CN111736764B - Storage system of database all-in-one machine and data request processing method and device - Google Patents
Storage system of database all-in-one machine and data request processing method and device Download PDFInfo
- Publication number
- CN111736764B CN111736764B CN202010466211.XA CN202010466211A CN111736764B CN 111736764 B CN111736764 B CN 111736764B CN 202010466211 A CN202010466211 A CN 202010466211A CN 111736764 B CN111736764 B CN 111736764B
- Authority
- CN
- China
- Prior art keywords
- pmem
- storage
- database
- data
- machine
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 12
- 238000013461 design Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims description 26
- 238000000034 method Methods 0.000 claims description 13
- 238000013507 mapping Methods 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 230000002085 persistent effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000002955 isolation Methods 0.000 description 2
- 101000771640 Homo sapiens WD repeat and coiled-coil-containing protein Proteins 0.000 description 1
- 102100029476 WD repeat and coiled-coil-containing protein Human genes 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
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
- G06F3/061—Improving I/O performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2228—Indexing structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2291—User-Defined Types; Storage management thereof
-
- 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
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
- G06F3/0679—Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]
Abstract
The invention provides a storage system of a database all-in-one machine, and a data request processing method and device. And configuring a PMEM for a storage node of the database, wherein the PMEM and the DRAM meet the requirement of consistency design, and the storage node is also configured with a flash disk. According to the invention, the PMEM is arranged on the storage node, and for the data with high frequency access, the data on the PMEM can be directly read by using a memory access mode instead of being loaded from a traditional flash memory disk through a file system and a storage stack, so that the delay of data access of the storage node of the all-in-one machine can be greatly reduced, the execution efficiency of the storage node is improved, and the product competitiveness is improved.
Description
Technical Field
The invention relates to the technical field of database storage, in particular to a storage system of a database all-in-one machine and a data request processing method and device.
Background
Generally, a high-performance database all-in-one machine is mainly oriented to enterprise-level key core services and used for achieving better performance improvement of a client system. Therefore, the basic requirements for designing the database all-in-one machine are high reliability, high performance and high expansibility. The K-DB integrated machine for the wave tide is an integrated solution which is developed and designed based on an open X86 fusion platform and wave database software K-DB and has the advantages of high performance, high expansion, high reliability and the like.
The K-DB database all-in-one machine is mainly composed of a computing node, a storage node and a high-speed network 3. The computing nodes are also called database nodes and are mainly used for processing database access requests of applications; the storage node generally consists of a plurality of nodes, each storage node is configured with a flash disk with high capacity or high performance, and a database instance is deployed at the same time and is mainly used for data management, access and processing; high-speed networks, generally chosen to be infiniband network switches, are used to accelerate data interactions between compute nodes and storage nodes.
KAS (K-DB Automated Storage Management), namely the automatic Storage Management of K-DB, is a Storage Volume Manager component of the K-DB of the Langchao database, is used for replacing the LVM (Logical Volume Manager) of the traditional linux operating system, can automatically manage disk groups used by the database, and provides data redundancy and performance optimization. The KAS combines similar storage technology characteristics of a traditional RAID (Redundant array of Independent Disks) and a LVM (Logical Volume Manager), and has functions of Mirroring (Mirroring) and Striping (Striping) of a disk. As shown in fig. 1, KAS may implement mutual mirroring of data files on disks in different failure domains by creating a single disk group (diskgroup) based on one or more independent disks and by specifying different failure domains (failgroups) on the disk group, thereby implementing data reliability. Therefore, the I/O balanced distribution of a single disk group is realized, and the parallel processing performance of I/O is improved.
At present, a high-performance database all-in-one machine, such as a wave database all-in-one machine, which is common in the industry, mainly adopts a full flash memory device, that is, all hard disks of all storage nodes are high-speed flash memory devices. Each storage node is independent of each other, and each storage node is set to be a failgroup through the setting of a storage management function-fault group (failgroup) of the database, so that the storage nodes are mutually mirrored (mirror), and the service access of the whole all-in-one machine is not influenced when any one of the storage nodes fails, and therefore the high performance and the high reliability of the all-in-one machine are achieved. Based on the existing storage design scheme of the existing database all-in-one machine, all application data access processes need to complete interaction among nodes from a disk, through a file system, then to a system memory, then through a network and the like. Such a policy limits the read-write speed of data, and the read-write performance of the flash Memory device has a large difference from the read-write speed of a CPU or a DRAM (Dynamic Random Access Memory), and the performance characteristics of the flash Memory device and the data Access manner become the performance bottleneck of the database all-in-one machine.
