CN112363676A - Control method and system based on low access delay distributed storage system - Google Patents
Control method and system based on low access delay distributed storage system Download PDFInfo
- Publication number
- CN112363676A CN112363676A CN202011294948.4A CN202011294948A CN112363676A CN 112363676 A CN112363676 A CN 112363676A CN 202011294948 A CN202011294948 A CN 202011294948A CN 112363676 A CN112363676 A CN 112363676A
- Authority
- CN
- China
- Prior art keywords
- client
- service
- service process
- storage system
- distributed storage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- 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
- G06F3/0655—Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a control method and a system based on a low-access-delay distributed storage system, which comprises the following steps: a client acquires a connection database of a plurality of service processes in a cluster where the client is located, wherein the connection database comprises matching information of the service processes and process parameters; the client determines the process parameters of the client and the service process according to the connection database; and the client establishes communication between the client and the service process corresponding to the process parameters according to the process parameters. According to the method and the device, the client side obtains the service parameters of the service process according to the connection database, the client side is directly communicated with the hard disk corresponding to the service process according to the service parameters, and the client side can directly send data to the hard disk corresponding to the service process without passing through the service side, so that IO delay is reduced.
Description
Technical Field
The invention relates to the technical field of distributed storage, in particular to a control method and a control system based on a low-access-delay distributed storage system.
Background
The control system based on the distributed storage system dispersedly stores data on a plurality of independent devices. The traditional network storage system adopts a centralized storage server to store all data, and the storage server becomes the bottleneck of system performance and is also the focus of reliability and safety.
In the application process of the control system based on the distributed storage system, mass data and files are often borne, the delay of data IO is influenced to a certain extent along with the increase of the data and the files, the reading and writing of the data and the files cannot be guaranteed to be completed quickly, and the experience of a user is reduced.
Disclosure of Invention
The present invention is directed to solving at least one of the problems of the prior art. Therefore, the invention provides a control method based on a low-access-delay distributed storage system, which can reduce the delay of data IO and improve the experience of users.
The invention also provides a control system based on the low-access-delay distributed storage system.
In a first aspect, an embodiment of the present invention provides a control method for a distributed storage system based on low access latency, including:
a client acquires a connection database of a plurality of service processes in a cluster where the client is located, wherein the connection database comprises matching information of the service processes and process parameters;
the client determines the process parameters of the client and the service process according to the connection database;
and the client establishes communication between the client and the service process corresponding to the process parameters according to the process parameters.
The control method based on the low-access-delay distributed storage system provided by the embodiment of the invention at least has the following beneficial effects: the service parameters of the service process are obtained through the client according to the connection database, the client is directly communicated with the hard disk corresponding to the service process according to the service parameters, and the client can directly send data to the hard disk corresponding to the service process without passing through the server, so that IO delay is reduced, and user experience is improved.
According to other embodiments of the present invention, the method for controlling a distributed storage system based on low access latency includes: a service process number and a service IP address.
According to another embodiment of the present invention, a control method for a low access latency distributed storage system, wherein the client determines the process parameters of the client and the service process according to the connection database, includes:
the client determines the service process according to the data distribution characteristics;
and determining the service process number and the service IP address of the service process according to the service process and the connection database.
According to other embodiments of the present invention, a control method for a low access latency distributed storage system, further comprises:
and after the client finishes the communication with the service process, the client directly stores the data to the hard disk of the communicated service process.
According to other embodiments of the present invention, a control method for a low access latency distributed storage system, further comprises:
the storage server loads a TCP data packet processing driver;
the storage server side starts the TCP data packet processor driver to process the TCP data packet.
According to other embodiments of the present invention, in the control method based on a low access latency distributed storage system, the TCP packet processing driver is a TOE driver.
In a second aspect, an embodiment of the present invention provides a control system for a low access latency distributed storage system, including:
a client, the client comprising:
an obtaining module, configured to obtain a connection database of multiple service processes in a cluster where the client is located, where the connection database includes: matching information of the service process and the process parameters;
the processing module is used for determining the process parameters of the client and the service process according to the connection database;
and the establishing module is used for establishing the communication between the client and the service process corresponding to the process parameters according to the process parameters.
