CN112532701A - Distributed accelerated storage method and system based on QEMU (quantum execution unit) - Google Patents

Distributed accelerated storage method and system based on QEMU (quantum execution unit) Download PDF

Info

Publication number
CN112532701A
CN112532701A CN202011302230.5A CN202011302230A CN112532701A CN 112532701 A CN112532701 A CN 112532701A CN 202011302230 A CN202011302230 A CN 202011302230A CN 112532701 A CN112532701 A CN 112532701A
Authority
CN
China
Prior art keywords
qemu
distributed
copy
issued
information
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
Application number
CN202011302230.5A
Other languages
Chinese (zh)
Other versions
CN112532701B (en
Inventor
马怀旭
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Suzhou Inspur Intelligent Technology Co Ltd
Original Assignee
Suzhou Inspur Intelligent Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Suzhou Inspur Intelligent Technology Co Ltd filed Critical Suzhou Inspur Intelligent Technology Co Ltd
Priority to CN202011302230.5A priority Critical patent/CN112532701B/en
Publication of CN112532701A publication Critical patent/CN112532701A/en
Application granted granted Critical
Publication of CN112532701B publication Critical patent/CN112532701B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45504—Abstract machines for programme code execution, e.g. Java virtual machine [JVM], interpreters, emulators
    • G06F9/45508—Runtime interpretation or emulation, e g. emulator loops, bytecode interpretation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
    • H04L41/0663—Performing the actions predefined by failover planning, e.g. switching to standby network elements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/104—Peer-to-peer [P2P] networks
    • H04L67/1061—Peer-to-peer [P2P] networks using node-based peer discovery mechanisms
    • H04L67/1065—Discovery involving distributed pre-established resource-based relationships among peers, e.g. based on distributed hash tables [DHT] 

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Hardware Redundancy (AREA)

Abstract

The invention relates to the technical field of distributed storage, and provides a distributed accelerated storage method and a system based on a QEMU (QEMU), wherein the method comprises the following steps: at the client side of the QEMU, calculating the position of the copy through a distributed hash table algorithm built in a driver of the QEMU; according to the position of the copy, IO (input/output) is issued to a host where the copy obtained by calculation is located through a driver of the QEMU; the host where the copy is located carries out IO processing on the received IO request issued by the driver of the QEMU and feeds the IO processing result back to the host where the QEMU issues the IO, so that the IO path of distributed storage provided by the QEMU for accessing the distributed hash table algorithm is shortened, the IO availability of the distributed storage is improved, the IO delay information is shortened, and the adaptability of the distributed storage based on a cloud computing scene is improved.