Disclosure of Invention
The invention provides a storage system of a database all-in-one machine, and a data request processing method and device, which are used for solving the problem that the writing speed of the existing storage system of the database all-in-one machine cannot meet the requirement.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides a storage system of a database all-in-one machine, which is characterized in that a PMEM is configured for a storage node of a database, the PMEM and a DRAM meet the requirement of consistency design, and the storage node is also configured with a flash disk.
Further, PMEM in each storage node is configured as an interleaved APP Direct Mode, and PMEM under the same physical CPU constitutes a storage area.
Further, PMEM positioned on one side of the CPUx in all the storage nodes takes the storage nodes as fault domains to create a double-copy disk group; and x in the CPUx represents a corresponding CPU in the storage node.
Further, the FLASH memory disk in the storage node takes the storage node as a fault domain, and adopts a double copy mirror mode to create a disk group FLASH _ DATA +.
The second aspect of the present invention provides a data request processing method for a database all-in-one machine, where the method includes the following steps:
counting the physical reading and writing frequency of data loaded from a flash disk to a DRAM in a service period to obtain IO time consumption;
if the IO time consumption exceeds a set threshold, moving the table data to a table space in the PMEM storage area, and reconstructing index information of the table;
and when the data requested to be accessed by the client side is in the PMEM storage area, bypassing the DRAM cache, and returning the result to the client side by the PMEM in a memory mapping mode.
Further, the method further comprises:
and monitoring data in the PMEM storage area, and mapping the data with the updating frequency exceeding a threshold value into the DRAM.
Further, before the step, the method further comprises:
and configuring PMEM and a flash disk for the database storage node, wherein the configured PMEM and the DRAM meet the requirement of consistent design.
Furthermore, PMEM in each storage node is configured as an APP Direct Mode of an interleaving type, and PMEM under the same physical CPU forms a storage area; the PMEM positioned on one side of the CPUx in all the storage nodes takes the storage nodes as fault domains to create a double-copy disk group; and x in the CPUx represents a corresponding CPU in the storage node.
A third aspect of the present invention provides a data request processing apparatus for a database all-in-one machine, where the apparatus includes:
the counting unit is used for counting the physical reading and writing frequency of data loaded from the flash disk to the DRAM in a service period to obtain the IO time consumption condition;
the data processing unit is used for moving the table data to the table space in the PMEM storage area and reconstructing the index information of the table if the IO time consumption exceeds a set threshold;
and the request processing unit bypasses the DRAM cache when the data requested to be accessed by the client is in the PMEM storage area, and returns the result to the client by the PMEM in a memory mapping mode.
The fourth aspect of the present invention provides a computer storage medium, in which computer instructions are stored, and the computer instructions are characterized in that when running on a data request processing device of a database all-in-one machine, the computer instructions cause the data request processing device of the database all-in-one machine to execute a data request processing method of the database all-in-one machine.
The effect provided in the summary of the invention is only the effect of the embodiment, not all the effects of the invention, and one of the above technical solutions has the following advantages or beneficial effects:
1. according to the invention, the PMEM is arranged on the storage node, and for the data with high frequency access, the data on the PMEM can be directly read by using a memory access mode instead of being loaded from a traditional flash memory disk through a file system and a storage stack, so that the delay of data access of the storage node of the all-in-one machine can be greatly reduced, the execution efficiency of the storage node is improved, and the product competitiveness is improved.
2. The direct access mode is applied to the application mode of the PMEM, so that the maximum high-speed storage space of 6TB can be increased for a single storage node of the all-in-one machine, and meanwhile, different PMEM storage areas are created by taking the physical CPU as isolation, so that when one or more PMEMs are in fault, only PMEM equipment on one side of the same physical CPU is influenced, other PMEMs on the whole node are not influenced, and the high reliability of the storage node is guaranteed to the maximum extent.
Drawings
In order to more clearly illustrate the embodiments or technical solutions in the prior art of the present invention, the drawings used in the description of the embodiments or prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained based on these drawings without creative efforts.
