, acquiring a state file of the container to be migrated by using a docker checkpoint create-checkpoint-dir ═ checkpoint-object > < container id > < checkpoint sequence number > command, and storing the state file in a checkpoint directory for migrating to a destination host.
T2, a container migration step, configured to migrate related components, state files, and memory raw data required by the migration container from the source host to the destination host, the specific step includes:
and step T2.1, respectively transmitting the mirror image file in the/var/lib/docker/image directory, the container configuration file in the/var/lib/docker/containers directory and the container operation metadata of the/var/run/doc-ker/execdriver/native to the destination host by using an Rsync command.
And step T2.2, acquiring a state file of the container to be migrated by using a command of docker checkpoint create-checkpoint-dir ═ checkpoint-destination } { container id } { container sequence number }, and storing the state file in a checkpoint directory for migrating to a destination host. And the source host sends an RPC command to the target host, and the reconstruction of the three corresponding directories is carried out at the target host.
And step T2.3, inputting a command p.haul Docker < id > -fdrpc < fd > -fdmem < fd > -pre-dump, starting the online pre-copy migration of the Docker container, and migrating according to a pre-copy stopping condition and a history-based transmission delay mechanism.
T3, container rebuilding and cleaning steps, which are used to rebuild a new container on the destination host according to the obtained container state file and the obtained memory original data, delete the migrated container in the source host, and delete the corresponding checkpoint directory, and specifically include:
and step T3.1, deleting the migrated container in the source host by using a docker rm < id > command.
And step T3.2, restarting a docker system of the target host to enable the newly migrated past container configuration information and the root file system to take effect.
And step T3.3, reconstructing a new container in the destination host by using a docker start-checkpoint-dir ═ checkpoint directory > - - -checkpoint ═ checkpoint sequence number > < container id >.
Tty terminal, which refers to a terminal interacting with the container, and the p.haul cannot transfer the container with the interactive terminal at present, an error message is returned in step T1 once it is detected that Tty is opened by the container, and the whole transfer process is exited.
And the Checkpoint state file directories are used for describing an information file, an opened file descriptor information file, a CPU information file and a memory mapping table by the state file containers, and the information is stored in corresponding files in a JSON data format. Meanwhile, the original data of the memory page during the operation of the container is acquired and stored in a binary file by taking the size of 4Kbyte as a unit.
The source host and the destination host are linked through a network, as can be seen from fig. 2, a p.haul framework is based on a C/S structure, and in order to ensure that data transmission and control operations can be performed between the source host and the destination host, two communication sockets are established between the source host and the destination host to respectively transmit control commands and data. The control command socket mainly adopts an RPC remote process control mode, and a source host sends a related control command to a target host. The data socket adopts a TCP protocol, and is mainly used for sending the running state file acquired from the source host computer to the destination host computer.
The pre-copy iteration stop condition is that, because the more the iteration times are, the smaller the last transmission downtime will be, but the excessive iteration times will cause the total migration time to be too high, different pre-copy iteration stop times need to be set according to the size of the memory occupied by the container: when the size of the migration memory data is smaller than 128M, generating no more than 64 memory dirty pages (256K) in the last pre-copying by adopting a condition (1); when the size of the migration memory data is larger than 128M and smaller than 512M, the page copy number of the last pre-copy is increased by less than 10% by adopting the condition (2); when the size of the migration memory data is larger than 512M, the iteration number of the condition (3) is not more than 8 times.
Based on a historical transmission delay mechanism, whether the corresponding memory page is transmitted in the iteration is determined by comparing the memory dirty page bitmaps which are continuously performed twice. As can be seen from table 1:
(1) when dirtymapnAnd dirtymapn-1When both are 0, it means that the page has never been modified and is not copied;
(2) when dirtymapnIs 1, dirtymapn-1A value of 0 indicates that the page is newly generated and needs to be copied;
(3) when dirtymapnIs 0, dirtymapn-1A value of 1 indicates that the page has been modified, and may be a memory page that has been continuously modified without being copied, and may still not be modified next time, so that the page needs to be copied;
(4) when dirtymapnAnd dirtymapn-1All 1's indicate that the page has been modified continuously, which indicates that the page is likely to be modified the next time, and therefore the page is not copied this time.