Description

Distributed accelerated storage method and system based on QEMU (quantum execution unit)
Technical Field
The invention belongs to the technical field of distributed storage, and particularly relates to a distributed accelerated storage method and system based on a QEMU (quantum execution unit).
Background
In the era of information explosion growth, mass data grows, the traditional storage cost is high, the efficiency is low, the growth speed of user data cannot be met, the pain is solved by a high-efficiency intelligent distributed storage technology, and the distributed storage has the following characteristics: high performance, high reliability, high expandability, transparency and autonomy. The distributed storage data storage firstly needs to be subjected to fragmentation and cutting processing, then the data storage position is calculated through a certain algorithm, and as the user data is divided into a plurality of data blocks, the data can not be used due to the fact that any data block is lost, therefore, a reasonable redundant storage model must be considered in the distributed storage, a plurality of redundant storage copies are provided for the data blocks of the user, and therefore the safety and the reliability of the data are guaranteed.
Two main schemes are provided for realizing distributed storage, one scheme is realized based on a DHT algorithm, and the other scheme is realized based on a metadata center; for storage provided by distributed storage, there are three directions: object storage, file storage, and block storage. The object storage is mainly used for storing unchangeable objects, the file storage is mainly used for storing files, and the block storage provides block equipment; block storage is generally an application that provides blocks to the emulated memory QEMU to use to create virtual machines or to provide databases or file storage; there are two ways for normally using block storage, one is that distributed storage provides iSCSI device mapping to a host, and the other is that distributed storage is used through a proprietary protocol direct connection, but these ways all face the disadvantages of relatively long IO path and relatively large delay.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a distributed accelerated storage method based on a QEMU (QEMU), aiming at solving the problems of long IO path and large time delay in distributed storage in the prior art.
The technical scheme provided by the invention is as follows: a method for distributed accelerated storage based on a QEMU emulation processor, the method comprising the steps of:
at the client side of the analog processor QEMU, calculating the position of the copy through a distributed hash table algorithm built in a driver of the analog processor QEMU;
according to the position of the copy obtained by calculation, IO (input/output) is issued to a host where the copy obtained by calculation is located through a driver of the QEMU;
and the host where the copy is located carries out IO processing on the received IO request issued by the driver of the QEMU, and feeds back the IO processing result to the host where the IO is issued by the QEMU.
As an improved scheme, the step of issuing an IO to the host where the calculated copy is located through the driver of the QEMU according to the calculated position of the copy specifically includes the following steps:
acquiring volume information of IO issued by the QEMU of the analog processor and offset distance value offset of the issued IO;
calculating target object information of the issued IO according to the acquired volume information of the IO issued by the QEMU of the analog processor and the offset value offset of the issued IO;
calculating a target object position parameter by using the distributed hash table algorithm according to the calculated target object information;
and according to the copy number information of the volume and the object position parameter of the target object, transmitting the IO to a host where the multiple copies are located through a socket.
As an improvement, the method further comprises the steps of:
configuring a plurality of IO driving modes in the QEMU, wherein the plurality of IO driving modes support a plurality of corresponding IO issuing schemes;
configuring IP information of a plurality of cluster nodes iNode based on distributed storage and configuring floating IP based on distributed storage according to a plurality of IO driving modes in the configured QEMU;
the IP information of the cluster nodes iNode based on the distributed storage is used for IP switching when an accessed IP fails, and the floating IP based on the distributed storage is used for actively performing floating IP switching when the nodes fail.
As an improvement, the method further comprises the steps of:
and when the analog processor QEMU issues IO abnormity, automatically executing IO retry action.
As an improvement, the method further comprises the steps of:
when the distributed storage system fails, checking a queue and an IO queue which are being requested, acquiring IO information of distributed storage which is being defended, and controlling to preferentially recover a block which is being accessed;
when the distributed storage system fails back, an IO retry action is performed on the IO being requested.
Another object of the present invention is to provide a distributed accelerated storage system based on a QEMU, the system comprising:
the copy position calculation module is used for calculating the position of the copy at the client side of the QEMU through a distributed hash table algorithm built in a driver of the QEMU;
the IO issuing module is used for issuing IO to a host where the copy obtained by calculation is located through a driver of the QEMU according to the position of the copy obtained by calculation;
and the IO processing module is used for carrying out IO processing on the received IO request sent by the driver of the QEMU by the host where the copy is located and feeding back an IO processing result to the host where the IO is sent by the QEMU.
As an improved scheme, the IO issue module specifically includes:
the information acquisition module is used for acquiring volume information of IO issued by the QEMU of the analog processor and offset distance value offset of the issued IO;
the target object information acquisition module is used for calculating target object information of the issued IO according to the acquired volume information of the IO issued by the QEMU and the offset distance value offset of the issued IO;
the object position parameter acquisition module is used for calculating object position parameters of the target object by utilizing the distributed hash table algorithm according to the calculated object information of the target object;
and the IO transmission module is used for transmitting the IO to a host where the plurality of copies are located through the socket according to the copy number information of the volume and the target object location parameter.
As an improvement, the system further comprises:
the first configuration module is used for configuring a plurality of IO driving modes in the QEMU, wherein the plurality of IO driving modes support a plurality of corresponding IO issuing schemes;
the second configuration module is used for configuring IP information of a plurality of cluster nodes iNode based on distributed storage and configuring floating IP based on distributed storage according to a plurality of IO driving modes in the configured QEMU;
the IP information of the cluster nodes iNode based on the distributed storage is used for IP switching when an accessed IP fails, and the floating IP based on the distributed storage is used for actively performing floating IP switching when the nodes fail.
As an improvement, the system further comprises:
and the first IO retry module is used for automatically executing IO retry actions when the analog processor QEMU issues IO abnormity.
As an improvement, the system further comprises:
the priority recovery control module is used for checking the queue and the IO queue which are being requested when the distributed storage system fails, acquiring the IO information of distributed storage which is being defended and controlling the priority recovery of the accessed block;
and the second IO retry module is used for executing IO retry actions on the IO being requested when the failure of the distributed storage system is recovered.
In the embodiment of the invention, at the client side of the QEMU, the position of the copy is calculated by a distributed hash table algorithm built in a driver of the QEMU; according to the position of the copy obtained by calculation, IO (input/output) is issued to a host where the copy obtained by calculation is located through a driver of the QEMU; and the host where the copy is located carries out IO processing on the received IO request issued by the driver of the QEMU, and feeds the IO processing result back to the host where the IO is issued by the QEMU, so that the IO path of the distributed storage provided by the QEMU for accessing the distributed hash table algorithm is shortened, the IO availability of the distributed storage is improved, the time delay information of the IO is shortened, and the adaptability of the distributed storage based on the cloud computing scene is improved.
Drawings
In order to more clearly illustrate the detailed description of the invention or the technical solutions in the prior art, the drawings that are needed in the detailed description of the invention or the prior art will be briefly described below. Throughout the drawings, like elements or portions are generally identified by like reference numerals. In the drawings, elements or portions are not necessarily drawn to scale.
FIG. 1 is a flow chart of an implementation of a distributed accelerated storage method based on a QEMU of an analog processor provided by the present invention;
fig. 2 is a flowchart illustrating an implementation according to the present invention, where an IO issue is performed to a host where a calculated copy is located through a driver of the QEMU according to the calculated position of the copy;
FIG. 3 is a block diagram of a distributed accelerated storage system based on a QEMU simulation processor provided by the present invention;
fig. 4 is a structural block diagram of an IO issue module provided in the present invention.
Detailed Description
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following examples are merely for illustrating the technical solutions of the present invention more clearly, and therefore are only examples, and the protection scope of the present invention is not limited thereby.
Fig. 1 is a flowchart of an implementation of the distributed accelerated storage method based on QEMU of the analog processor according to the present invention, which specifically includes the following steps:
in step S101, at the client of the analog processor QEMU, the location of the copy is calculated by the distributed hash table algorithm built in the driver of the analog processor QEMU;
in this step, the DHT algorithm of distributed storage is transplanted to the driver inside of the QEMU, and the client side of the QEMU can directly calculate the location of the copy through the DHT algorithm, specifically:
distributed storage based on the DHT algorithm has no metadata service, the IO copy position is calculated according to the DHT algorithm used by the distributed storage and little cluster information, and then IO is directly distributed to a host where the copy is located to directly issue IO;
among other things, DHTs use distributed hash algorithms to solve structured distributed storage problems. The core idea of the distributed Hash algorithm is to obtain a Key value (Hash Key) by performing Hash operation on the characteristics (keywords) of a storage object, and the distributed storage of the object is performed according to the Key value.
In step S102, according to the calculated position of the copy, an IO issue is performed to a host where the calculated copy is located through a driver of the QEMU;
in step S103, the host where the copy is located performs IO processing on the received IO request issued by the driver of the QEMU, and feeds back an IO processing result to the host where the analog processor QEMU issues IO.
In the embodiment, methods of cluster configuration information acquisition, IO retransmission and the like are added at the same time, IO distribution is placed on the client, and IO flows are simplified by using the computing capacity of the client.
In the embodiment of the present invention, as shown in fig. 2, the step of issuing an IO to the host where the calculated copy is located through the driver of the QEMU according to the calculated position of the copy specifically includes the following steps:
in step S201, volume information of the IO issued by the QEMU and an offset distance value offset of the IO issued by the QEMU are acquired;
in this step, the distributed storage is generally performed by fragmentation, information of the target object to be issued can be calculated according to information of the volume of the IO issued by the QEMU of the analog processor and the offset value of the IO issued, and the position of the target object can be directly calculated by using the DHT algorithm according to the information of the target object.
In step S202, target object information of the IO to be issued is calculated according to the obtained volume information of the IO issued by the QEMU and the offset distance value offset of the issued IO;
in step S203, calculating a target object location parameter by using the distributed hash table algorithm according to the calculated target object information;
in step S204, according to the copy number information of the volume and the object location parameter of the target object, the IO is transmitted to the host where the multiple copies are located through the socket.
In the step, IO is directly carried out through the socket according to the number of the copies of the volume and the calculated object position, and the IO is transmitted to the host where the multiple copies are located, IO forwarding is not needed, and IO path information is simplified.
In the embodiment of the present invention, multiple IO driving modes are configured in the analog processor QEMU, where the multiple IO driving modes support multiple corresponding IO delivery schemes, that is:
the analog processor QEMU supports various IO modes, and can support more IO schemes by customizing the IOdriver mode of the analog processor QEMU;
configuring IP information of a plurality of cluster nodes iNode based on distributed storage and configuring floating IP based on distributed storage according to a plurality of IO driving modes in the configured QEMU;
the IP information of the cluster nodes iNode based on the distributed storage is used for IP switching when an accessed IP fails, and the floating IP based on the distributed storage is used for actively performing floating IP switching when the nodes fail;
namely: the QEMU supports various configurations, fault tolerance is required for distributed storage, namely IP information of a plurality of cluster nodes iNode is required to be supported and configured, IP switching is carried out when an accessed IP fails, distributed storage node switching is realized, and single-point failure is avoided; meanwhile, the configuration of a distributed storage end floating IP is supported, and high availability is realized by switching the IP at the storage end;
when the node where the distributed storage floating IP is located fails, the floating IP can be actively switched to a normal distributed storage node, the client side of the QEMU directly accesses the floating IP, the distributed storage failure can not be sensed, and the cluster information can be normally acquired.
In the embodiment of the invention, an IO mechanism of distributed storage is moved upwards and moved to a client of a QEMU (QEMU) of a simulation processor for processing, so that the IO stability and the supporting fault redundancy of the client of the QEMU are increased;
the distributed storage copy distribution mechanism is moved up to the client side, copy distribution is carried out through the client side, multiple data distribution and data forwarding are avoided, the copy distribution mechanism is directly sent to a host side where IO is located, IO is directly carried out, and multiple memory copying is avoided;
moreover, because the distributed storage has more nodes, fault redundancy needs to be performed, so that an IO retransmission mechanism needs to be moved to a client, and when the client is abnormal and does not return, the IO retransmission mechanism is performed for multiple times of retransmission, so that more fault redundancy is supported.
In the embodiment of the invention, when the analog processor QEMU issues IO abnormity, the IO retry action is automatically executed, so that the IO success possibility is improved;
meanwhile, when the distributed storage system has a fault, checking the queue and the IO queue which are being requested, acquiring the IO information of the distributed storage which is being defended, and controlling to preferentially recover the block which is being accessed;
when the distributed storage system fails back, an IO retry action is performed on the IO being requested. The IO efficiency is improved, and the loss of IO performance in fault is reduced.