FIG. 1 is a schematic diagram of the structure relationship between a disk group and a fault domain in the prior art;
FIG. 2 is a schematic diagram of logical topology after the system of the present invention configures PMEM;
FIG. 3 is a schematic diagram of the structure of a storage area of the system according to the present invention;
FIG. 4 is a schematic diagram of the topological relationship after the system of the present invention creates a double copy disk group;
FIG. 5 is a schematic diagram of a data request processing path according to the method of the present invention;
FIG. 6 is a schematic flow diagram of the method of the present invention.
Detailed Description
In order to clearly explain the technical features of the present invention, the present invention will be explained in detail by the following embodiments and the accompanying drawings. The following disclosure provides many different embodiments, or examples, for implementing different features of the invention. To simplify the disclosure of the present invention, the components and arrangements of specific examples are described below. Moreover, the present disclosure may repeat reference numerals and/or letters in the various examples. This repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed. It should be noted that the components illustrated in the figures are not necessarily drawn to scale. Descriptions of well-known components and processing techniques and processes are omitted so as to not unnecessarily limit the invention.
Non-volatile Memory (NVM), also known as Persistent Memory (PMEM) or Memory-level Memory (SCM), is between the Memory and storage hierarchy and can provide larger capacity than DRAM and faster speed than Memory. With the persistent memory, a new hierarchy is applied to place data, so that the persistent memory can be accessed like the conventional memory, and data blocks do not need to be switched back and forth between the memory and the storage, thereby filling the performance/capacity difference of the existing hierarchy. The invention improves the existing database all-in-one machine based on PMEM.
As shown in fig. 2, based on the requirement of coherent design of PMEM and DRAM memory, 12 pieces of PMEM hardware devices are fully inserted into each storage node of the existing database all-in-one machine (when a PMEM device is fully inserted with 12 pieces, 12 memory banks are also necessarily inserted), and then the total capacity of the PMEM device is total _ command _ of _ PMEM — 12 total _ command _ per _ PMEM. The capacity of each PMEM hardware device can be selected from 3 specifications such as 128GB, 256GB, 512GB and the like, namely the maximum storage capacity which can be increased by each storage node is 6 TB. Meanwhile, in consideration of cost, the flash memory disks with different capacities can be collocated according to the requirement of business data volume, and the total capacity of the corresponding PMEM equipment is planned and collocated according to the total capacity of the flash memory disks, and the total capacity of the PMEM equipment is preferably 1/3-1/10 of the total capacity (total _ capacity _ of _ flash) of the flash memory equipment.
Considering that the more PMEM devices the better the balance performance, the performance of PMEM devices is about 3-10 times that of a general flash memory device under the same capacity requirement. Therefore, when planning configuration, all available slots are preferably selected to be fully inserted, and then 1/3-1/10 with the capacity close to that of a flash memory is selected, so that the method can be flexibly applied to most of customer service scene requirements.
As shown in fig. 3, at each storage node operating system level, PMEM devices are configured as interleaved type APP Direct Mode. Meanwhile, different memory areas pmem0 and pmem1 are created according to different physical CPUs, and fig. 3 illustrates a memory node as a dual-channel CPU.
The staggered App Direct Mode configuration is to uniformly manage PMEM equipment on one side of the same CPU and divide the PMEM equipment into a large storage space.
When the PMEM device is configured as the interleaved APP Direct Mode, any one or more of the 6 PMEM devices on the same physical CPU side may cause the memory area to be rebuilt.
As shown in fig. 4, in this embodiment, the nonvolatile memory in all storage nodes is located in the PMEM storage area on the CPU0 side, and a double-copy disk group disk _ PMEM0 is created with the storage node as a failure domain; the nonvolatile memory in all storage nodes is located in the PMEM storage area on the side of the CPU1, and a double-copy disk group disk _ PMEM1 is created with the storage node as a failure domain. Thus, the PMEM0 from different storage nodes are mirror images of each other, and the PMEM1 from different storage nodes are mirror images of each other, which can ensure that when the PMEM memory region on the CPU0 side fails, corresponding copy data still exists on the PMEM memory region on the CPU0 side of another node, and similarly, the PMEM memory region on the CPU1 side. And the failure of the whole PMEM storage area has no influence. The normal operation of a single storage node is not influenced when any non-volatile memory hardware in any storage node fails.
The invention combines the characteristic of PMEM equipment failure, creates a disk group taking different CPUs and different storage nodes as isolation, and maximally ensures the high reliability of the storage nodes.
In order to further guarantee the high reliability of the storage nodes, the FLASH memory disks of the storage nodes use the KAS management function of the wave DATA software, the storage nodes are used as fault domains, a double-copy mirror image mode is selected, and the FLASH _ DATA + of the disk group is created and used for storing the service DATA of the database, so that when a single storage node fails, the continuous operation of the whole storage of the all-in-one machine is not influenced.
As shown in fig. 5, in the database instance portion of the storage node, an interface of the mmap (memory mapped) access mode of the database storage instance to the nonvolatile memory is opened, so that after the PMEM device is formatted and mounted, the database process can normally read and write access to the storage area of the PMEM.
When detecting that the data to be accessed is cached in the DRAM, directly processing and updating and returning the result to the client; and when the data is detected to be in the flash memory area, processing and updating are carried out in the DRAM by loading the data from the flash memory area into the DRAM, and the result is returned to the client. Meanwhile, the K-DB database can automatically count the physical read-write frequency of data loaded from a flash to a DRAM in a business low-peak period, and the IO time consumption condition is obtained. If IO time consumption exceeds 60% of total time consumption of certain business SQL processing, automatically moving table data to a table space in a PMEM storage area, and reconstructing index information of the table. When the next request detects that the data to be accessed by the client read-write request is in the PMEM storage area, the data can directly bypass the DRAM cache and directly return the client result in a memory Mapped mode, meanwhile, tracking monitoring is automatically started on the data stored in the PMEM storage area, and the data updated by the data frequently read from the PMEM equipment is Mapped and updated into the DRAM, so that the higher access speed is achieved.
The interface and the cache updating mode of the MMAP (memory mapped) access mode of the storage instance to the nonvolatile memory are stored in the storage node open database, so that the database all-in-one machine is convenient to use the memory access mode to directly read the data on the nonvolatile memory for the data with high frequency access between the inside of the storage node and the nodes, and the data is not loaded from the traditional flash memory disk through a file system and a storage stack, thereby greatly reducing the delay of the data access of the storage node of the all-in-one machine and improving the execution efficiency of the storage node.
Establishing a rapid table space tablespace _ Flash at a database node (or a computing node) based on a Flash disk group, wherein the rapid table space tablespace _ Flash is used for establishing a data table and storing actual service data; a super speed table space and corresponding data files tablespace _ PMEM0 and tablespace _ PMEM1 are created based on a PMEM disk device and used for storing key tables (hot data) which are periodically moved from a flash table space in a move table mode and other data frequently accessed by a core. By utilizing the functional characteristics that the KDB can transparently move the table data in different table spaces, the original database command can be used for automatically establishing the hot point database and maintaining the hot point table storage information by only establishing different table spaces without considering whether the equipment formats of the PMEM and the Flash are compatible, whether the data organization and management modes are consistent and the like.
As shown in fig. 6, based on the storage system, the data request processing method of the database all-in-one machine of the present invention includes the following steps:
s1, counting the physical read-write frequency of data loaded from a flash disk to a DRAM in a service period to obtain IO time consumption; one on-site rush hour period is typically counted.
S2, if the IO consumed time exceeds the set threshold, moving the table data to the table space in the PMEM storage area, and reconstructing the index information of the table; wherein the threshold is set to be that IO time consumption exceeds 60% of the total time consumption of certain business SQL processing.
S3, when the data requested to be accessed by the client is in the PMEM storage area, bypassing the DRAM cache, and returning the result to the client by the PMEM in a memory mapping mode.
Meanwhile, data in the PMEM storage area are monitored, and data with the updating frequency exceeding a threshold value are mapped into the DRAM. The threshold of the update frequency is set according to actual requirements.
Before executing the above steps S1-S3, the method further includes: and configuring a PMEM and a flash disk for the database storage node, wherein the configured PMEM and the DRAM meet the requirement of a consistent design. PMEM in each storage node is configured into an APP Direct Mode of a staggered type, and PMEM under the same physical CPU forms a storage area; the PMEM positioned on one side of the CPUx in all the storage nodes takes the storage nodes as fault domains to create a double-copy disk group; and x in the CPUx represents a corresponding CPU in the storage node.