Table 1 history-based transmission delay mechanism description table
dirtymapn-1 | 0 | 0 | 1 | 1 |
dirtymapn | 0 | 1 | 0 | 1 |
Whether to copy | Whether or not | Is that | Is that | Whether or not |
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (6)
1. A cloud container pre-copy online migration method based on a P.haul framework is characterized by comprising the following steps:
t1, container configuration acquisition and migration network establishment steps: acquiring a configuration file of a container to be migrated, and establishing network connection from an origin host to a destination host so as to facilitate related migration work in the container migration step;
t2, a container transferring step, comprising:
a container assembly migration substep: the system comprises a storage unit, a storage unit and a server, wherein the storage unit is used for storing container component related files and working state files;
memory migration substep: the system comprises a target host, a migration module and a migration module, wherein the migration module is used for iteratively sending a current working state file and memory original data of a container to be migrated to the target host;
t3, container rebuilding and cleaning steps: the method is used for reconstructing a new container in the target host according to the state file and the memory data obtained by the transmission in the step T2, and cleaning the container and the configuration file in the source host;
the container component migration substep described in T2, which is used to transfer the relevant components of the container in the source host to the destination host, requires 3 directories to be migrated during the migration process of Docker: the root file system, the container configuration directory and the container operation metadata directory specifically include:
step 1, transmitting mirror image files in a/var/lib/docker/image directory, container configuration files in a/var/lib/docker/containers directory and/var/run/docker-r/execdriver/native container operation metadata to a target host by using an Rsync command;
step 2, the source host sends an RPC command to the target host, and the target host reconstructs the three corresponding directories in the step 1;
the memory migration substep described in T2 is configured to transmit the memory raw data of the container to be migrated to the destination host in an iterative manner, so as to reduce the downtime during container migration, and specifically includes:
step 1', setting iteration pre-copy stopping conditions, and determining one of different stopping conditions to be adopted according to the size of memory data derived from a container to be migrated;
step 2', inputting a command p.haul Docker < id > -fdrpc < fd > -fdmem < fd > -pre-dump, using a p.haul frame, and starting the online pre-copy migration of the Docker container;
step 3', the p.haul transmits the newly generated dirty memory page to the target host in an increment mode during each iteration;
setting an iteration pre-copy stopping condition, and determining to adopt one of three stopping conditions according to the size of the memory data derived from the container to be migrated: when the size of the migration memory data is smaller than 128M, generating no more than 64 memory dirty pages in the last pre-copying by adopting a condition (1); when the size of the migration memory data is larger than 128M and smaller than 512M, the page copy number of the last pre-copy is increased by less than 10% by adopting the condition (2); when the size of the migration memory data is larger than 512M, the iteration number of the condition (3) is not more than 8 times;
the memory dirty page increment transmission adopts a history-based transmission delay mechanism, and specifically comprises the following steps:
step 1', storing dirty page bitmap value dirtyymap in n-1 th iterationn-1;
Step 2', the bitmap value in the n-th iteration is dirtymapn;
And 3', determining whether to derive the bitmap or not by comparing the specific values of the corresponding bits of the two bitmaps.
2. The p.haul framework-based cloud container pre-copy online migration method according to claim 1, wherein in step T1, acquiring configuration information of a container to be migrated specifically includes:
step T1.1, according to the id of the container to be migrated, using a docker insert command to read the Tty condition of the current container, if the container is opened, the container cannot be migrated, and returning error information;
step T1.2, establishing a network link between the source host and the target host; the network link needs to establish two network links: a command transfer link and a data transfer link; the command transmission link adopts an RPC remote call mode, namely, a command is sent in a source host and runs in a target host; the data link adopts an Rsync data synchronization command to transmit related data from the source host to the destination host;
and step T1.3, acquiring a state file of the container to be migrated by using a command of docker checkpoint create-checkpoint-dir ═ checkpoint purpose } { container id } { container serial number }, and storing the state file in a checkpoint directory for migrating to a destination host.
