CN111506271B - Data segment unit transmission differentiation management method and distributed block storage system - Google Patents
Data segment unit transmission differentiation management method and distributed block storage system Download PDFInfo
- Publication number
- CN111506271B CN111506271B CN202010603735.9A CN202010603735A CN111506271B CN 111506271 B CN111506271 B CN 111506271B CN 202010603735 A CN202010603735 A CN 202010603735A CN 111506271 B CN111506271 B CN 111506271B
- Authority
- CN
- China
- Prior art keywords
- data segment
- group
- segment unit
- metadata
- groups
- 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
- 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
-
- 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
- G06F3/0644—Management of space entities, e.g. partitions, extents, pools
-
- 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]
Abstract
The invention discloses a data segment unit transmission differentiation management method, which comprises the following steps: s1, grouping a plurality of data segment units, wherein the number of the grouped units is between [ M/20] and [ M/10], and is an integer, and M is the maximum value of the number of the data segment units which can be managed in one group; s2, grouping a data segment unit into the group to which the data segment unit belongs S3, if the total amount of the metadata in the group to which the data segment unit belongs in the step S2 is larger than C, grouping the group again, wherein C is the size of the maximum data packet reported to the information management service node; s4, if the metadata in the two groups are respectively larger than C/20 and smaller than C/5, merging the two groups; and S5, reporting the data segment unit to the information management service node by taking the group as a unit. The invention can better utilize network resources and simultaneously solve the problem of network congestion in the prior art. The invention also provides a distributed block storage system, which is used for realizing the data segment unit transmission differentiation management method.
Description
Technical Field
The invention belongs to the field of computer information security, and particularly relates to a data segment unit transmission differentiation management method and a distributed block storage system.
Background
In DBS (Distributed Block Storage, Distributed Block Storage system), metadata information of all data segment unit segment units needs to be reported to an information management service node in a unified manner, and the information management service node collects information of all data segment unit segment units, analyzes, sorts, filters, and then displays the information to a user. In the prior art, metadata information of all segment units is reported to an information management service node at regular time, so that the following problems exist: on one hand, the network load pressure is high, and network congestion is easily caused; on the other hand, the information management service node needs to continuously process the metadata information of all the segment units, and the information management service node is exposed to a great processing pressure.
The granted announcement date is 22 3 and 2017, the granted announcement number is CN104967639B, and the patent name is an information reporting method, an information reporting device and a chinese patent of a data node server, which disclose a technical scheme for reducing the burden on the metadata server caused by reporting a large amount of metadata information. The technical scheme comprises the following steps: when the metadata information of the data block needs to be reported to the metadata server, determining the metadata information of the data block reported to the metadata server; if the data node server reports the metadata information of the data blocks for the first time, the reported metadata information of the data blocks comprises the metadata information of all the data blocks in the data node server, and if the data node server does not report the metadata information of the data blocks for the first time, the reported metadata information of the data blocks is the metadata information of partial data blocks in the data node server; and the data node server reports the reported metadata information of the data block to the metadata server.
The application publication date is 2017, 8, 18, the application publication number is CN107070713A, and a Chinese patent with a patent name of a data monitoring processing method discloses another technical scheme, which can filter unnecessary local state warning reports and whole-node alternate query triggered by a short-time or low-intensity exceeding threshold, thereby reducing the number of times of reporting state information by each monitoring node, reducing the network load, improving the reliability of data safety monitoring in the whole system, reducing the probability of false alarm and false alarm, and reducing the resource consumption of the whole system.
Different from the technical scheme, the invention provides another technical scheme to solve the problems of network congestion and excessive data processing pressure of the information management service node in the prior art.
Disclosure of Invention
1. Problems to be solved
The invention provides a data segment unit transmission differentiation management method, aiming at the problems of network congestion and excessive data processing pressure of an information management service node in the prior art.