The invention also provides a data request processing device of the database all-in-one machine, which corresponds to the data request processing method and comprises a statistical unit, a data processing unit and a request processing unit.
A statistical unit counts the physical reading and writing frequency of data loaded from a flash disk to a DRAM in a service period to obtain IO time consumption; if the IO consumed time exceeds a set threshold value, the data processing unit moves the table data to a table space in the PMEM storage area and reconstructs the index information of the table; and when the data requested to be accessed by the client is in the PMEM storage area, the request processing unit bypasses the DRAM cache and returns the result to the client by the PMEM in a memory mapping mode.
The computer storage medium of the invention stores computer instructions, and when the computer instructions run on the data request processing device of the database all-in-one machine, the data request processing device of the database all-in-one machine executes the data request processing method of the database all-in-one machine.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive changes in the technical solutions of the present invention.
Claims (6)
1. A storage system of a database all-in-one machine is characterized in that a PMEM is configured for a storage node of a database, the PMEM and a DRAM meet the requirement of consistency design, and the storage node is further configured with a flash disk; the consistency design requirement is that the number of the PMEM hardware equipment in each storage node of the database all-in-one machine is the same as that of the DRAMs;
PMEM in each storage node is configured into an APP Direct Mode of a staggered type, and PMEM under the same physical CPU forms a storage area;
the PMEM positioned on one side of the CPUx in all the storage nodes takes the storage nodes as fault domains, and a double-copy disk group is created; and x in the CPUx represents a corresponding CPU in the storage node.
2. The storage system of the database all-in-one machine as claimed in claim 1, wherein FLASH memory disks in the storage nodes adopt a double copy mirror mode to create a disk group FLASH _ DATA + by taking the storage nodes as fault domains.
3. A data request processing method of a database all-in-one machine is characterized by comprising the following steps:
counting the physical reading and writing frequency of data loaded from a flash disk to a DRAM in a service period to obtain IO time consumption;
if the IO consumed time exceeds a set threshold, moving table data to a table space in the PMEM storage area, and reconstructing index information of the table;
when the data requested to be accessed by the client side is in the PMEM storage area, bypassing the DRAM cache, and returning a result to the client side by the PMEM in a memory mapping mode;
before the step, the method further comprises the following steps:
configuring PMEM and a flash disk for a database storage node, wherein the configured PMEM and DRAM meet the requirement of consistency design, and the requirement of consistency design is that the number of PMEM hardware equipment in each storage node of the database all-in-one machine is the same as that of the DRAM;
PMEM in each storage node is configured into an APP Direct Mode of a staggered type, and PMEM under the same physical CPU forms a storage area; the PMEM positioned on one side of the CPUx in all the storage nodes takes the storage nodes as fault domains, and a double-copy disk group is created; and x in the CPUx represents a corresponding CPU in the storage node.
4. The database all-in-one machine as claimed in claim 3, wherein the method further comprises:
and monitoring data in the PMEM storage area, and mapping the data with the updating frequency exceeding a threshold value into the DRAM.
5. A data request processing device of a database all-in-one machine is characterized by comprising:
the counting unit is used for counting the physical reading and writing frequency of data loaded from the flash disk to the DRAM in a service period to obtain the IO time consumption condition;
the data processing unit is used for moving the table data to the table space in the PMEM storage area and reconstructing the index information of the table if the IO time consumption exceeds a set threshold;
the request processing unit bypasses a DRAM cache when the data requested to be accessed by the client side is in a PMEM storage area, and returns a result to the client side by the PMEM in a memory mapping mode;
the PMEM in each storage node is configured into an APP Direct Mode of a staggered type, and the PMEM under the same physical CPU forms a storage area; the PMEM positioned on one side of the CPUx in all the storage nodes takes the storage nodes as fault domains, and a double-copy disk group is created; and x in the CPUx represents a corresponding CPU in the storage node.