In this embodiment, data recovery is performed when a distributed storage fails, errors such as timeout occur when IO is sent according to old cluster information in a recovery process, and therefore a failure of a distributed storage end needs to be sensed, IO pending is performed when a failure occurs, IO issue is performed after data recovery of an object to be written is completed, and multiple retransmissions are avoided when a failure occurs;
the lost copy data needs to be recovered after the failure, but the data is out of order during the recovery, so a priority object recovery mechanism of the client needs to be supported, that is, the object of the target object which is issuing the IO is recovered as quickly as possible, and the IO timeout caused by the overlong IO pending time is avoided.
In the embodiment of the invention, the driver of the private protocol used by the QEMU is modified, the DHT algorithm of the distributed storage is adapted, the IO calculation of the distributed storage is moved upwards, and the moved IO calculation is transplanted to the driver of the QEMU, an IO high-availability and IO retransmission mechanism is realized in the driver of the QEMU, and the storage algorithm is modified, so that the client is supported to improve the priority of the data recovery object, the IO driver of the QEMU is enabled to issue a short IO path, and the IO performance of the distributed storage used by the IO QEMU is improved.
Fig. 3 is a block diagram of a distributed accelerated storage system based on a QEMU of an analog processor according to the present invention, and for convenience of illustration, only the parts related to the embodiment of the present invention are shown in the diagram.
The distributed acceleration storage system based on the QEMU of the analog processor comprises:
the copy position calculation module 11 is used for calculating the position of the copy at the client of the QEMU through a distributed hash table algorithm built in a driver of the QEMU;
the IO issuing module 12 is configured to issue IO to a host where the calculated copy is located through a driver of the QEMU according to the calculated position of the copy;
and the IO processing module 13 is configured to perform IO processing on the IO request sent by the driver of the received analog processor QEMU by the host where the copy is located, and feed an IO processing result back to the host where the IO is sent by the analog processor QEMU.
In this embodiment of the present invention, as shown in fig. 4, the IO issue module 12 specifically includes:
an information obtaining module 14, configured to obtain volume information of the IO issued by the QEMU and an offset distance value offset of the IO issued by the QEMU;
a target object information obtaining module 15, configured to calculate target object information of an issued IO according to the obtained volume information of the IO issued by the QEMU and the offset distance value offset of the issued IO;
an object location parameter obtaining module 16, configured to calculate, according to the calculated target object information, a target object location parameter by using the distributed hash table algorithm;
and the IO transfer module 17 is configured to transfer the IO to a host where multiple copies are located through a socket according to the copy number information of the volume and the target object location parameter.
As shown in fig. 3, the system further includes:
a first configuration module 18, configured to configure multiple IO driving modes in the QEMU, where the multiple IO driving modes support multiple corresponding IO delivery schemes;
a second configuration module 19, configured to configure IP information of multiple cluster nodes iNode based on distributed storage and configure a floating IP based on distributed storage according to multiple IO driving modes in the configured QEMU;
the IP information of the cluster nodes iNode based on the distributed storage is used for IP switching when an accessed IP fails, and the floating IP based on the distributed storage is used for actively performing floating IP switching when the nodes fail.
As shown in fig. 3, the system further includes:
the first IO retry module 20 is configured to automatically execute an IO retry action when the analog processor QEMU issues an IO exception;
the priority recovery control module 21 is configured to, when the distributed storage system fails, check a queue and an IO queue that are being requested, acquire IO information that is being defended in the distributed storage, and control to preferentially recover a block that is being accessed;
and a second IO retry module 22, configured to perform an IO retry action on the IO being requested when the distributed storage system is recovered from the failure.
The functions of the modules are described in the above embodiments, and are not described herein again.
In the embodiment of the invention, at the client side of the QEMU, the position of the copy is calculated by a distributed hash table algorithm built in a driver of the QEMU; according to the position of the copy obtained by calculation, IO (input/output) is issued to a host where the copy obtained by calculation is located through a driver of the QEMU; and the host where the copy is located carries out IO processing on the received IO request issued by the driver of the QEMU, and feeds the IO processing result back to the host where the IO is issued by the QEMU, so that the IO path of the distributed storage provided by the QEMU for accessing the distributed hash table algorithm is shortened, the IO availability of the distributed storage is improved, the time delay information of the IO is shortened, and the adaptability of the distributed storage based on the cloud computing scene is improved.
The above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the present invention, and they should be construed as being included in the following claims and description.

Claims (10)

1. A distributed accelerated storage method based on a QEMU (quantum execution unit) of an analog processor, which is characterized by comprising the following steps:
at the client side of the analog processor QEMU, calculating the position of the copy through a distributed hash table algorithm built in a driver of the analog processor QEMU;
according to the position of the copy obtained by calculation, IO (input/output) is issued to a host where the copy obtained by calculation is located through a driver of the QEMU;
and the host where the copy is located carries out IO processing on the received IO request issued by the driver of the QEMU, and feeds back the IO processing result to the host where the IO is issued by the QEMU.
2. The distributed accelerated storage method based on the QEMU of the analog processor of claim 1, wherein the step of issuing an IO to a host where the calculated copy is located through a driver of the QEMU according to the calculated position of the copy specifically includes the steps of:
acquiring volume information of IO issued by the QEMU of the analog processor and offset distance value offset of the issued IO;
calculating target object information of the issued IO according to the acquired volume information of the IO issued by the QEMU of the analog processor and the offset value offset of the issued IO;
calculating a target object position parameter by using the distributed hash table algorithm according to the calculated target object information;
and according to the copy number information of the volume and the object position parameter of the target object, transmitting the IO to a host where the multiple copies are located through a socket.
3. The method for QEMU-based distributed accelerated storage according to claim 1, further comprising the steps of:
configuring a plurality of IO driving modes in the QEMU, wherein the plurality of IO driving modes support a plurality of corresponding IO issuing schemes;
configuring IP information of a plurality of cluster nodes iNode based on distributed storage and configuring floating IP based on distributed storage according to a plurality of IO driving modes in the configured QEMU;
the IP information of the cluster nodes iNode based on the distributed storage is used for IP switching when an accessed IP fails, and the floating IP based on the distributed storage is used for actively performing floating IP switching when the nodes fail.
4. The method for QEMU-based distributed accelerated storage according to claim 1, further comprising the steps of:
and when the analog processor QEMU issues IO abnormity, automatically executing IO retry action.
5. The method for QEMU-based distributed accelerated storage according to claim 1, further comprising the steps of:
when the distributed storage system fails, checking a queue and an IO queue which are being requested, acquiring IO information of distributed storage which is being defended, and controlling to preferentially recover a block which is being accessed;
when the distributed storage system fails back, an IO retry action is performed on the IO being requested.
6. A QEMU-based distributed accelerated storage system, the system comprising:
the copy position calculation module is used for calculating the position of the copy at the client side of the QEMU through a distributed hash table algorithm built in a driver of the QEMU;
the IO issuing module is used for issuing IO to a host where the copy obtained by calculation is located through a driver of the QEMU according to the position of the copy obtained by calculation;
and the IO processing module is used for carrying out IO processing on the received IO request sent by the driver of the QEMU by the host where the copy is located and feeding back an IO processing result to the host where the IO is sent by the QEMU.
7. The distributed acceleration storage system based on analog processor QEMU of claim 6, characterized in that, the IO issue module specifically includes:
the information acquisition module is used for acquiring volume information of IO issued by the QEMU of the analog processor and offset distance value offset of the issued IO;
the target object information acquisition module is used for calculating target object information of the issued IO according to the acquired volume information of the IO issued by the QEMU and the offset distance value offset of the issued IO;
the object position parameter acquisition module is used for calculating object position parameters of the target object by utilizing the distributed hash table algorithm according to the calculated object information of the target object;
and the IO transmission module is used for transmitting the IO to a host where the plurality of copies are located through the socket according to the copy number information of the volume and the target object location parameter.
8. The QEMU-based distributed accelerated storage system of claim 6, further comprising:
the first configuration module is used for configuring a plurality of IO driving modes in the QEMU, wherein the plurality of IO driving modes support a plurality of corresponding IO issuing schemes;
the second configuration module is used for configuring IP information of a plurality of cluster nodes iNode based on distributed storage and configuring floating IP based on distributed storage according to a plurality of IO driving modes in the configured QEMU;
the IP information of the cluster nodes iNode based on the distributed storage is used for IP switching when an accessed IP fails, and the floating IP based on the distributed storage is used for actively performing floating IP switching when the nodes fail.
9. The QEMU-based distributed accelerated storage system of claim 6, further comprising:
and the first IO retry module is used for automatically executing IO retry actions when the analog processor QEMU issues IO abnormity.
10. The QEMU-based distributed accelerated storage system of claim 6, further comprising:
the priority recovery control module is used for checking the queue and the IO queue which are being requested when the distributed storage system fails, acquiring the IO information of distributed storage which is being defended and controlling the priority recovery of the accessed block;
and the second IO retry module is used for executing IO retry actions on the IO being requested when the failure of the distributed storage system is recovered.
CN202011302230.5A 2020-11-19 2020-11-19 Distributed accelerated storage method and system based on QEMU (quantum execution unit) Active CN112532701B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011302230.5A CN112532701B (en) 2020-11-19 2020-11-19 Distributed accelerated storage method and system based on QEMU (quantum execution unit)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011302230.5A CN112532701B (en) 2020-11-19 2020-11-19 Distributed accelerated storage method and system based on QEMU (quantum execution unit)