2. Technical scheme
In order to solve the problems, the technical scheme adopted by the invention is as follows: a data segment unit transmission differentiation management method comprises the following steps:
s1, grouping a plurality of data segment units, wherein the number of the grouped units is between [ M/20] and [ M/10], and is an integer, and M is the maximum value of the number of the data segment units which can be managed in one group;
s2, grouping a data segment unit into the group to which it belongs,
s3, if the total amount of metadata in the group to which the data segment unit belongs in the step S2 is larger than C, grouping the group again, wherein C is the size of the maximum data packet reported to the information management service node;
s4, if the metadata in the two groups are respectively larger than C/20 and smaller than C/5, combining the two groups into one group;
and S5, reporting the data segment unit to the information management service node by taking the group as a unit.
Compared with the prior art that metadata in all data segment units are reported to an information management service node at one time at regular time, the technical scheme can better utilize network resources and solve the problem of network congestion in the prior art.
Further, when reporting the data segment unit, the method further includes the following steps:
s21, carrying out hash transformation on the metadata in all the data segment units in the current group to obtain hash values, and comparing the hash value of each data segment unit with the hash value of the data segment unit reported last time;
s22, if the hash values of the two times are equal, only reporting the ID of the data segment unit in the current group to the information management service node;
and S23, if the hash values of the two times are not equal, reporting all metadata in the data segment units with unequal hash values to the information management service node.
The technical scheme can solve the problem of repeated reporting of the metadata in the data segment unit and further relieve network congestion pressure.
Further, if the metadata of the current group is reported for the first time, the hash value obtained by performing hash transformation on the metadata in the data segment unit is compared with G, wherein G is a non-positive integer.
Further, the step S2 is specifically: and searching a group ID related to Y, wherein Y is a remainder obtained after X/M, X is the ID of a data segment unit needing to be put into the group, and then putting the data segment unit into the searched group for management. For the newly added data segment unit, it needs to be put into an appropriate group for management.
Further, the search is performed in a red-black tree.
Further, the data segment units are divided into two groups in step S1. Since the splitting and merging of the groups can be performed according to the size of the metadata in the groups, all data segment units are divided into two groups in an ideal state, and then the splitting and merging of the groups are performed according to the situation, so that the method is simpler to implement.
Further, the group IDs of two groups into which the data segment units are divided are M/2 and M respectively, wherein M is an even number; or (M + 1)/2 and M, wherein M is an odd number;
or, the group IDs of the two groups are 1 and M/2, respectively, wherein M is an even number, or 1 and (M + 1)/2, wherein M is an odd number.
The group IDs of the two groups are respectively set to 1, M/2 or (M + 1)/2, M, so that the data segment units managed in the two groups are balanced as much as possible, and the possibility of splitting of the subsequent groups can be reduced as much as possible.
Further, if the group IDs of the two groups into which the data segment unit is divided are M/2 and M respectively, wherein M is an even number, or (M + 1)/2 and M, wherein M is an odd number, searching a group with the group ID larger than or equal to Y in the red-black tree, and then putting the data segment unit into the searched group;
or if the group IDs of the two groups into which the data segment unit is divided are 1 and M/2, respectively, where M is an even number, or 1 and (M + 1)/2, where M is an odd number, then searching for a group whose group ID is less than or equal to Y, and then placing the data segment unit into the searched group, where Y is a remainder obtained after X/M, and X is an ID of the data segment unit that needs to be placed into the group.
The invention also provides a distributed block storage system for realizing the data segment unit transmission differentiation management method.
3. Advantageous effects
Compared with the prior art, the invention has the beneficial effects that:
(1) according to the invention, the metadata information of the segment unit of the data segment is managed in a differentiated manner, so that the reporting performance of the metadata information is improved;
(2) the invention manages the metadata of the segment unit of the data segment, only reports the metadata information which changes, and can effectively solve the problems of network congestion and overlarge data processing pressure of the information management service node in the process of reporting the metadata information in an extremely large scale.