6. A computer storage medium having computer instructions stored therein, wherein the computer instructions, when executed on a data request processing device of a database all-in-one machine, cause the data request processing device of the database all-in-one machine to execute the data request processing method of the database all-in-one machine according to claim 3 or 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010466211.XA CN111736764B (en) | 2020-05-28 | 2020-05-28 | Storage system of database all-in-one machine and data request processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010466211.XA CN111736764B (en) | 2020-05-28 | 2020-05-28 | Storage system of database all-in-one machine and data request processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111736764A CN111736764A (en) | 2020-10-02 |
| CN111736764B true CN111736764B (en) | 2022-07-22 |
Family
ID=72648002
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010466211.XA Active CN111736764B (en) | 2020-05-28 | 2020-05-28 | Storage system of database all-in-one machine and data request processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111736764B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113590047B (en) * | 2021-08-11 | 2024-01-26 | 中国建设银行股份有限公司 | Database screening method and device, electronic equipment and storage medium |
| CN113590049B (en) * | 2021-09-28 | 2021-12-31 | 苏州浪潮智能科技有限公司 | Storage volume cross-node cloning method, device, equipment and readable medium |
| CN117032596B (en) * | 2023-10-09 | 2024-01-26 | 苏州元脑智能科技有限公司 | Data access method and device, storage medium and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106775947A (en) * | 2016-12-06 | 2017-05-31 | 中国电子科技集团公司第三十二研究所 | Large-scale virtual computing dynamic load balancing method based on openstack |
| CN109324933A (en) * | 2018-09-06 | 2019-02-12 | 郑州云海信息技术有限公司 | A kind of AEP memory reboot test method, device, terminal and storage medium |
| CN109460332A (en) * | 2018-11-01 | 2019-03-12 | 郑州云海信息技术有限公司 | A kind of SWRaid detection method and detection system based on AEP memory techniques |
-
2020
- 2020-05-28 CN CN202010466211.XA patent/CN111736764B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106775947A (en) * | 2016-12-06 | 2017-05-31 | 中国电子科技集团公司第三十二研究所 | Large-scale virtual computing dynamic load balancing method based on openstack |
| CN109324933A (en) * | 2018-09-06 | 2019-02-12 | 郑州云海信息技术有限公司 | A kind of AEP memory reboot test method, device, terminal and storage medium |
| CN109460332A (en) * | 2018-11-01 | 2019-03-12 | 郑州云海信息技术有限公司 | A kind of SWRaid detection method and detection system based on AEP memory techniques |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111736764A (en) | 2020-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111736764B (en) | Storage system of database all-in-one machine and data request processing method and device | |
| US10031703B1 (en) | Extent-based tiering for virtual storage using full LUNs | |
| US8966476B2 (en) | Providing object-level input/output requests between virtual machines to access a storage subsystem | |
| US9846544B1 (en) | Managing storage space in storage systems | |
| CN111587423B (en) | Hierarchical data policies for distributed storage systems | |
| US7562251B1 (en) | Flexibly combining mirroring, concatenation and striping in virtual storage devices | |
| CN100524235C (en) | Recovery operations in storage networks | |
| US20130159651A1 (en) | Storage system and storage migration method | |
| WO2015015550A1 (en) | Computer system and control method | |
| CN106066890B (en) | Distributed high-performance database all-in-one machine system | |
| CN107798130A (en) | A kind of Snapshot Method of distributed storage | |
| US11061788B2 (en) | Storage management method, electronic device, and computer program product | |
| KR20130083356A (en) | A method for metadata persistence | |
| US8140886B2 (en) | Apparatus, system, and method for virtual storage access method volume data set recovery | |
| CN114610232A (en) | Storage system, memory management method and management node | |
| US20230236761A1 (en) | Read-disturb-based logical storage read temperature information identification system | |
| WO2023035646A1 (en) | Method and apparatus for expanding memory, and related device | |
| US11809720B2 (en) | Techniques for storage management | |
| CN112379825B (en) | Distributed data storage method and device based on data feature sub-pools | |
| US11922067B2 (en) | Read-disturb-based logical storage read temperature information maintenance system | |
| US11379326B2 (en) | Data access method, apparatus and computer program product | |
| US11625183B1 (en) | Techniques for data layout on rotating disk drives | |
| US11922020B2 (en) | Read-disturb-based read temperature information persistence system | |
| CN116009761A (en) | Data writing method and related equipment | |
| US11561695B1 (en) | Using drive compression in uncompressed tier |
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 |