The control system based on the low-access-delay distributed storage system provided by the embodiment of the invention at least has the following beneficial effects: the service parameters of the service process are obtained through the client according to the connection database, the client is directly communicated with the hard disk corresponding to the service process according to the service parameters, and the client can directly send data to the hard disk corresponding to the service process without passing through the server.
According to other embodiments of the present invention, a control system based on a low access latency distributed storage system further comprises: a storage server, the storage server comprising: TCP packet processing driver;
and the storage server loads the TCP data packet processing driver and starts the TCP data packet processing driver to process the TCP data packet.
According to another embodiment of the present invention, in the control system based on the low access latency distributed storage system, the TCP packet processing driver is a toe driver, and the toe driver is integrated on a load reduction engine network card.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the application. The objectives and other advantages of the application may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
FIG. 1 is a flow chart illustrating a method for controlling a low access latency distributed storage system according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating a method for controlling a low access latency distributed storage system according to another embodiment of the present invention;
FIG. 3 is a flow chart illustrating a method for controlling a low access latency distributed storage system according to another embodiment of the present invention;
FIG. 4 is a flow chart illustrating a method for controlling a low access latency distributed storage system according to another embodiment of the present invention;
FIG. 5 is a block diagram of an embodiment of a control system of a low access latency distributed storage system according to the present invention.
Reference numerals: 100. a client; 110. an acquisition module; 120. a processing module; 130. establishing a module; 200. and storing the server side.
Detailed Description
The concept and technical effects of the present invention will be clearly and completely described below in conjunction with the embodiments to fully understand the objects, features and effects of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments, and those skilled in the art can obtain other embodiments without inventive effort based on the embodiments of the present invention, and all embodiments are within the protection scope of the present invention.
In the description of the embodiments of the present invention, if "a number" is referred to, it means one or more, if "a plurality" is referred to, it means two or more, if "greater than", "less than" or "more than" is referred to, it is understood that the number is not included, and if "greater than", "lower" or "inner" is referred to, it is understood that the number is included. If reference is made to "first" or "second", this should be understood to distinguish between features and not to indicate or imply relative importance or to implicitly indicate the number of indicated features or to implicitly indicate the precedence of the indicated features.
In the application process of the distributed storage system, a large amount of data and files are often carried, how to quickly retrieve and access the data is a vital technology, and particularly how to reduce the access delay of each IO or file is vital. Especially, on some domestic chip technology platforms, how to ensure that the delay of IO is not affected by software optimization in the process of slow processor processing.
Based on the above, the application discloses a control method and system based on a low-access-delay distributed storage system, which can reduce IO delay and improve user experience.
In a first aspect, the present application discloses a control method for a low access latency distributed storage system, including:
s100, a client acquires a connection database of a plurality of service processes in a cluster where the client is located, wherein the connection database comprises matching information of the service processes and process parameters;
s200, the client determines process parameters of the client and the service process according to the connection database;
s300, the client establishes communication between the client and the service process corresponding to the process parameters according to the process parameters.
In a general storage scheme, a protocol gateway service module similar to a controller generally exists for supporting block storage, so as to provide block storage protocols such as iSCSI and FC SCSI externally, and distribute IO to different nodes and hard disks internally. However, one upper layer service IO needs to complete data dropping at least twice through network transmission. Because the delay consumption of the network IO is high, if the number of network interactions can be minimized, the overall delay can be reduced. Therefore, the client side directly obtains the connection database, determines the service process needing to store data according to the connection database, then obtains the process parameters of the service process, and finally establishes communication directly according to the process parameters and the service process corresponding to the process parameters, so that the client side can directly communicate with the hard disk corresponding to the service process to directly complete data reading and writing, a third party is not needed to transfer data, only one-time network IO (input/output) is needed in the whole process, and the network delay expense is greatly reduced.
In some embodiments, the process parameters include: service process number and service IO address. Each hard disk of each storage server side is provided with an independent storage service, the independent storage service is a service process, all the service processes are assembled into a storage service cluster, and each service process is provided with a corresponding service process number and a corresponding service IP address, so that the corresponding service process can be inquired according to the service process number and the service IO address.
Referring to fig. 2, in some embodiments, step S200 includes:
s210, the client determines a service process according to the data distribution characteristics;
s220, determining the service process number and the service IP address of the service process according to the service process and the connection database.
The connection database stores matching information of the service processes and the process parameters, the client divides each file into a plurality of objects, then determines a hard disk corresponding to the service process required to be stored by each object, and obtains the service process number and the service IP address of the service process according to the service process and the matching information, so that the client can directly establish communication connection with the service process according to the service process number and the service IP address.
Specifically, the client directly sends the data to the corresponding service process according to the CRUSH algorithm. The CRUSH algorithm aims to locate the position of stored data, and calculate the distribution of data objects through the weight of each device, so that a corresponding service process is determined, and corresponding process parameters are obtained according to the service process, so that a client can establish communication connection with the corresponding service process according to the connection parameters, and the data can be directly sent to the service process. The communication connection between the client and the service process is mainly provided for the upper layer application by an API mode to provide a service process number and a service IP address related to the service process, and the upper layer application realizes the IO butt joint through the service process number and the service IP address according to the API mode so as to realize the hard disk communication connection corresponding to the client and the service process.
Referring to fig. 3, in some embodiments, the control method based on the low access latency distributed storage system further includes:
s400, after the client finishes the communication with the service process, the client directly stores the data to a hard disk of the communicated service process.
After the upper application of the client establishes communication with the service process, the hard disk corresponding to the client and the service process is connected, so that the client can store the hard disk of the service process which is in communication connection after splitting the data, and the data is directly stored in the corresponding hard disk. The data can be directly communicated with the service process of each hard disk of each host in the storage cluster, data reading and writing can be directly finished without a third party, the whole process only needs to carry out network IO once, and network delay overhead is greatly reduced.
Referring to fig. 4, in some embodiments, a control method of a low access latency-based distributed storage system further includes:
s500, loading a TCP data packet processing driver by a storage server;
s600, the storage server side starts a TCP data packet processor driver to process the TCP data packet.
Under the condition of high load in the common network card driving mode, the storage service end can be seen to generate a huge amount of CPU interruption, and 10% -20% of kernel CPU overhead is needed to process TCP data packets, so that certain CPU overhead is consumed, and IO delay is reduced. The storage node is provided with a load reduction engine network card with a TCP data packet processing driver, when a CPU is interrupted, the storage server loads the TCP data packet processing driver, and the storage server starts the TCP data packet processing driver to process a TCP data packet through the TCP data packet processing driver, so that the CPU consumption is saved, and the interruption condition of the network card can be reduced. Because TCP data packet processing drive is not provided, the unpacking process of TCP/IP is completed in a user state, and a large amount of memory copying actions are carried out, so that the efficiency is lost.
In some embodiments, the TCP packet processing driver is a TOE driver, and the offload engine network card has a TOE property, and the TOE property can be enabled by migrating and recompiling a driver of a higher version of the offload engine network card. And the storage server side starts up and loads the TOE driver, so that the storage software layer can use the TOE capability of the network card without sensing.
In this embodiment, the model of the load shedding engine network card is Chelsio T5.
A control method of a low access latency-based distributed storage system according to an embodiment of the present invention is described in detail below with reference to fig. 1 to 4 in a specific embodiment. It is to be understood that the following description is only exemplary, and not a specific limitation of the invention.
The client calculates the distribution of the data objects through the weight of each device according to a CRUSH algorithm, determines a corresponding service process, acquires a corresponding process parameter service process number and a service IP address according to the service process and a connection database, and then the client mainly provides the service process number and the service IP address of the service process for the upper layer application in an API mode, the upper layer application realizes the butt joint with the service process through the service process number and the service IP address according to the API mode so as to realize the hard disk communication connection corresponding to the client and the service process, and the client can store the hard disk of the service process which is in communication connection after splitting the data. When the CPU is interrupted, the storage server loads the TOE driver, the storage server starts the TOE driver to process the TCP data packet through the TOE driver, and the TCP data packet is processed through the TOE driver, so that the consumption of the CPU is saved.
In a second aspect, referring to fig. 5, an embodiment of the present invention discloses a control system based on a low access latency distributed storage system, including: the client 100, the client 100 includes: an obtaining module 110, a processing module 120, and an establishing module 130, where the obtaining module 110 is configured to obtain a connection database of a plurality of service processes in a cluster in which the client 100 is located, where the connection database includes: matching information of the service process and the process parameters; the processing module 120 is configured to determine process parameters of the client 100 and the service process according to the connection database; the establishing module 130 is configured to establish communication between the client 100 and a service process corresponding to the process parameter according to the process parameter.
The client 100 directly obtains the service parameters of the service process according to the connection database, the client 100 directly communicates with the hard disk corresponding to the service process according to the service parameters, and the client 100 can directly send data to the hard disk corresponding to the service process without passing through a server.
The control system based on the low access delay distributed storage system further comprises: the storage server 200, the storage server 200 includes: the storage server 200 loads the TCP packet processing driver and starts the TCP packet processing driver to process the TCP packet. The TCP data packet processing driver is a toe driver which is integrated on the load reduction engine network card.
The specific operation process of the control system based on the low-access-latency distributed storage system refers to the control method based on the low-access-latency distributed storage system of the first aspect, and is not described herein again.
The above-described embodiments of the apparatus are merely illustrative, wherein the units illustrated as separate components may or may not be physically separate, i.e. may be located in one place, or may also be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
One of ordinary skill in the art will appreciate that all or some of the steps, systems, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as is well known to those of ordinary skill in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, Digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can accessed by a computer. In addition, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media as known to those skilled in the art.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention. Furthermore, the embodiments of the present invention and the features of the embodiments may be combined with each other without conflict.
Claims (9)
1. The control method based on the low access delay distributed storage system is characterized by comprising the following steps:
a client acquires a connection database of a plurality of service processes in a cluster where the client is located, wherein the connection database comprises matching information of the service processes and process parameters;
the client determines the process parameters of the client and the service process according to the connection database;
and the client establishes communication between the client and the service process corresponding to the process parameters according to the process parameters.
2. The control method of claim 1, wherein the process parameters comprise: a service process number and a service IP address.
3. The control method based on the low access delay distributed storage system according to claim 2, wherein the determining, by the client, the process parameters of the client and the service process according to the connection database comprises:
the client determines the service process according to the data distribution characteristics;
and determining the service process number and the service IP address of the service process according to the service process and the connection database.
4. The control method based on the low access latency distributed storage system according to any one of claims 1 to 3, further comprising:
and after the client finishes the communication with the service process, the client directly stores the data to the hard disk of the communicated service process.
5. The control method of the distributed storage system based on low access latency according to claim 1, further comprising:
the storage server loads a TCP data packet processing driver;
the storage server side starts the TCP data packet processor driver to process the TCP data packet.
6. The control method based on the low access latency distributed storage system according to claim 5, wherein the TCP packet processing driver is a TOE driver.
7. A control system for a low access latency distributed storage system, comprising:
a client, the client comprising:
an obtaining module, configured to obtain a connection database of multiple service processes in a cluster where the client is located, where the connection database includes: matching information of the service process and the process parameters;
the processing module is used for determining the process parameters of the client and the service process according to the connection database;
and the establishing module is used for establishing the communication between the client and the service process corresponding to the process parameters according to the process parameters.
8. The low access latency distributed storage system based control system according to claim 7, further comprising: a storage server, the storage server comprising: the TCP packet processing driver is driven by the TCP packet processing,
and the storage server loads the TCP data packet processing driver and starts the TCP data packet processing driver to process the TCP data packet.
9. The low access latency distributed storage system based control system according to claim 8, wherein the TCP packet processing driver is a toe driver, and the toe driver is integrated on a load reduction engine network card.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011294948.4A CN112363676A (en) | 2020-11-18 | 2020-11-18 | Control method and system based on low access delay distributed storage system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011294948.4A CN112363676A (en) | 2020-11-18 | 2020-11-18 | Control method and system based on low access delay distributed storage system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112363676A true CN112363676A (en) | 2021-02-12 |
Family
ID=74533455
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011294948.4A Pending CN112363676A (en) | 2020-11-18 | 2020-11-18 | Control method and system based on low access delay distributed storage system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112363676A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110099219A1 (en) * | 2009-10-23 | 2011-04-28 | International Business Machines Corporation | Universal architecture for client management extensions on monitoring, control, and configuration |
| CN106385448A (en) * | 2016-09-13 | 2017-02-08 | 郑州云海信息技术有限公司 | Method and device for communicating between client and server |
| CN109213611A (en) * | 2018-08-01 | 2019-01-15 | 天津字节跳动科技有限公司 | The striding course means of communication, device, terminal and storage medium |
| CN111930315A (en) * | 2020-08-21 | 2020-11-13 | 北京天融信网络安全技术有限公司 | Data access method, data access device and storage medium |
-
2020
- 2020-11-18 CN CN202011294948.4A patent/CN112363676A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110099219A1 (en) * | 2009-10-23 | 2011-04-28 | International Business Machines Corporation | Universal architecture for client management extensions on monitoring, control, and configuration |
| CN106385448A (en) * | 2016-09-13 | 2017-02-08 | 郑州云海信息技术有限公司 | Method and device for communicating between client and server |
| CN109213611A (en) * | 2018-08-01 | 2019-01-15 | 天津字节跳动科技有限公司 | The striding course means of communication, device, terminal and storage medium |
| CN111930315A (en) * | 2020-08-21 | 2020-11-13 | 北京天融信网络安全技术有限公司 | Data access method, data access device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111277616B (en) | RDMA-based data transmission method and distributed shared memory system | |
| US20200356277A1 (en) | De-duplication of client-side data cache for virtual disks | |
| JP6607783B2 (en) | Distributed cache cluster management | |
| CN108028833A (en) | A kind of method, system and the relevant device of NAS data accesses | |
| US20180013826A1 (en) | Resource node interface protocol | |
| CN114201421B (en) | Data stream processing method, storage control node and readable storage medium | |
| CN112130748A (en) | Data access method, network card and server | |
| US11966614B2 (en) | Object tiering in a distributed storage system | |
| CN111460504A (en) | Service processing method, device, node equipment and storage medium | |
| CN108777718B (en) | Method and device for accessing read-write-more-less system through client side by service system | |
| CN106973091B (en) | Distributed memory data redistribution method and system, and master control server | |
| US10387043B2 (en) | Writing target file including determination of whether to apply duplication elimination | |
| US8549274B2 (en) | Distributive cache accessing device and method for accelerating to boot remote diskless computers | |
| US7213245B2 (en) | Software on demand system | |
| CN111262904B (en) | Service agent system and method | |
| CN112363676A (en) | Control method and system based on low access delay distributed storage system | |
| WO2024021470A1 (en) | Cross-region data scheduling method and apparatus, device, and storage medium | |
| CN111506254A (en) | Distributed storage system and management method and device thereof | |
| CN112106327B (en) | Resource management and resource allocation method in self-optimization network of heterogeneous processing nodes | |
| CN112748850B (en) | Method, apparatus and computer program product for storage management | |
| CN103051975A (en) | P2P (peer to peer) cache data elimination method | |
| JP6423491B1 (en) | Method for providing service to customer by at least one operator using network and network thereof | |
| CN108733805B (en) | File interaction method, system, computer equipment and storage medium | |
| CN105912477A (en) | Directory reading method, device and system | |
| US10169363B2 (en) | Storing data in a distributed file 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 |