Drawings
FIG. 1 is a schematic diagram of a packet flow for newly creating a segment unit of a data segment in the present invention;
FIG. 2 is a flow chart illustrating the deletion of a segment unit from a group according to the present invention;
FIG. 3 is a diagram illustrating metadata of a segment unit according to the present invention.
Detailed Description
The invention is further described with reference to specific examples.
In the field of data storage, a disk storing data is usually divided into a plurality of volumes, and then the volumes are divided into a plurality of data segment units; the information management service node is used for collecting all segment unit information, and displaying the information to a user after certain filtering processing, so that the user feels whether operations such as capacity expansion maintenance of the storage node are needed or not by browsing the information. The invention provides differential management of segment unit of data segment, aiming at managing two parts of segment unit grouping of data segment and metadata information of segment unit of data segment in group.
The data segment unit grouping is mainly used for grouping management of the data segment unit segment units, and due to the fact that the number of the data segment units is large, the size of a data packet reported each time can be reduced by reporting the data segment units according to the groups, and congestion pressure of network bandwidth is reduced.
The management of the metadata information of the segment unit of the data segment unit in the group mainly manages the metadata in the group, distinguishes the metadata information needing to be reported, and correspondingly processes the feedback information of the information management service node so as to reduce the processing pressure of the information management node.
The following describes the data segment unit grouping and the intra-group data segment unit metadata information management in detail. First, a data segment unit packet includes the following steps:
1. boundary value of prediction set
The invention manages the data segment unit in groups, divides the data segment unit segment into several groups, because the number A of the data segment unit that the information management service node can manage is determined by the calculation performance of the service node, the total capacity of the managed disk, the size of the data segment unit segment, etc. the value of A is not detailed here, but we can calculate and obtain the value when the system is started, the maximum value B of the number of the data segment unit managed in each group is also fixed, B = C/D, wherein C is the size of the largest data packet reported to the information management service node, the value of C mainly depends on the network environment, if the value of C is too small, the time consumption is too long when reporting to the information management service node, if the value of C is too small, the metadata may need to be divided into more data packets for uploading, which causes waste of network resources; d is the maximum metadata size of each segment unit, then the upper limit M of the number of data segment units that can be managed within a group, i.e. the maximum boundary value of the group, M = a/B, where a is the number of data segment units that can be managed by the information management service node and B is the maximum value of the number of data segment units that can be managed within each group. And after the maximum boundary value M of the groups is obtained, grouping the segment units, wherein the number of the grouped groups is an integer between [ M/20] and [ M/10 ].
For ease of understanding, the present invention is illustrated with an upper limit of the number of segment units being 128 (actually, the number is much larger than this value), the size of the segment unit managed in each group being 4 (actually, the number is also much larger than this value), and the boundary value of the group being 32, that is, we need to create 32 groups at most, and the group ID being 1 to 32.
If 32 groups are created from the beginning, when the number of segment units is small (the creation and deletion of the segment units are determined by the user), the result is that the number of segment units in each group is small, which results in that multiple TCP (Transmission Control Protocol) connections need to be established but small packets are uploaded each time, which results in wasted performance. Therefore, at the beginning, only a small number of groups need to be created, the number is between [32/20] and [32/10], namely 1 group and 3 groups, and the creation of 1 group does not help to solve the problem of network congestion, so that 3 groups or 2 groups can be created in the present embodiment, and in the present embodiment, two groups are created, the group IDs of the two groups are respectively 16 and 32, and the two group IDs are put into the red-black tree and managed by the red-black tree. Since the subsequent split or merge is performed according to the number of the segment units contained in the group, in the present embodiment, regardless of the number of the segment units, only two groups are always created at the beginning, and the group IDs of the two groups created at the beginning are M/2 and M (M is an even number), or (M + 1)/2 and M (M is an odd number), respectively; of course, the group IDs of the two groups created at the beginning may be set to 1 and M/2 (M is an even number), or 1 and (M + 1)/2 (M is an odd number), respectively, and the specific set of the group ID determines whether to search the segment unit in the red and black tree for the subsequent segment unit, and determines whether to delete the group with the smaller group ID or the group with the larger group ID when the combination of the groups occurs.