3. The p.haul framework-based cloud container pre-copy online migration method according to claim 2, wherein in step T1.3, the state file of the container to be migrated that needs to be acquired includes: the system comprises a container description information file, an opened file descriptor information file, a CPU information file and a memory mapping table, wherein the information is stored in corresponding files in a JSON data format; meanwhile, the original data of the memory page during the operation of the container is acquired and stored in a binary file by taking the size of 4Kbyte as a unit.
4. The cloud container pre-copy online migration method based on the p.haul framework as claimed in claim 2, wherein in step T1.2, two network links are constructed in the p.haul framework based on a C/S structure, and in order to ensure that data transmission and control operations can be performed between the source host and the destination host, the two network links transmit control commands and data respectively by establishing two communication sockets; the control command socket adopts an RPC remote procedure control mode, and a source host sends a related control command to a target host; the data socket adopts a TCP protocol and is used for sending the running state file acquired from the source host computer to the destination host computer.
5. The cloud container pre-copy online migration method based on the p.haul framework, as claimed in claim 1, wherein the internal dirty page increment transmission employs a history-based transmission delay mechanism:
(1) when dirtymapnAnd dirtymapn-1When both are 0, no copy is needed;
(2) when dirtymapnIs 1, dirtymapn-1When 0, it needs to be copied;
(3) when dirtymapnIs 0, dirtymapn-1When 1, it needs to be copied;
(4) when dirtymapnAnd dirtymapn-1When both are 1, the page is not copied this time.
6. The p.haul-frame-based cloud container pre-copy online migration method according to claim 1, wherein in the T3, container reconstruction and cleaning steps, the method is used for reconstructing a new container on the destination host according to the acquired container state file and the acquired memory original data, deleting a migrated container in the source host, and deleting a corresponding checkpoint directory, and specifically includes:
step T3.1, deleting the migrated container in the source host by using a docker rm < id > command;
step T3.2, restarting a docker system of the target host to enable the newly migrated container configuration information and the root file system to take effect;
and step T3.3, reconstructing a new container in the destination host by using a docker start-checkpoint-dir ═ checkpoint directory > - - -checkpoint ═ checkpoint sequence number > < container id >.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810840047.7A CN109271249B (en) | 2018-07-27 | 2018-07-27 | Cloud container pre-copy online migration method based on P.haul framework |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810840047.7A CN109271249B (en) | 2018-07-27 | 2018-07-27 | Cloud container pre-copy online migration method based on P.haul framework |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109271249A CN109271249A (en) | 2019-01-25 |
CN109271249B true CN109271249B (en) | 2022-02-08 |
Family
ID=65153302
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810840047.7A Active CN109271249B (en) | 2018-07-27 | 2018-07-27 | Cloud container pre-copy online migration method based on P.haul framework |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109271249B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110119377B (en) * | 2019-04-24 | 2021-01-15 | 华中科技大学 | Docker container-oriented online migration system implementation and optimization method |
CN110321198B (en) * | 2019-07-04 | 2020-08-25 | 广东石油化工学院 | Container cloud platform computing resource and network resource cooperative scheduling method and system |
CN111190702A (en) * | 2020-01-14 | 2020-05-22 | 山东超越数控电子股份有限公司 | Cross-node hot migration method for Docker |
US11748153B2 (en) | 2020-11-25 | 2023-09-05 | International Business Machines Corporation | Anticipated containerized infrastructure used in performing cloud migration |
CN114416284A (en) * | 2021-12-24 | 2022-04-29 | 北京百度网讯科技有限公司 | Distributed operating system control method, apparatus, device, medium, and program product |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107526626A (en) * | 2017-08-24 | 2017-12-29 | 武汉大学 | A kind of Docker containers thermomigration process and system based on CRIU |
CN107992353A (en) * | 2017-07-31 | 2018-05-04 | 南京邮电大学 | A kind of container dynamic migration method and system based on minimum transition amount |
CN108182105A (en) * | 2017-12-12 | 2018-06-19 | 苏州大学 | Local dynamic station moving method and control system based on Docker container techniques |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9852137B2 (en) * | 2015-09-23 | 2017-12-26 | International Business Machines Corporation | Container storage migration |
CN108279969A (en) * | 2018-02-26 | 2018-07-13 | 中科边缘智慧信息科技(苏州)有限公司 | Stateful service container thermomigration process based on memory compression transmission |
-
2018
- 2018-07-27 CN CN201810840047.7A patent/CN109271249B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107992353A (en) * | 2017-07-31 | 2018-05-04 | 南京邮电大学 | A kind of container dynamic migration method and system based on minimum transition amount |
CN107526626A (en) * | 2017-08-24 | 2017-12-29 | 武汉大学 | A kind of Docker containers thermomigration process and system based on CRIU |
CN108182105A (en) * | 2017-12-12 | 2018-06-19 | 苏州大学 | Local dynamic station moving method and control system based on Docker container techniques |
Non-Patent Citations (2)
Title |
---|
Voyager: Complete Container State Migration;Shripad Nadgowda et al;《2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS)》;20170717;第2137-2142页 * |
一种基于Linux容器的动态迁移方法;姜渠;《信息技术与信息化》;20150228(第2期);第66-68页 * |
Also Published As
Publication number | Publication date |
---|---|
CN109271249A (en) | 2019-01-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109271249B (en) | Cloud container pre-copy online migration method based on P.haul framework | |
US9170892B2 (en) | Server failure recovery | |
CN111182067B (en) | Data writing method and device based on interplanetary file system IPFS | |
US20230079868A1 (en) | Techniques for coordinating parallel performance and cancellation of commands in a storage cluster system | |
CN103268318B (en) | A kind of distributed key value database system of strong consistency and reading/writing method thereof | |
US9286298B1 (en) | Methods for enhancing management of backup data sets and devices thereof | |
CN108073656A (en) | A kind of method of data synchronization and relevant device | |
US20120130950A1 (en) | Data replication to multiple data nodes | |
TWI510932B (en) | Block storage gateway module, mediator system for storage, cloud storage system, method for providing access to block storage, mediating method for storage, and content delivery apparatus | |
CN111274252A (en) | Block chain data chaining method, device, storage medium and server | |
WO2018108158A1 (en) | Method and device for storing data based on majority, and storage medium and apparatus | |
CN107295030B (en) | Data writing method and device, data processing method, device and system | |
CN111026510A (en) | Cross-storage thermal migration method for virtual machine in cloud environment | |
CN109597903A (en) | Image file processing apparatus and method, document storage system and storage medium | |
CN107181774B (en) | Data movement between distributed data centers | |
CN114564339A (en) | Disk image file cross-platform migration method and system | |
CN109597671B (en) | Desktop cloud implementation method compatible with centralized and distributed operation | |
CN103856511A (en) | Data packet upload method, client terminal, nodes, information server and system | |
WO2024148824A1 (en) | Data processing method, system and apparatus, storage medium, and electronic device | |
CN105471955A (en) | Writing method of distributed file system, client device and distributed file system | |
US11144232B2 (en) | Storage system with efficient snapshot pair creation during synchronous replication of logical storage volumes | |
CN115048190B (en) | Virtual machine snapshot management method, system and computer readable medium | |
CN115562805A (en) | Resource migration method and device and electronic equipment | |
CN115167776A (en) | Method and system for migrating distributed storage data | |
CN113965582B (en) | Mode conversion method and system, and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |