CN106066890A - A kind of distributed high-performance data storehouse integrated machine system - Google Patents
A kind of distributed high-performance data storehouse integrated machine system Download PDFInfo
- Publication number
- CN106066890A CN106066890A CN201610427650.3A CN201610427650A CN106066890A CN 106066890 A CN106066890 A CN 106066890A CN 201610427650 A CN201610427650 A CN 201610427650A CN 106066890 A CN106066890 A CN 106066890A
- Authority
- CN
- China
- Prior art keywords
- storage device
- data
- storage
- performance
- oracle
- 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.)
- Granted
Links
Classifications
-
- 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
- G06F16/21—Design, administration or maintenance of databases
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Memory System Of A Hierarchy Structure (AREA)
Abstract
The present invention relates to a kind of distributed high-performance data storehouse integrated machine system, including data base's all-in-one and the supplementary module being erected on data base's all-in-one, data base's all-in-one includes: subscriber's main station, is used for carrying out Business Processing；Oracle applied host machine, is used for subscriber's main station and returns result；Storage device, is used for storing subscriber's main station desired data；Data storage network, is used for connecting Oracle applied host machine and storage device；Supplementary module includes: storage device management module, for United Dispatching storage device；Distributed high-performance intelligent storage module；For with storage device management module communication, it is achieved the distributed high-performance intelligent storage of storage device；Oracle main frame adaptation module, for being mapped as standard storage hardware resource by storage device.Compared with prior art, the present invention have process data volume is big, processing speed is fast, process performance is strong and high reliability.
Description
Technical field
The present invention relates to oracle database storage management domain, particularly relate to a kind of distributed high-performance data storehouse one
Machine system.
Background technology
Oracle database is the most outstanding DMBS, current Oracle DBMS of Oracle company release and is correlated with
Product almost has application in each industrial circle of the whole world.No matter online transaction and the data warehouse of large enterprise should
With, or middle-size and small-size online transaction business, there is model's application that numerous oracle database system successfully uses.
In order to dispose oracle database, modal structure is multinode RAC (Real Application
Cluster) mode adding shared storage realizes.Fig. 1 is typical case's construction mode, is subscriber's main station 1, Oracle RAC including 1
Node 2 and shared storage device 3, there are more than ten years in such structural models in data center, mature and reliable, the most surely
Fixed.
It is typically chosen minicomputer or high-end PC server disposition is Oracle RAC Node, the data namely often said
Storehouse High Availabitity node, stores network by FC at a high speed, accesses shared storage device (Shared Storage).User (User)
Connected the RAC node of oracle database by TCP/IP network, Oracle RAC node can return user with result, when
When one of them database node occurs abnormal, do not affect user and use.
Today of new technique high speed development, along with the rapid growth of data volume, the data creating magnanimity need to process,
Data scale tens TB easily, the data of TB up to a hundred, even up to PB.The shared storage rack of the most traditional Oracle
Structure, just as there is the wait of too many storage input/output not only to produce huge waste, simultaneously because of the performance bottleneck produced, makes
Obtain this inefficient framework mode and can not meet the business demand of user.The most this system structure at storage performance, capacity
Extension and management aspect increasingly become the bottleneck of Informatization Development.
Shared storage device under current conventional architectures, even if under a large amount of disks are supported, IOPS value maintains 5～10
Ten thousand/second.Current FC networking can enough reach the interface rate of 16Gb/s, even if the storage of middle and high end, uses multiport parallel,
Its throughput also can only be to the level of 4～5GB/s, and this is for processing in the face of the data of TB rank, and the general business time will
More than 20 hours, hence it is evident that be to meet customer service demand.
So, the data base's deployment mode under tradition Oracle RAC structure, exist relatively in present big data quantity processes
For serious performance bottleneck problem.
Summary of the invention
Data volume is big, processing speed is fast, process performance is strong to it is an object of the invention to provide one to process for the problems referred to above
And the distributed high-performance data storehouse integrated machine system that reliability is high.
For realizing purpose of the present invention, the present invention provides a kind of distributed high-performance data storehouse integrated machine system, including
Data base's all-in-one and the supplementary module being erected on data base's all-in-one, described data base's all-in-one includes:
Subscriber's main station, is used for carrying out Business Processing；
Oracle applied host machine, connects subscriber's main station by TCP/IP network, for returning result to subscriber's main station；
Storage device, for storing the data needed for subscriber's main station；
Data storage network, connects Oracle applied host machine and storage device respectively, is used for making the Oracle applied host machine can
To access storage device；
Described supplementary module includes:
Storage device management module, is set up in storage device, for United Dispatching storage device, completes storage device
Fundamental management；
Distributed high-performance intelligent storage module, is set up in storage device, is used for and storage device management module communication,
Realize the distributed high-performance intelligent storage of storage device；
Oracle main frame adaptation module, is set up on Oracle applied host machine, deposits for storage device is mapped as standard
Storage hardware resource makes Oracle applied host machine access storage device.
Described Oracle applied host machine is the server of open X86-based.
Described data storage network is InfiniBand network.
Described storage device is that distributed high performance parallel stores system, comprises multiple memory node, each memory node
Including PC server local disk and the high-performance Flash SSD of respective volume.
Described storage device management module includes:
Storage scheduling unit, for being divided into different blocks, and by each block labelling by PC server local disk
For object data space, Flash SSD is labeled as spatial cache simultaneously；
Memory communicating unit, is used for and distributed high-performance intelligent storage module communication, it is achieved the high property to storage device
Can intelligent storage；
Storage driver element, for according to scheduling unit and the order of communication unit, it is achieved the read-write operation of storage device.
Described distributed high-performance intelligent storage module includes:
Communication unit, is used for and storage device management module communication, it is achieved the data access operation to storage device；
Distribution memory cells, for being evenly distributed in data among all object data space, and according to accessing frequency
Degree is called in Flash SSD, prevents node failure from causing number at 2 parts or 3 parts data trnascriptions of each memory node reservation simultaneously
According to loss；
Intelligent scheduling unit, being used for the PC server local disk label of fault is to lose efficacy and by this failed disk
Data policy redundant distributions is to remaining disk, and Uniform-distributed Data again after failed disk is replaced.
The exchange rate of the every port of switch in described InfiniBand network is 56Gb/s.
The quantity of the PC server local disk that described each memory node comprises is between 12 pieces to 24 pieces.
Compared with prior art, the method have the advantages that
(1) Oracle applied host machine uses the server of open X86-based, meet on a large scale TB DBMS amount with
On OLTP/OLAP application need.
(2) data storage network have selected that technology is novel, the network bandwidth high, network delay is little and wide in supercomputer
The InfiniBand network switching technologies of general application, and use the switch of the exchange rate of every port 56Gb/s, relatively conventional
FC network switch bandwidth promotes 3.5 times, improves the IO handling capacity of data base greatly.
(3) storage device is distributed high performance parallel memory system, relative to traditional special shared storage architecture,
IOPS and the handling capacity of storage all have increased significantly.
(4) high-performance intelligent storage module it is provided with, it is achieved that the high-performance distribution to storage data, and in storage device
Automatically according to new redundancy strategy distributed data when going wrong, it is ensured that the high efficiency of data base's all-in-one.
(5) high-performance intelligent storage module is provided with backup to data, there is not Single Point of Faliure, any one memory node
Damage do not interfere with service application, it is ensured that the reliability of data base's all-in-one.
(6), in storage device management module, introduce high-performance Sasobit and SBR D disk and high efficiency hot spot data caching technology, make
Obtain memory node and be provided simultaneously with high-performance Sasobit and SBR D characteristic and traditional magnetic disk storage characteristics.
Accompanying drawing explanation
Fig. 1 is tradition Oracle RAC configuration diagram；
Fig. 2 is that distributed high-performance stores Oracle RAC configuration diagram；
Fig. 3 is the structural representation of the present invention；
Fig. 4 is that memory node internal disk manages logical schematic；
Fig. 5 is cache management schematic diagram.
In figure: 1 is subscriber's main station, 2 is Oracle RAC node, and 3 is shared storage device, and 4 deposit for distributed high-performance
Storage system, 5 is Oracle main frame adaptation module, and 6 is distributed high-performance intelligent storage module, and 7 is storage device management module,
8 is disk management software, and 9 is memory dram, and 10 is NVDIMM, and 11 is Flash SDD, and 12 is disk HDD.
Detailed description of the invention
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.The present embodiment is with technical solution of the present invention
Premised on implement, give detailed embodiment and concrete operating process, but protection scope of the present invention be not limited to
Following embodiment.
Being illustrated in figure 1 tradition Oracle RAC framework, the present invention is to apply based on oracle database, is ensureing it
While Oracle RAC High Availabitity, by realizing a series of new architecture integrated software system on standard hardware, machine is at open mark
Standardization PC server and distributed, high performance software, set up a set of high performance database technology framework.
As shown in Fig. 2-Fig. 3, the present embodiment provides a kind of distributed high-performance data storehouse integrated machine system, its storage device
For distributed High Performance Cache and Memory System 4, its core component includes that Oracle main frame adaptation module 5, distributed high-performance intelligence are deposited
Storage module 6 and storage device management module 7, by the collaborative work having cooperated to data base's all-in-one of these assemblies
Make, promote database application performance.
In this programme, Oracle RAC node uses the server of open X86-based, meets large-scale OLTP/
OLAP applies needs, disposes based on Oracle RAC service application.Data storage network selection technique is novel, the network bandwidth is high,
Network delay is little, and wide variety of InfiniBand network switching technologies in supercomputer, this programme use every
The switch of the exchange rate of port 56Gb/s, the relatively conventional FC network switch network bandwidth promotes 3.5 times, improves greatly
The IO handling capacity of data base.
Storage system applications abandons traditional special shared storage architecture, by using open standardization PC server originally
The pattern of local disk serves as memory node, sets up a distributed high performance parallel memory system.New distribution type height
Can will have 12～24 pieces of physical disks by storage PC server node, and configure the high-performance Flash SSD of respective volume.
By the hot spot data buffer scheduling optimisation technique within memory node, promote IOPS and the handling capacity of storage.
It is illustrated in figure 3 the main logic structure of the Technical Architecture of high-performance data storehouse all-in-one.Wherein modules is complete
The major function become is as follows:
1, storage device management module
The disk of storage device management module (disk management module) United Dispatching management PC server and high speed Flash
The significant components of SSD, completes the fundamental management of disk.Local disk is divided into different blocks by this module
(Unit), each block will become the object data space (Data Unit is called for short DU) of intelligent storage software management, simultaneously
Flash SSD is labeled as spatial cache (Flash Unit), and device driver module undertakes the data write work of reality simultaneously.
Storage device management module software is responsible for and distributed high-performance intelligent storage software communication.
It is illustrated in figure 4 the Data Migration of the hybrid storage medium access that the present embodiment uses, slow in memory node configuration
Deposit the important method being to improve performance of storage system.By shared data being temporarily stored in DRAM internal memory at a high speed, magnanimity
The speed of operation sharing data is significantly improved.Although having the read or write speed of nanosecond (ns) level, storage relatively low for DRAM is held
Amount and high cost limit its ability promoting HDD read and write access for extensive caching.(flash memory is solid for Flash SSD
State hard disk) as novel storage medium, have the advantage that power consumption is little, speed is fast, capacity is big, its cost and performance between DRAM and
Between HDD, therefore can be effectively improved I/O bandwidth and IOPS as the L2 cache between DRAM and HDD.Although Flash
SSD is the most attractive as caching, but it exists bigger unsymmetry (i.e. reading far faster than writing) on read-write operation, dodges
The erasing times deposited is limited.Existing cache contents is replaced algorithm and is operated mainly in DRAM, and less consideration flash memory is as slow
The feature deposited, and the framework of multi-level buffer.Therefore, the present embodiment is based on traffic performance, co-design two-level cache management plan
Slightly.
2, distributed high-performance intelligent storage software
Distributed high-performance intelligent storage software is the core place of whole data storage procedure.Intelligent storage software is responsible for
Data are evenly distributed among all of DU by the balancing performance distribution storage of many memory nodes data, and by data
Call in Flash SSD according to visiting frequency, improve system entirety access performance.
Another big function of distributed high-performance intelligent storage software is responsible for the reliability of data, intelligent storage software root
According to service strategy, 2～3 parts of data trnascriptions can be retained between different storage node, thus ensure when a node failure, both
Any loss of data will not be caused, the most do not affect the node application of upper layer data storehouse.Equally, when single disk failures
Waiting, while failed disk can be labeled as failure state by intelligent storage software, data are the most tactful superfluous in remaining disk
Remaining distribution, to ensure the reliability of data.After failed disk is replaced, the equilibrium completing to be newly added data in magnetic disk automatically divides
Cloth, restPoses.Distributed high-performance intelligent storage software is responsible for and storage device management module and device driver module
Mutually communication, has coordinated the access of data.
Therefore, the present embodiment is based on traffic performance, co-design cache management strategy.As it is shown in figure 5, it is relatively low for DRAM
Capacity and Flash SSD write operation on not enough, introduce the super internal memory being made up of DRAM and Nonvolatile memory chip
NVDIMM, designs novel read and write abruption technology.Design novel read and write abruption method: have more for finance data business
, random small documents write operation feature, introduce super internal memory NVDIMM.Little dirty data is first write non-by the DRAM of volatibility
The NVDIMM of volatibility, when the dirty data of NVDIMM is pooled to certain threshold value again by continuous blocks batch write HDD, and randomized block is write
Enter Flash SSD.Compared with the writing direct of the present embodiment and DRAM to HDD, significantly reduce I/O operation number of times；Arrive with DRAM
The write of Flash SSD is compared, and reduces the erasing times to Flash SSD, improves data consistency.NVDIMM mainly uses
In the caching of writing of HDD, and Flash SDD is mainly used in the reading caching of HDD, it is achieved thereby that the read-write operation gone up largely divides
From.
3, Oracle main frame internet adapter software
Oracle database can not the distributed High Performance Cache and Memory System in Direct Recognition rear end, so need Oracle should
With main frame is disposed Oracle main frame internet adapter software.By Oracle main frame internet adapter software, distributed high-performance storage is mapped as
Standard storage hardware resource, i.e. maps the tradition discernible standard disk of Oracle ASM.Make Oracle applied host machine the most not
Need to do any change, directly access storage hardware resource by this software module.Oracle main frame internet adapter software module is responsible for
With distributed high-performance intelligent storage software communication.
Adaptive by above-mentioned storage device management module, distributed high-performance intelligent storage software module and Oracle main frame
The organic assembling of software module, based on the open-standards PC server, with high efficiency and the reliability of data storage
Distributed high-performance intelligent storage software is core, in conjunction with Flash SSD at a high speed, at a high speed, the InfiniBand of low latency hands over
Change planes.By software and hardware consolidation and innovation, remote ultra-traditional main frame and the performance of memory system data library structure, streamlining management are reached
Burden.
Claims (8)
1. a distributed high-performance data storehouse integrated machine system, including data base's all-in-one and being erected on data base's all-in-one
Supplementary module, it is characterised in that described data base's all-in-one includes:
Subscriber's main station, is used for carrying out Business Processing；
Oracle applied host machine, connects subscriber's main station by TCP/IP network, for returning result to subscriber's main station；
Storage device, for storing the data needed for subscriber's main station；
Data storage network, connects Oracle applied host machine and storage device respectively, is used for making Oracle applied host machine to visit
Ask storage device；
Described supplementary module includes:
Storage device management module, is set up in storage device, for United Dispatching storage device, completes the basis of storage device
Management work；
Distributed high-performance intelligent storage module, is set up in storage device, is used for and storage device management module communication, it is achieved
The distributed high-performance intelligent storage of storage device；
Oracle main frame adaptation module, is set up on Oracle applied host machine, hard for storage device is mapped as standard storage
Part resource makes Oracle applied host machine access storage device.
Distributed high-performance data storehouse the most according to claim 1 integrated machine system, it is characterised in that described Oracle should
With the server that main frame is open X86-based.
Distributed high-performance data storehouse the most according to claim 1 integrated machine system, it is characterised in that described data store
Network is InfiniBand network.
Distributed high-performance data storehouse the most according to claim 1 integrated machine system, it is characterised in that described storage device
Storing system for distributed high performance parallel, comprise multiple memory node, each memory node includes PC server local disk
High-performance Flash SSD with respective volume.
Distributed high-performance data storehouse the most according to claim 4 integrated machine system, it is characterised in that described storage device
Management module includes:
Storage scheduling unit, for PC server local disk being divided into different blocks, and each block is labeled as right
Image data space, is labeled as spatial cache by Flash SSD simultaneously；
Memory communicating unit, is used for and distributed high-performance intelligent storage module communication, it is achieved the high-performance intelligence to storage device
Can storage；
Storage driver element, for according to scheduling unit and the order of communication unit, it is achieved the read-write operation of storage device.
Distributed high-performance data storehouse the most according to claim 5 integrated machine system, it is characterised in that described distributed height
Performance intelligent storage module includes:
Communication unit, is used for and storage device management module communication, it is achieved the data access operation to storage device；
Distribution memory cells, for data being evenly distributed among all object data space, and adjusts according to visiting frequency
Enter in Flash SSD, prevent node failure from causing data to be lost in each memory node reservation 2 parts or 3 parts of data trnascriptions simultaneously
Lose；
Intelligent scheduling unit, being used for the PC server local disk label of fault is to lose efficacy and by the data in this failed disk
Strategy redundant distributions is to remaining disk, and Uniform-distributed Data again after failed disk is replaced.