2. Searching the group of segment units
Each data segment unit has a unique immutable data segment unit ID that is generated at creation time and is continuously self-incrementing, so that the group ID in which the data segment unit is located can be determined using the data segment unit ID. The method specifically comprises the following steps: X/M, resulting in a remainder Y, where X is the ID of the segment unit, then, the group ID equal to or larger than this remainder Y is searched in the red-black tree, if the group ID is selected to be set to 1 and M/2 (M is an even number) respectively when grouping, or 1 and (M + 1)/2 (M is an odd number), it is necessary to find a group ID equal to or less than this remainder in the red-black tree, the time complexity of the search in the red and black tree is log (n), n is the number of groups into which the segment unit is divided, several times of time can be saved compared to the non-grouped search in the prior art, and in particular, if 4 segment units are managed within one group, the temporal complexity of normal non-grouping is log (4 n), the time complexity of the grouping is log (n), and the found group ID is the group to be searched.
For example, if the user newly creates a segment unit with segment unit ID of 90 and the remainder of 90 ÷ 32 is 26, when looking up, we find a group ID greater than or equal to 26, and therefore manage the segment unit in the group with group ID of 32.
3. Splitting of groups
The embodiment is described by taking two existing groups as an example, where the group IDs are respectively 16 and 32, the group ID 32 manages segment units with segment unit IDs respectively 17, 30, 50, and 59, the user newly creates a segment unit with segment unit ID 60, and should place the segment unit with segment unit ID 60 in the group with group ID 32 for management, but at this time, the group with group ID 32 already contains 4 segment units, if a total amount of metadata possibly greater than the upper limit of the number of segment units that can be contained in the group is added, specifically, if the total amount of metadata in the segment units contained in one group is greater than C, then C is the size of the largest packet to be reported to the information management service node, then the segment unit in the group can be considered as containing the upper limit of the segment unit, the value of C mainly depends on the network environment, the network bandwidth, etc., for example, if it is a gigabit network, the value of C is set to about 5M; if the network is a ten-trillion network, setting the value of C to be about 16M; the value of C is set to balance between alleviating network congestion and fully utilizing network bandwidth, and thus can be set by the user on his or her own experience. If the data segment unit contained within a group has reached an upper limit, the group needs to be split. As shown in fig. 1, the splitting manner is specifically as follows:
finding the maximum group ID smaller than the group ID, and taking a median of the two group IDs, and taking the median as the newly reconstructed group ID, in this embodiment, the maximum group ID smaller than the group ID is 16, and we take a median of 24 for 16 and 32, so we need to newly create a group with a group ID of 24.
All data segment units contained within the group are reallocated. In this embodiment, the group with the group ID of 32 contains data segment units with ID of 17, 30, 50, 59, 60; the data segment units newly calculated to have segment unit IDs of 17 and 50 should be moved to the group having group ID of 24, and the data segment units having segment unit IDs of 30, 59, 60 are still stored in the group having group ID of 32.
4. Merging of groups
As shown in fig. 2, when the data segment unit in the group is deleted, which results in that the number of the data segment unit segment units in the group is reduced to a certain number, the group and the adjacent group can be merged to achieve the purpose of reducing the number of reporting times.