Publications (2)

Publication Number Publication Date
CN112532701A true CN112532701A (en) 2021-03-19
CN112532701B CN112532701B (en) 2022-07-15

Family

ID=74981710

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011302230.5A Active CN112532701B (en) 2020-11-19 2020-11-19 Distributed accelerated storage method and system based on QEMU (quantum execution unit)

Country Status (1)

Country Link
CN (1) CN112532701B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113254166A (en) * 2021-07-13 2021-08-13 云宏信息科技股份有限公司 Method for processing IO request, storage medium and virtualization simulator
CN116401020A (en) * 2023-06-07 2023-07-07 四川大学 KVM virtual machine I/O filter framework implementation method, system and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104050015A (en) * 2014-06-27 2014-09-17 国家计算机网络与信息安全管理中心 Mirror image storage and distribution system for virtual machines
US20150020071A1 (en) * 2013-07-12 2015-01-15 Bluedata Software, Inc. Accelerated data operations in virtual environments
CN106802839A (en) * 2016-12-13 2017-06-06 龚平 A kind of virtual-machine data guard method driven based on block
CN109327539A (en) * 2018-11-15 2019-02-12 上海天玑数据技术有限公司 A kind of distributed block storage system and its data routing method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150020071A1 (en) * 2013-07-12 2015-01-15 Bluedata Software, Inc. Accelerated data operations in virtual environments
CN104050015A (en) * 2014-06-27 2014-09-17 国家计算机网络与信息安全管理中心 Mirror image storage and distribution system for virtual machines
CN106802839A (en) * 2016-12-13 2017-06-06 龚平 A kind of virtual-machine data guard method driven based on block
CN109327539A (en) * 2018-11-15 2019-02-12 上海天玑数据技术有限公司 A kind of distributed block storage system and its data routing method

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113254166A (en) * 2021-07-13 2021-08-13 云宏信息科技股份有限公司 Method for processing IO request, storage medium and virtualization simulator
CN116401020A (en) * 2023-06-07 2023-07-07 四川大学 KVM virtual machine I/O filter framework implementation method, system and storage medium
CN116401020B (en) * 2023-06-07 2023-08-11 四川大学 KVM virtual machine I/O filter framework implementation method, system and storage medium

Also Published As

Publication number Publication date
CN112532701B (en) 2022-07-15

Similar Documents

Publication Publication Date Title
US10983860B2 (en) Automatic prefill of a storage system with conditioning of raid stripes
US10592161B1 (en) Storage system with flexible scanning supporting storage volume addition and/or recovery in asynchronous replication
US10846178B2 (en) Hash-based remote rebuild assistance for content addressable storage systems
US11327834B2 (en) Efficient computation of parity data in storage system implementing data striping
US11288286B2 (en) Storage system with data consistency checking in synchronous replication using active snapshot set
US11074129B2 (en) Erasure coded data shards containing multiple data objects
US11144399B1 (en) Managing storage device errors during processing of inflight input/output requests
US11494103B2 (en) Memory-efficient processing of RAID metadata bitmaps
CN112532701B (en) Distributed accelerated storage method and system based on QEMU (quantum execution unit)
US20190042103A1 (en) Resilient external memory
JP2017531857A (en) Distributed active hybrid storage system
US11531498B2 (en) Peer storage device messaging over control bus
US20200379899A1 (en) Methods for facilitating efficient storage operations using host-managed solid-state disks and devices thereof
CN111587420A (en) Method and system for rapid failure recovery of distributed storage system
US10951699B1 (en) Storage system with asynchronous messaging between processing modules for data replication
US11645174B2 (en) Recovery flow with reduced address lock contention in a content addressable storage system
WO2018080875A1 (en) Deduplication aware scalable content placement
US11144232B2 (en) Storage system with efficient snapshot pair creation during synchronous replication of logical storage volumes
US10324652B2 (en) Methods for copy-free data migration across filesystems and devices thereof
US11194501B2 (en) Standby copies withstand cascading fails
US11442894B2 (en) Methods for scalable file backup catalogs and devices thereof
US11216204B2 (en) Degraded redundant metadata, DRuM, technique
US11467906B2 (en) Storage system resource rebuild based on input-output operation indicator
US11481335B2 (en) Methods for using extended physical region page lists to improve performance for solid-state drives and devices thereof
US10996871B2 (en) Hash-based data recovery from remote storage system responsive to missing or corrupted hash digest

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