Distributed high-performance data storehouse the most according to claim 3 integrated machine system, it is characterised in that described
The exchange rate of the every port of switch in InfiniBand network is 56Gb/s.
Distributed high-performance data storehouse the most according to claim 4 integrated machine system, it is characterised in that described each storage
The quantity of the PC server local disk that node comprises is between 12 pieces to 24 pieces.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610427650.3A CN106066890B (en) | 2016-06-16 | 2016-06-16 | Distributed high-performance database all-in-one machine system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610427650.3A CN106066890B (en) | 2016-06-16 | 2016-06-16 | Distributed high-performance database all-in-one machine system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106066890A true CN106066890A (en) | 2016-11-02 |
| CN106066890B CN106066890B (en) | 2020-02-18 |
Family
ID=57421256
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610427650.3A Active CN106066890B (en) | 2016-06-16 | 2016-06-16 | Distributed high-performance database all-in-one machine system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106066890B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106843770A (en) * | 2017-01-23 | 2017-06-13 | 北京思特奇信息技术股份有限公司 | A kind of distributed file system small file data storage, read method and device |
| CN106933514A (en) * | 2017-03-15 | 2017-07-07 | 郑州云海信息技术有限公司 | A kind of automatic location mode of cold and hot data of ASM disks group and system |
| CN106991120A (en) * | 2017-02-22 | 2017-07-28 | 杭州沃趣科技股份有限公司 | A kind of method for realizing oracle database with computer room zero-data loss |
| CN107463336A (en) * | 2017-08-07 | 2017-12-12 | 郑州云海信息技术有限公司 | The processing method and system that solid magnetic disc fails in a kind of SSD cachings |
| CN107544757A (en) * | 2017-08-11 | 2018-01-05 | 郑州云海信息技术有限公司 | A kind of method and system based on MCS system quick search disk id and Firmware versions |
| CN108153759A (en) * | 2016-12-05 | 2018-06-12 | 中国移动通信集团公司 | A kind of data transmission method of distributed data base, middle tier server and system |
| CN109144412A (en) * | 2018-07-26 | 2019-01-04 | 郑州云海信息技术有限公司 | A kind of iSCSI adapter batch scanning method and system |
| CN109213827A (en) * | 2017-07-03 | 2019-01-15 | 阿里巴巴集团控股有限公司 | A kind of data processing system, method, router and from database |
| CN110161930A (en) * | 2019-06-05 | 2019-08-23 | 安徽三实信息技术服务有限公司 | A kind of data monitoring system and its data monitoring method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102385628A (en) * | 2011-11-14 | 2012-03-21 | 北京锐安科技有限公司 | Java data base connectivity (JDBC)-based data distributed processing method |
| US20120239730A1 (en) * | 2010-09-15 | 2012-09-20 | Oracle International Corporation | System including a middleware machine environment |
| CN103516809A (en) * | 2013-10-22 | 2014-01-15 | 浪潮电子信息产业股份有限公司 | High-scalability and high-performance distributed storage system structure |
| CN103873559A (en) * | 2014-02-13 | 2014-06-18 | 南京斯坦德通信股份有限公司 | Database all-in-one machine capable of realizing high-speed storage |
| CN104035522A (en) * | 2014-06-16 | 2014-09-10 | 南京云创存储科技有限公司 | Large database appliance |
-
2016
- 2016-06-16 CN CN201610427650.3A patent/CN106066890B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120239730A1 (en) * | 2010-09-15 | 2012-09-20 | Oracle International Corporation | System including a middleware machine environment |
| CN102385628A (en) * | 2011-11-14 | 2012-03-21 | 北京锐安科技有限公司 | Java data base connectivity (JDBC)-based data distributed processing method |
| CN103516809A (en) * | 2013-10-22 | 2014-01-15 | 浪潮电子信息产业股份有限公司 | High-scalability and high-performance distributed storage system structure |
| CN103873559A (en) * | 2014-02-13 | 2014-06-18 | 南京斯坦德通信股份有限公司 | Database all-in-one machine capable of realizing high-speed storage |
| CN104035522A (en) * | 2014-06-16 | 2014-09-10 | 南京云创存储科技有限公司 | Large database appliance |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108153759B (en) * | 2016-12-05 | 2021-07-09 | 中国移动通信集团公司 | Data transmission method of distributed database, intermediate layer server and system |