Taking three existing groups as an example, a group ID is 16, 24, 32, respectively, a group with ID 16 manages one data segment unit, a group with ID 24 manages 2 data segment units, and a group with ID 32 manages 3 data segment units, when the data segment unit in the group with ID 24 is deleted, only one data segment unit remains in the group, and a group ID in the adjacent group is 16 containing one data segment unit and ID 32 containing 3 data segment unit segment units, the groups with ID 16 and 32 are merged, if merging is to be performed, the size of the total metadata of the data segment units contained in the group is to be seen specifically, in specific implementation, the size of the total metadata of the data segment units contained in the group can be set to be a K group size, which is required for merging, and then deleting the group with small ID, wherein the threshold value K can be set to C/20-C/5, and C is the size of the maximum data packet reported to the information management service node. Specifically, in this embodiment, it is necessary to delete the group with the group ID of 16, and move the remaining data segment unit in the group with the group ID of 16 to the group with the group ID of 24 for management. Only two groups with group IDs of 24 and 32 exist in the merged service. It should be noted that when a group needs to be deleted because of the occurrence of group merging, a group with a small ID may be deleted, or a group with a large ID may also be deleted, and the specific deletion depends on whether an upward search or a downward search is performed when the search is performed in the red and black tree, and more specifically, depends on whether a smaller ID or a larger ID is used for the ID used for grouping. Explained in connection with the embodiment, that is, if the IDs of the two groups at the time of initial grouping are 16 and 32, respectively, then the upward search is performed in the red-black tree, and the group with the smaller ID is deleted at the time of group merging; if the IDs of the two groups are 1 and 16, respectively, at the time of initial grouping, then a downward lookup is performed in the red-black tree, deleting the group with the larger ID at the time of group merging.
Next, the step of managing the metadata information of the segment unit is described, where the metadata of the segment unit is shown in fig. 3, where segment unit metadata represents metadata of a segment unit, the lower half parameter in fig. 3 represents a state of the metadata, used Space represents used Space, status represents state, and enum represents enumeration, and the state not shown in fig. 3 includes: (1) during creating, (2) created creating is completed, (3) deletion deleting is completed, (4) deleted deletion is completed, (5) broken damage is mainly irreparable abnormality caused by disk abnormality or other abnormality; snapshot capacity represents the capacity of a snapshot, is RollBack represents whether the current snapshot is rolling back, booten represents that the current snapshot is rolling back, a disk in the rolling back does not allow new writing, and a hash value is generated by carrying out hash transformation on the value of segment unit metadata of a data segment unit and is stored in a memory.
When the reporting module reports the segment unit segment units by taking a group as a unit, the metadata of each segment unit in the current group needs to be subjected to hash transformation to generate a hash value, the newly generated hash value is compared with the hash value reported last time (if the hash value is created for the first time, the last time is-1), if the values of the two times are equal, only the ID of the segment unit is reported, if the values of the two times are not equal, all metadata reported this time need to be reported to the information management service, and the newly generated hash value is stored in the memory.
It should be noted that, because a group includes a plurality of data segment units, a plurality of hash values are generated when hash conversion is performed, each data segment unit has an ID and a hash value after conversion, and when comparison is performed, if two hash values before and after the data segment unit with the same ID are different, metadata of the data segment unit needs to be reported; otherwise, only the ID of the segment unit needs to be reported, so as to save resources. Because the group of the segment unit segment units will be split and merged, and the number of the segment units contained in one group will not be consistent when the two hash transformations are performed, for example, when the last hash transformation is performed, there are 3 segment units in one group, and there are 4 segment units in the group, after the hash transformation is performed, there will be a case that the ID of one segment unit is only the hash value obtained after the hash transformation, but there is no hash value of the last hash transformation corresponding to the ID of the segment unit, and it is considered that the metadata of the segment unit needs to be reported, except that the hash values of the metadata in the two uploaded segment units are not the same because the number of the segment units in the group generated by the split and merge of the group is not consistent, there are also situations such as adding or deleting data segment unit in the group. If the metadata of the current group is reported for the first time, the hash value obtained by performing hash transformation on the metadata in the data segment unit is compared with a value G, wherein the value G is a non-positive integer, that is, the value G includes 0 and all negative integers, and in the specific implementation, a value can be selected, and in the embodiment, the value G is-1.
The invention also provides a distributed block storage system for realizing the data segment unit transmission differentiation management method.
The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications or additions may be made to the described embodiments or alternatives may be employed by those skilled in the art without departing from the spirit or ambit of the invention as defined in the appended claims.
Claims (8)
1. A data segment unit transmission differentiation management method is characterized in that: the method comprises the following steps:
s1, grouping a plurality of data segment units, wherein the number of the grouped units is between [ M/20] and [ M/10], and is an integer, and M is the maximum value of the number of the data segment units which can be managed in one group;
s2, grouping a data segment unit into its group, wherein the step S2 specifically includes: searching a group ID related to Y, wherein Y is a remainder obtained after X/M, X is the ID of a data segment unit needing to be put into a group, and then putting the data segment unit into the searched group for management;
s3, if the total amount of metadata in the group to which the data segment unit belongs in the step S2 is larger than C, grouping the group again, wherein C is the size of the maximum data packet reported to the information management service node;
s4, if the metadata in the two groups are respectively larger than C/20 and smaller than C/5, combining the two groups into one group;
and S5, reporting the data segment unit to the information management service node by taking the group as a unit.
2. The data segment unit transmission differentiation management method according to claim 1, characterized in that: when reporting the data segment unit, the method also comprises the following steps:
s21, carrying out hash transformation on the metadata in all the data segment units in the current group to obtain hash values, and comparing the hash value of each data segment unit with the hash value of the data segment unit reported last time;
s22, if the hash values of the two times are equal, only reporting the ID of the data segment unit in the current group to the information management service node;
and S23, if the hash values of the two times are not equal, reporting all metadata in the data segment units with unequal hash values to the information management service node.
3. The data segment unit transmission differentiation management method according to claim 2, characterized in that: and if the metadata of the current group is reported for the first time, comparing a hash value obtained after hash transformation is carried out on the metadata in the data segment unit with G, wherein G is a non-positive integer.
4. The data segment unit transmission differentiation management method according to claim 3, characterized in that: the search is performed in a red and black tree.
5. The data segment unit transmission differentiation management method according to any of claims 1-4, characterized in that: in step S1, the data segment units are divided into two groups.
6. The data segment unit transmission differentiation management method according to claim 5, characterized in that: the group IDs of the two groups into which the data segment unit is divided are M/2 and M respectively, wherein M is an even number; or (M + 1)/2 and M, wherein M is an odd number;
or the group IDs of the two groups are 1 and M/2 respectively, wherein M is an even number; or 1 and (M + 1)/2, wherein M is an odd number.
7. The data segment unit transmission differentiation management method according to claim 6, characterized in that: if the group IDs of the two groups into which the data segment unit is divided are M/2 and M respectively, wherein M is an even number, or (M + 1)/2 and M, wherein M is an odd number, searching the group with the group ID larger than or equal to Y in the red and black tree, and then putting the data segment unit into the searched group;
alternatively, the first and second electrodes may be,
if the group IDs of the two groups into which the data segment unit is divided are 1 and M/2 respectively, wherein M is an even number, or 1 and (M + 1)/2, wherein M is an odd number, searching the group with the group ID less than or equal to Y, and then putting the data segment unit into the searched group, wherein Y is the remainder obtained after X/M, and X is the ID of the data segment unit needing to be put into the group.
8. A distributed block storage system, characterized by: the method for implementing the data segment unit transmission differentiation management of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010603735.9A CN111506271B (en) | 2020-06-29 | 2020-06-29 | Data segment unit transmission differentiation management method and distributed block storage system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010603735.9A CN111506271B (en) | 2020-06-29 | 2020-06-29 | Data segment unit transmission differentiation management method and distributed block storage system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111506271A CN111506271A (en) | 2020-08-07 |
| CN111506271B true CN111506271B (en) | 2020-10-09 |
Family
ID=71878818
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010603735.9A Active CN111506271B (en) | 2020-06-29 | 2020-06-29 | Data segment unit transmission differentiation management method and distributed block storage system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111506271B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112260874A (en) * | 2020-10-23 | 2021-01-22 | 南京鹏云网络科技有限公司 | Management system and method based on distributed storage unit |
| CN112256202B (en) * | 2020-10-23 | 2023-09-26 | 南京鹏云网络科技有限公司 | Distributed storage system and method for deleting volumes in distributed storage system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190155523A1 (en) * | 2016-10-28 | 2019-05-23 | Pure Storage, Inc. | Efficient Volume Replication In A Storage System |
| CN110263038A (en) * | 2019-06-11 | 2019-09-20 | 中国人民大学 | A kind of Hash multi-table join implementation method based on grouping vector |
| EP3647928A1 (en) * | 2018-11-02 | 2020-05-06 | Wipro Limited | A method and system for transmitting data using small computer system interface (scsi) |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103914522A (en) * | 2014-03-20 | 2014-07-09 | 电子科技大学 | Data block merging method applied to deleting duplicated data in cloud storage |
| CN105204781B (en) * | 2015-09-28 | 2019-04-12 | 华为技术有限公司 | Compression method, device and equipment |
| CN108124112B (en) * | 2017-12-28 | 2020-04-03 | 福州瑞芯微电子股份有限公司 | Performance improving method for multi-path storage stream media data |
-
2020
- 2020-06-29 CN CN202010603735.9A patent/CN111506271B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190155523A1 (en) * | 2016-10-28 | 2019-05-23 | Pure Storage, Inc. | Efficient Volume Replication In A Storage System |
| EP3647928A1 (en) * | 2018-11-02 | 2020-05-06 | Wipro Limited | A method and system for transmitting data using small computer system interface (scsi) |
| CN110263038A (en) * | 2019-06-11 | 2019-09-20 | 中国人民大学 | A kind of Hash multi-table join implementation method based on grouping vector |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111506271A (en) | 2020-08-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4896593B2 (en) | Performance monitoring method, computer and computer system | |
| CN111885040A (en) | Distributed network situation perception method, system, server and node equipment | |
| CN111506271B (en) | Data segment unit transmission differentiation management method and distributed block storage system | |
| WO2016149552A1 (en) | Compaction policy | |
| CN107391770B (en) | Method, device and equipment for processing data and storage medium | |
| CN107103068A (en) | The update method and device of service buffer | |
| US11816029B2 (en) | Adjustment of garbage collection parameters in a storage system | |
| CN105512129A (en) | Method and device for mass data retrieval, and method and device for storing mass data | |
| CN106407224A (en) | Method and device for file compaction in KV (Key-Value)-Store system | |
| CN107844402A (en) | A kind of resource monitoring method, device and terminal based on super fusion storage system | |
| CN110389967A (en) | Date storage method, device, server and storage medium | |
| CN112286903A (en) | Containerization-based relational database optimization method and device | |
| CN108123840A (en) | Log processing method and system | |
| CN110955704A (en) | Data management method, device, equipment and storage medium | |
| CN112764993A (en) | Node information collection method, device, equipment and readable storage medium | |
| WO2021258360A1 (en) | On-board data storage method and system | |
| CN110196787B (en) | Data backup and recovery system and data backup and recovery method thereof | |
| DE112020005227T5 (en) | MEMORY HEALTH MONITORING FOR DIFFERENT DATA RECOVERY CONFIGURATIONS | |
| CN116389156A (en) | Real-time abnormality detection system in large-data-volume environment | |
| CN115993932A (en) | Data processing method, device, storage medium and electronic equipment | |
| US20180246673A1 (en) | Method and storage system for storing a multiplicity of data units | |
| US20230177043A1 (en) | Generation of a predictive model for selection of batch sizes in performing data format conversion | |
| CN115525603A (en) | Storage statistics method and device, computer readable storage medium and AI device | |
| CN114153695A (en) | Loose coupling and high-expansibility non-buried point data acquisition method based on Android | |
| CN113656367A (en) | BMC log collection method and system |
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 |