| CN108153759A (en) * | 2016-12-05 | 2018-06-12 | 中国移动通信集团公司 | A kind of data transmission method of distributed data base, middle tier server and system |
| CN106843770A (en) * | 2017-01-23 | 2017-06-13 | 北京思特奇信息技术股份有限公司 | A kind of distributed file system small file data storage, read method and device |
| CN106991120A (en) * | 2017-02-22 | 2017-07-28 | 杭州沃趣科技股份有限公司 | A kind of method for realizing oracle database with computer room zero-data loss |
| CN106933514A (en) * | 2017-03-15 | 2017-07-07 | 郑州云海信息技术有限公司 | A kind of automatic location mode of cold and hot data of ASM disks group and system |
| CN109213827A (en) * | 2017-07-03 | 2019-01-15 | 阿里巴巴集团控股有限公司 | A kind of data processing system, method, router and from database |
| CN109213827B (en) * | 2017-07-03 | 2022-07-08 | 阿里云计算有限公司 | Data processing system, method, router and slave database |
| CN107463336B (en) * | 2017-08-07 | 2020-09-22 | 苏州浪潮智能科技有限公司 | Method and system for processing failure of solid state disk in SSD (solid state disk) cache |
| CN107463336A (en) * | 2017-08-07 | 2017-12-12 | 郑州云海信息技术有限公司 | The processing method and system that solid magnetic disc fails in a kind of SSD cachings |
| CN107544757A (en) * | 2017-08-11 | 2018-01-05 | 郑州云海信息技术有限公司 | A kind of method and system based on MCS system quick search disk id and Firmware versions |
| CN109144412A (en) * | 2018-07-26 | 2019-01-04 | 郑州云海信息技术有限公司 | A kind of iSCSI adapter batch scanning method and system |
| CN110161930A (en) * | 2019-06-05 | 2019-08-23 | 安徽三实信息技术服务有限公司 | A kind of data monitoring system and its data monitoring method |
| CN110161930B (en) * | 2019-06-05 | 2021-11-09 | 安徽三实信息技术服务有限公司 | Data monitoring system and data monitoring method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106066890B (en) | 2020-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106066890A (en) | A kind of distributed high-performance data storehouse integrated machine system | |
| US9823980B2 (en) | Prioritizing data reconstruction in distributed storage systems | |
| US11163699B2 (en) | Managing least recently used cache using reduced memory footprint sequence container | |
| CN102012791B (en) | Flash based PCIE (peripheral component interface express) board for data storage | |
| US20180024964A1 (en) | Disaggregated compute resources and storage resources in a storage system | |
| CN103873559A (en) | Database all-in-one machine capable of realizing high-speed storage | |
| CN105843557B (en) | Redundant storage system, redundant storage method and redundant storage device | |
| CN107967124A (en) | A kind of distribution persistence memory storage system and method | |
| CN102117248A (en) | Caching system and method for caching data in caching system | |
| CN100452046C (en) | Storage method and system for mass file | |
| CN105657066A (en) | Load rebalance method and device used for storage system | |
| CN103455577A (en) | Multi-backup nearby storage and reading method and system of cloud host mirror image file | |
| JP2018520420A (en) | Cache architecture and algorithm for hybrid object storage devices | |
| US20100180066A1 (en) | Electronically addressed non-volatile memory-based kernel data cache | |
| CN105516313A (en) | Distributed storage system used for big data | |
| CN111736764B (en) | Storage system of database all-in-one machine and data request processing method and device | |
| Xu et al. | Rethink the storage of virtual machine images in clouds | |
| US11379326B2 (en) | Data access method, apparatus and computer program product | |
| Zhang et al. | Leveraging glocality for fast failure recovery in distributed RAM storage | |
| CN102929813A (en) | Method for designing peripheral component interconnect express (PCI-E) interface solid hard disk controller | |
| US20230237029A1 (en) | Data deduplication in a storage system | |
| WO2022121274A1 (en) | Metadata management method and apparatus in storage system, and storage system | |
| CN101859232A (en) | Variable length data memory interface | |
| CN105978987B (en) | A kind of virtual method for Exadata memory node | |
| Jummah | Enhancing System Performance Using Persistent RAM Modules as Storage Class Memory |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |