WO2014101197A1 - 一种存储资源调度方法及终端 - Google Patents

一种存储资源调度方法及终端 Download PDF

Info

Publication number
WO2014101197A1
WO2014101197A1 PCT/CN2012/088071 CN2012088071W WO2014101197A1 WO 2014101197 A1 WO2014101197 A1 WO 2014101197A1 CN 2012088071 W CN2012088071 W CN 2012088071W WO 2014101197 A1 WO2014101197 A1 WO 2014101197A1
Authority
WO
WIPO (PCT)
Prior art keywords
data object
priority
qos policy
storage system
storage
Prior art date
Application number
PCT/CN2012/088071
Other languages
English (en)
French (fr)
Inventor
万富元
张冬
崔文林
何凌智
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2012/088071 priority Critical patent/WO2014101197A1/zh
Priority to CN201280002837.9A priority patent/CN104040477A/zh
Publication of WO2014101197A1 publication Critical patent/WO2014101197A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input 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
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0668Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/0671In-line storage system
    • G06F3/0683Plurality of storage devices
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input 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
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0602Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/061Improving I/O performance
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input 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
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0628Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0655Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
    • G06F3/0659Command handling arrangements, e.g. command buffers, queues, command scheduling

Definitions

  • the present invention relates to the field of data processing technologies, and in particular, to a storage resource scheduling method and a terminal. Background technique
  • the storage resources (including the memory, the CPU, the disk processing capability, and so on) in the storage system are limited.
  • the QoS (Quality of Service) of a LUN may be damaging the QoS of other services. This requires reasonable planning and allocation of storage resources according to the characteristics of various services.
  • QoS includes IOPS, bandwidth, and delay for the storage system.
  • the QoS is controlled by using the partition or the LUN as the granularity.
  • the resource scheduling is performed according to the priority of each LUN to ensure that the service with higher priority can compete for the storage resource.
  • the storage system such as a disk array, calculates the priority of the LUN according to the number of times of requesting the LUN data, for example, to determine the priority of the LUN where the data is located, for example, to meet the access performance of the hotspot area, where the hotspot area is located.
  • the LUN sets a higher priority.
  • the manner in which the storage system directly sets the priority is likely to cause priority misjudgment, thereby affecting storage resource scheduling and reducing the processing performance of the storage system. Summary of the invention
  • a storage resource scheduling method and a terminal are provided, which can reduce the misjudgment of priorities, thereby reducing the adverse impact on storage resource scheduling.
  • the first aspect provides a storage resource scheduling method, including:
  • the host acquires the priority of the data object set by the user
  • the acquiring, by the host, a priority of a data object set by a user includes:
  • the host searches for setting information of a user in a tab attribute page of a storage attribute of the data object; determining a priority of the data object according to setting information in the tab attribute page.
  • the data object is a file or directory in a file system.
  • the sending the QoS policy of the data object to The storage system is specifically:
  • the QoS policy of the data object is sent to a record carrier of the storage system.
  • a storage resource scheduling apparatus including:
  • a priority determining module configured to acquire a priority of a data object set by a user
  • a search module configured to find a correspondence between a priority and a quality of service QoS policy
  • An analysis module configured to determine a QoS policy of the data object according to a correspondence between the priority and a QoS policy and a priority of the data object;
  • a transmission module configured to send the QoS policy of the data object to the storage system, so that the storage system performs storage resource scheduling on the data object according to the QoS policy of the data object.
  • the priority determining module includes: an information searching unit, configured to search for setting information of a user in a tab attribute page of a storage attribute of the data object;
  • an information determining unit configured to determine a priority of the data object according to the setting information in the tab attribute page.
  • the data object is a file or directory in a file system.
  • the transmission module is specifically configured to use the data
  • the object's QoS policy is sent to the record carrier of the storage system.
  • a storage resource scheduling system including a host and a storage system
  • the host is configured to obtain a priority of the data object set by the user; and search for a correspondence between the priority and the QoS policy; and determine, according to the correspondence between the priority and the QoS policy and the priority of the data object, a QoS policy of the data object; sending a QoS policy of the data object to the storage system
  • the storage system is configured to perform storage resource scheduling on the data object according to a QoS policy of the data object.
  • the host is specifically configured to search for setting information of a user in a tab attribute page of a storage attribute of the data object; according to the tab attribute page The setting information determines the priority of the data object.
  • the data object is a file or directory in a file system.
  • a terminal comprising a memory and a processor, the memory for storing computer instructions or code, the processor and the memory being coupled for executing computer instructions or code in the memory, Implement the following methods:
  • the embodiment of the present invention determines the priority of the data object set by the host, determines the QoS policy of the data object, and sends the QoS policy of the data object to the storage system, so that the storage system can use the data object according to the QoS policy.
  • Resource scheduling is implemented to implement linkage between the host and the storage system.
  • the method performs resource scheduling according to the data object priority set by the user, which more accurately reflects the user's will, compared with the prior art storage system.
  • the method of setting priorities and performing resource scheduling reduces the misjudgment of data object priorities, reduces the impact on storage resource scheduling caused by priority misjudgment, and improves the processing performance of the storage system.
  • FIG. 1 is a flowchart of a storage resource scheduling method according to an embodiment of the present invention
  • 2 is a flowchart of another storage resource scheduling method according to an embodiment of the present invention
  • FIG. 3 is a flowchart of a method for obtaining a priority of a file or a directory in a file system in the embodiment shown in FIG. 2;
  • FIG. 4 is a schematic structural diagram of a storage resource scheduling apparatus according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of another storage resource scheduling apparatus according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of a storage resource scheduling system according to an embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of a terminal according to an embodiment of the present invention.
  • FIG. 1 is a flowchart of a storage resource scheduling method according to an embodiment of the present invention.
  • the method can include:
  • Step 101 The host acquires a priority of a data object set by the user.
  • the agent running software can be installed in the host running the service.
  • the user configures the data object of the host to configure the storage attribute according to the requirements of the agent.
  • the storage attribute includes the priority of the data object.
  • the host obtains the priority of the data object, and then performs the next step.
  • the data object is a collection of data or data, which may be a LUN, a data in the LUN, or a file system.
  • a file or directory, etc. which is a collection of data under a storage path or a storage address, specifically all files under a folder.
  • the priority can have three levels: high, medium, and low, or multiple levels such as the first level and the second level. The specific level of the priority can be set as needed.
  • Step 102 Find a correspondence between the priority and the QoS policy.
  • a plurality of QoS policies and a correspondence between the priorities and the QoS policies are preset in the host. For example, the high priority corresponds to the first QoS policy, and the low priority corresponds to the second QoS policy.
  • the preset QoS policies in the host can be various, for example: controlling the CPU service time occupied by different priority data objects based on different priorities; managing the local page quota of the cache based on different priorities;
  • the priority is the control granularity, and the brush water level, the number of Req concurrency, the number of concurrent pages, and the like corresponding to each priority quota are controlled.
  • the data object corresponding to the different priority is triggered by the QoS time period, which may be High-priority data objects can preferentially trigger QoS policies for storage resource scheduling.
  • QoS control can be performed on low-priority data objects after QoS control on high-priority data objects.
  • Priority corresponding to the data object in the QoS time period The performance value, which may include IOPS (Input/Output Operations Per Second), bandwidth, and delay.
  • IOPS Input/Output Operations Per Second
  • bandwidth bandwidth
  • delay delay
  • a high-priority data object may have a QoS policy when it is triggered. Higher IOPS, greater bandwidth, smaller latency, etc.
  • the specific QoS policy can be set according to requirements.
  • the QoS policy is set according to the priority of the data object, and different priorities correspond to different policies.
  • Step 103 Determine a QoS policy of the data object according to the correspondence between the priority and the QoS policy and the priority of the data object.
  • the host may determine the QoS policy of the data object according to the priority of the data object. For example, the priority of the data object A is a high priority, and according to the correspondence between the priority and the QoS policy, Obtaining a first QoS policy corresponding to the data object A.
  • Step 104 Send a QoS policy of the data object to the storage system.
  • the QoS policy of the data object is transmitted to the storage system.
  • the storage system may be a disk array.
  • the host sends the first QoS policy corresponding to the data object A to the storage system.
  • the storage resource is scheduled according to the data object's QoS policy to provide differentiated services. For example, when the storage system performs storage resource scheduling on the data object A, the storage system performs storage resource scheduling on the data object A according to the corresponding first QoS policy of the data object A.
  • the storage system performs storage resource scheduling. For example, the disk array adjusts the resources occupied by the host data object on the disk array according to the QoS policy delivered by the host, including: CPU, 10 priority, CACHE resource, queue depth, and number of queues. Parameters such as concurrent size; When system resources are tight, policies such as low-priority traffic and concurrent Cache are used to ensure that important service quality is not affected.
  • the embodiment of the present invention determines the priority of the data object set by the host, determines the QoS policy of the data object, and sends the QoS policy of the data object to the storage system, so that the storage system can use the data object according to the QoS policy.
  • Resource scheduling is implemented to implement linkage between the host and the storage system.
  • the method performs resource scheduling according to the data object priority set by the user, which more accurately reflects the user's will, compared with the prior art storage system.
  • the method of setting priorities and performing resource scheduling reduces the misjudgment of data object priorities, reduces the impact on storage resource scheduling caused by priority misjudgment, and improves the processing performance of the storage system.
  • FIG. 2 is a flowchart of another storage resource scheduling method according to an embodiment of the present invention.
  • the method can include: Step 201: The host acquires a priority of a file or a directory in the file system.
  • the host is installed with agent software.
  • agent software For each file or directory in the host's file system, the agent software adds a Tab property page specifically developed for the storage system to its storage properties.
  • the user can right-click on the file or directory in the file system of the host through the agent software, and then select the "Properties" item in the pop-up storage property menu, and a property menu will pop up.
  • a tab property page for setting priorities was added when the agent software was not installed before.
  • the user can configure the priority of the file or directory on the host according to the tab attribute page, and control whether the file or the directory is preferentially processed as a key service, preferentially allocate resources, or delay processing and allocate resources as a general service.
  • the host obtains the priority of the file or directory in the file system by the following steps:
  • Step 301 The host searches for setting information of the user in the Tab property page of the storage attribute of the data object.
  • Step 302 Determine a priority of the data object according to the setting information in the tab attribute page.
  • the setting information may be a pre-agreed information corresponding to the priority. For example, if the setting information is High, the corresponding high priority, and if the setting information is Low, the low priority; the setting information may also be represented by a number such as 1, 0. , 1 corresponds to high priority, 0 corresponds to low priority.
  • the specific form of the setting information can be set as needed.
  • the host can obtain information similar to the following table, which is mainly used to identify the priority of data objects such as files or directories.
  • Step 202 The host searches for a correspondence between the priority and the QoS policy.
  • Step 203 The host determines a corresponding QoS policy of the file or the directory according to the priority of the file or the directory and the correspondence between the priority and the QoS policy.
  • This step 202 203 is similar to step 102 103 in the previous embodiment.
  • Step 204 Send the QoS policy of the file or directory to the record carrier of the storage system.
  • a record carrier is created in the storage system for recording a QoS policy of a file or directory delivered from the host agent software, and the record carrier may specifically be a configuration file in the storage system.
  • the host in this step, can also prioritize the file or directory and the QoS.
  • the policy is sent to the record carrier.
  • the host can send the QoS policy of the file or directory to the record carrier of the storage system in-band or out-of-band.
  • the storage system performs refined data management on the received 10 requests based on the information in the record carrier. Specifically, the storage system uses the corresponding search algorithm to quickly obtain the QoS policy of the corresponding file requested in the record carrier to differentiate, and achieve the goal of prioritizing the quality of service of the important service.
  • the storage system may also obtain a final control policy of the storage system according to a QoS policy of the corresponding file in the record carrier, and a related policy preset in the storage system, and then perform differentiated processing according to the final control policy.
  • the host can obtain the record carrier in the storage system through the proxy software to query the corresponding QoS policy configuration, thereby implementing interaction between the host and the storage system.
  • the storage system can also monitor the performance of the host. When the system resources are tight, the storage system can perform storage resource scheduling according to the received QoS policy. For example: Ensure the important services by suppressing the low-priority traffic and the concurrent Cache. Quality of service is not affected.
  • the QoS policy of the data object is sent to the storage system by the host, so that the storage system performs resource scheduling according to the policy, and the linkage between the host and the storage system is implemented, and the Tab is added in the file storage attribute menu.
  • the attribute page describes the priority attribute of the file or directory.
  • the host agent can identify the priority attribute and enable the storage system to implement QoS for a single file or directory by sending the QoS policy of the priority attribute to the storage system.
  • the method of the present embodiment eliminates the disadvantage that the traditional storage system cannot perform QoS control on the lower layer data object of the LI, as compared with the prior art.
  • the host may also monitor the change of the priority of the data object after setting the priority of the data object such as a file or a directory, for example, from intermediate to advanced, or from advanced.
  • the QoS policy of the data object is updated according to the change of the priority, and the QoS policy of the updated data object is sent to the storage system.
  • the host can perform real-time interaction with the storage system to achieve refined QoS control of the storage system for file or directory granularity, for example, fine QoS control for the LUN hotspot area instead of the entire LUN.
  • the data objects in the foregoing embodiments include a hotspot area that is not limited to applications, files, and LUNs.
  • the application may refer to a storage file corresponding to the application.
  • FIG. 4 is a schematic structural diagram of a storage resource scheduling apparatus according to an embodiment of the present invention.
  • the terminal can include:
  • the priority determining module 401 is configured to obtain a priority of a data object set by the user.
  • the searching module 402 is configured to find a correspondence between the priority and the QoS policy.
  • the analyzing module 403 is configured to determine a QoS policy of the data object according to a correspondence between a priority and a quality of service QoS policy and a priority of the data object.
  • the transmission module 404 is configured to send a QoS policy of the data object obtained by the analysis module 403 to the storage system, so that the storage system performs storage resource scheduling on the data object according to the QoS policy of the data object.
  • the embodiment of the present invention obtains the priority of the data object set by the user, determines the QoS policy of the data object, and sends the QoS policy of the data object to the storage system, so that the storage system can be based on the QoS policy of the data object.
  • the data object performs resource scheduling, and realizes linkage with the storage system.
  • the device performs resource scheduling according to the obtained data object priority set by the user, which more accurately reflects the user's will, compared with the prior art storage.
  • the system sets the priority and performs resource scheduling, which reduces the misjudgment of the data object priority, reduces the impact on the storage resource scheduling caused by the priority misjudgment, and improves the processing performance of the storage system. .
  • FIG. 5 is a schematic structural diagram of another storage resource scheduling apparatus according to an embodiment of the present invention.
  • the terminal may also include a priority determination module 501, a lookup module 502, an analysis module 503, and a transmission module 504.
  • the search module 502 and the analysis module 503 are similar to the search module 402 and the analysis module 403 in the foregoing embodiment, and are not described herein again.
  • the priority determining module 501 may further include:
  • An information searching unit 5011 configured to find setting information of a user in a tab attribute page of a storage attribute of the data object
  • the information determining unit 5012 is configured to determine a priority of the data object according to the setting information in the tab attribute page.
  • the data object may be a file or a directory in the file system.
  • the transmission module 504 is specifically configured to send the QoS policy determined by the analysis module 503 to the record carrier of the storage system.
  • the terminal may further include a monitoring module, configured to monitor a change in the priority of the data object in the priority determining module, for example, from intermediate to advanced, or from advanced to low, etc.
  • the analysis module updates the QoS policy corresponding to the priority of the data object according to the change of the priority, and the QoS policy of the updated data object is sent by the transmission module to the QoS policy.
  • the embodiment of the present invention not only delivers the QoS policy of the data object to the storage system through the foregoing unit, so that the storage system performs resource scheduling according to the policy, implements linkage with the storage system, and adds a tab in the file storage attribute menu.
  • the attribute page describes the priority attribute of the file or directory.
  • the host agent can identify the priority attribute and enable the storage system to implement QOS for a single file or directory by sending the QOS policy corresponding to the priority attribute to the storage system.
  • the device of the present embodiment eliminates the disadvantage that the traditional storage system cannot perform QoS control on the lower layer data objects of the LUN, in a manner that the QoS control is implemented in the prior art.
  • FIG. 6 is a schematic structural diagram of a storage resource scheduling system according to an embodiment of the present invention.
  • the network device can include a host 601 and a storage system 602.
  • the host 601 is configured to obtain a priority of the data object set by the user, and obtain a correspondence between the priority and the QoS policy, and determine the relationship according to the correspondence between the priority and the QoS policy and the priority of the data object.
  • a QoS policy of the data object; a QoS policy of the data object is sent to the storage system 602.
  • the storage system 602 is configured to perform storage resource scheduling on the data object according to a QoS policy of the data object.
  • the host 601 is specifically configured to find setting information of a user in a tab attribute page of a storage attribute of the data object; and determining a priority of the data object according to the setting information in the tab attribute page.
  • the data object is a file or a directory in a file system.
  • the storage system 602 may further include a receiving module, a performance monitoring module, and a decision module, where the receiving module is configured to receive a QoS policy of the data object sent by the host side, and the receiving module may further store the QoS policy of the data object to the storage system.
  • the performance monitoring module is used to monitor the performance of the host to provide real-time data to the decision module; the decision module is used to determine the QoS control of the current system according to the QoS policy received by the receiving module according to the condition monitored by the performance monitoring module. Points and specific control strategies, for example: When system resources are tight, the quality of service is not affected by curbing low-priority traffic and concurrent Cache.
  • the priority of the data object set by the user is obtained by the host, the QoS policy of the data object is determined, and the QoS policy of the data object is sent to the storage system, so that the storage system performs resources on the data object according to the QoS policy. Scheduling, the linkage between the host and the storage system is realized.
  • the network device performs resource scheduling according to the priority of the data object set by the user, which more accurately reflects the user's will, reduces the misjudgment of the priority of the data object, and reduces the misjudgment caused by the priority. Tune on storage resources The impact of the degree improves the processing performance of the storage system.
  • FIG. 7 a schematic structural diagram of a terminal according to an example of the present invention is shown.
  • the terminal can include a memory 701 and a processor 702.
  • the memory 701 is configured to store computer instructions or code, and the processor 702 is coupled to the memory 701 for executing computer instructions or code in the memory 701 to implement the following methods:
  • the processor may search for setting information of a user in a tab attribute page of a storage attribute of the data object; and determining, according to the setting information in the Tab attribute page, the priority of the data object. level.
  • the data object can be a file or directory in the file system.
  • the processor can also send the QoS policy of the data object to a record carrier of the storage system.
  • the disclosed systems, devices, and methods may be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to On multiple network elements. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the functions, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium.
  • the technical solution of the present invention which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) or a processor to perform all or part of the steps of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes: a U disk, a removable hard disk, a Read-Only Memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and the like, which can store program codes.

Abstract

公开一种存储资源调度方法及终端。一种存储资源调度方法,包括:主机获取用户设置的数据对象的优先级;查找优先级与服务质量QoS策略的对应关系;根据所述优先级与QoS策略的对应关系以及所述数据对象的优先级,确定所述数据对象的QoS策略;将所述数据对象的QoS策略发送至存储系统,以使所述存储系统根据所述数据对象的QoS策略对所述数据对象进行存储资源调度。本发明实施例减少了对数据对象优先级的误判,也减小了因优先级误判而产生的对存储资源调度的影响,提升了存储系统的处理性能。

Description

一种存储资源调度方法及终端
技术领域 本发明涉及数据处理技术领域, 尤其涉及一种存储资源调度方法及终端。 背景技术
存储系统中的存储资源 (包括内存、 CPU、 磁盘处理能力等)是有限的, 存在着 各种业务或者各个 LUN (logic unit number, 逻辑单元号) 之间的资源抢占情况, 在 保证某类业务或某个 LUN的业务的 QoS (Quality of Service, 服务质量) 的同时, 可 能就是在损害其它业务的 QoS,这就需要根据各种业务的特点来对存储资源合理的规 划和分配。 其中, QoS对于存储系统来说包括 IOPS、 带宽、 时延。
现有技术中是以分区或 LUN为粒度来进行 QoS控制的, 在存储系统分配 CPU 和内存资源时根据各个 LUN的优先级进行资源调度, 以保证优先级较高的业务在竞 争存储资源时能够具有相对的优势。 其中, 由存储系统如磁盘阵列根据对 LUN数据 的 10请求次数等相关参数进行计算, 来确定该数据所在的 LUN的优先级, 例如, 为 了满足对热点区域的访问性能,对该热点区域所在的 LUN设置较高的优先级。然而, 该由存储系统直接设置优先级的方式容易产生优先级误判, 进而影响存储资源调度, 降低存储系统的处理性能。 发明内容
本发明实施例中提供了一种存储资源调度方法及终端, 能够减少优先级的误判, 进而减小对存储资源调度的不利影响。
为了解决上述技术问题, 本发明实施例公开了如下技术方案:
第一方面, 提供一种存储资源调度方法, 包括:
主机获取用户设置的数据对象的优先级;
查找优先级与服务质量 QoS策略的对应关系;
根据所述优先级与 QoS策略的对应关系以及所述数据对象的优先级, 确定所述 数据对象的 QoS策略;
将所述数据对象的 QoS策略发送至存储系统, 以使所述存储系统根据所述数据 对象的 QoS策略对所述数据对象进行存储资源调度。 结合上述第一方面, 在第一种可能的实现方式中, 所述主机获取用户设置的数据 对象的优先级, 包括:
所述主机查找用户在所述数据对象的存储属性的 Tab属性页中的设置信息; 根据所述 Tab属性页中的设置信息确定所述数据对象的优先级。
结合上述第一方面, 和 /或第一种可能的实现方式, 在第二种可能的实现方式中, 所述数据对象为文件系统中的文件或目录。
结合上述第一方面,和 /或第一种可能的实现方式,和 /或第二种可能的实现方式, 在第三种可能的实现方式中, 所述将所述数据对象的 QoS策略发送至存储系统, 具 体为:
将所述数据对象的 QoS策略发送至存储系统的记录载体。
第二方面, 提供一种存储资源调度装置, 包括:
优先级确定模块, 用于获取用户设置的数据对象的优先级;
查找模块, 用于查找优先级与服务质量 QoS策略的对应关系;
分析模块, 用于根据所述优先级与 QoS策略的对应关系以及所述数据对象的优 先级, 确定所述数据对象的 QoS策略;
传输模块, 用于将所述数据对象的 QoS策略发送至存储系统, 以使所述存储系 统根据所述数据对象的 QoS策略对所述数据对象进行存储资源调度。
结合上述第二方面, 在第一种可能的实现方式中, 所述优先级确定模块包括: 信息查找单元,用于查找用户在所述数据对象的存储属性的 Tab属性页中的设置 信息;
信息确定单元,用于根据所述 Tab属性页中的设置信息确定所述数据对象的优先 级。
结合上述第二方面, 和 /或第一种可能的实现方式, 在第二种可能的实现方式中, 所述数据对象为文件系统中的文件或目录。
结合上述第二方面,和 /或第一种可能的实现方式,和 /或第二种可能的实现方式, 在第三种可能的实现方式中, 所述传输模块, 具体用于将所述数据对象的 QoS策略 发送至存储系统的记录载体。
第三方面, 还提供一种存储资源调度系统, 包括主机和存储系统,
所述主机, 用于获取用户设置的数据对象的优先级; 查找优先级与服务质量 QoS 策略的对应关系; 根据所述优先级与 QoS策略的对应关系以及所述数据对象的优先 级, 确定所述数据对象的 QoS策略; 将所述数据对象的 QoS策略发送至所述存储系 所述存储系统, 用于根据所述数据对象的 QoS策略对所述数据对象进行存储资 源调度。
结合上述第三方面, 在第一种可能的实现方式中, 所述主机, 具体用于查找用户 在所述数据对象的存储属性的 Tab属性页中的设置信息;根据所述 Tab属性页中的设 置信息确定所述数据对象的优先级。
结合上述第二方面, 和 /或第一种可能的实现方式, 在第二种可能的实现方式中, 所述数据对象为文件系统中的文件或目录。
第三方面, 还提供一种终端, 包括存储器和处理器, 所述存储器用于存储计算机 指令或代码,所述处理器和所述存储器耦合,用于执行所述存储器中的计算机指令或 代码, 实现以下方法:
获取用户设置的数据对象的优先级;
查找优先级与服务质量 QoS策略的对应关系;
根据所述优先级与 QoS策略的对应关系以及所述数据对象的优先级, 确定所述 数据对象的 QoS策略;
将所述数据对象的 QoS策略发送至存储系统, 以使所述存储系统根据所述数据 对象的 QoS策略对所述数据对象进行存储资源调度。
本发明实施例通过由主机获取用户设置的数据对象的优先级,确定该数据对象的 QoS策略, 并将该数据对象的 QoS策略发送至存储系统, 使得存储系统可以依据该 QoS策略对该数据对象进行资源调度, 实现了主机与存储系统之间的联动, 该方法根 据主机获取的用户设置的数据对象优先级来进行资源调度,更准确地体现了用户的意 愿,相比较现有技术中存储系统自行设定优先级并进行资源调度的方法,减少了对数 据对象优先级的误判, 也减小了因优先级误判而产生的对存储资源调度的影响,提升 了存储系统的处理性能。 附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现 有技术描述中所需要使用的附图作简单地介绍, 显而易见地, 下面描述中的附图仅仅 是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前 提下, 还可以根据这些附图获得其他的附图。
图 1为本发明实施例一种存储资源调度方法流程图; 图 2为本发明实施例另一种存储资源调度方法流程图;
图 3为图 2所示实施例中, 获得文件系统中文件或目录的优先级的方法流程图; 图 4为本发明实施例一种存储资源调度装置的结构示意图;
图 5为本发明实施例另一种存储资源调度装置的结构示意图;
图 6为本发明实施例一种存储资源调度系统的结构示意图;
图 7为本发明实施例一种终端的结构示意图。 具体实施方式 为了使本技术领域的人员更好地理解本发明实施例中的技术方案,并使本发明实 施例的上述目的、特征和优点能够更加明显易懂, 下面结合附图对本发明实施例中技 术方案作进一步详细的说明。
参见图 1, 为本发明实施例一种存储资源调度方法流程图。
该方法可以包括:
步骤 101, 主机获取用户设置的数据对象的优先级。
运行业务的主机中可以安装有代理软件,用户根据自身的需求通过该代理软件对 主机需要配置存储属性的数据对象进行配置, 该存储属性中包括数据对象的优先级。 用户配置完成后, 主机获取该数据对象的优先级, 然后执行下一步操作, 其中, 数据 对象是数据或数据的集合, 具体可以是 LUN, 也可以是 LUN中的数据, 还可以是文 件系统的文件或目录等, 该目录是存储路径或存储地址下的数据的集合, 具体可以是 文件夹下的所有文件。 优先级可以有高、 中、 低三个级别, 也可以有第一级、 第二级 等多个级别, 该优先级的具体级别可以根据需要设定。
步骤 102, 查找优先级与 QoS策略的对应关系。
主机中预先设置有多个 QoS策略以及优先级与 QoS策略的对应关系, 例如高优 先级对应第一 QoS策略, 低优先级对应第二 QoS策略。
主机中预先设置的 QoS策略可以有多种, 例如: 基于不同的优先级, 控制不同 优先级数据对象所占用的 CPU服务时间; 基于不同的优先级, 管理 cache的本地页 面配额; 亦或是以优先级为控制粒度, 控制各个优先级配额所对应的刷盘水位、 Req 并发数、 页面并发数等等, 再例如, 不同的优先级对应的数据对象被触发 QoS 的时 间段, 具体可以是对于高优先级的数据对象可以优先触发 QoS策略进行存储资源调 度, 对于低优先级的数据对象可以在对高优先级的数据对象进行 QoS控制之后再对 低优先级的数据对应进行 QoS控制等; 不同的优先级对应的数据对象在 QoS时间段 内的性能值, 该性能值可以包括 IOPS ( Input/Output Operations Per Second, 即每秒 进行读写操作的次数)、 带宽、 时延, 例如对于高优先级的数据对象在触发 QoS策略 时可以具有更高的 IOPS, 更大的带宽, 更小的时延等。
其中, 具体 QoS策略的制定可以根据需要设置, QoS策略根据数据对象的优先 级设定, 不同的优先级对应不同的策略。
步骤 103,根据优先级与 QoS策略的对应关系以及数据对象的优先级,确定数据 对象的 QoS策略。
主机在确定数据对象的优先级后,可以根据数据对象的优先级确定该数据对象的 QoS策略, 例如, 数据对象 A的优先级为高优先级, 则根据优先级与 QoS策略的对 应关系, 可以获得该数据对象 A对应的第一 QoS策略。
步骤 104, 将数据对象的 QoS策略发送至存储系统。
在主机确定该数据对象的 QoS策略后, 将该数据对象的 QoS策略传输至存储系 统, 该存储系统具体可以是磁盘阵列, 例如主机将数据对象 A对应的第一 QoS策略 发送至存储系统。
该存储系统在对数据对象进行存储资源调度时, 根据数据对象的 QoS策略对数 据对象进行存储资源调度, 以提供差异化服务。 例如, 存储系统在对数据对象 A进 行存储资源调度时, 根据数据对象 A的对应的第一 QoS策略对数据对象 A进行存储 资源调度。
该存储系统进行存储资源调度, 例如, 磁盘阵列根据主机下发的 QoS策略, 调 整主机的数据对象在磁盘阵列上所占用的资源, 包括: CPU、 10优先级、 CACHE资 源、 队列深度、 队列数量、 并发大小等参数; 当系统资源出现紧张, 通过遏制低优先 级的流量和 Cache下盘的并发等策略来保证重要业务服务质量不受影响。
本发明实施例通过由主机获取用户设置的数据对象的优先级,确定该数据对象的 QoS策略, 并将该数据对象的 QoS策略发送至存储系统, 使得存储系统可以依据该 QoS策略对该数据对象进行资源调度, 实现了主机与存储系统之间的联动, 该方法根 据主机获取的用户设置的数据对象优先级来进行资源调度,更准确地体现了用户的意 愿,相比较现有技术中存储系统自行设定优先级并进行资源调度的方法,减少了对数 据对象优先级的误判, 也减小了因优先级误判而产生的对存储资源调度的影响,提升 了存储系统的处理性能。
参见图 2, 为本发明实施例另一种存储资源调度方法流程图。
该方法可以包括: 步骤 201, 主机获取文件系统中文件或目录的优先级。
主机安装有代理软件, 该代理软件针对主机的文件系统中的每个文件或目录,在 其存储属性中增加一个专门针对存储系统开发的 Tab属性页。
用户可以通过该代理软件在主机的文件系统中,对文件或目录单击右键, 再在弹 出的存储属性菜单中选择 "属性"这一项, 会弹出一个属性菜单, 在该属性菜单中, 较之前未安装代理软件时多出了一个用于设置优先级的 Tab属性页。用户可以根据该 Tab属性页在主机配置此文件或目录的优先级, 用于控制文件或目录是作为重点业务 来优先处理、 优先分配资源, 还是作为一般业务来延迟处理、 少分配资源。
该 Tab属性页的开发过程与现有技术类似, 此处不再赘述。
具体的, 主机通过以下步骤获得文件系统中文件或目录的优先级:
步骤 301, 主机查找用户在数据对象的存储属性的 Tab属性页中的设置信息。 步骤 302, 根据该 Tab属性页中的设置信息确定该数据对象的优先级。
该设置信息可以是预先约定的与优先级对应的信息, 例如设置信息为 High, 则 对应高优先级, 设置信息为 Low, 则对应低优先级; 该设置信息也可以用 1、 0等数 字表示, 1对应高优先级, 0对应低优先级。 该设置信息的具体形式可以根据需要设 定。
通过上述两步骤, 主机可以获得类似下表的信息, 主要用于标识文件或目录等数 据对象的优先级。
数据对象 ID 数据对象名 优先级
0X00000001 文件 A High
0X00000002 文件 B Low
0X00000003 目录 1 High
步骤 202, 主机查找优先级与 QoS策略的对应关系。
步骤 203, 主机根据该文件或目录的优先级以及优先级与 QoS策略的对应关系, 确定该文件或目录的对应的 QoS策略。
该步骤 202 203与前述实施例中的步骤 102 103类似。
步骤 204, 将该文件或目录的 QoS策略发送至存储系统的记录载体。
存储系统中创建有记录载体,用于记录从主机代理软件下发的文件或目录的 QoS 策略等, 该记录载体具体可以是存储系统中的配置文件。
在另一实施例中, 该步骤中, 主机还可以同时将文件或目录的优先级以及 QoS 策略发送至该记录载体中。
具体的, 主机可以将文件或目录的 QoS策略通过带内或带外发送至该存储系统 的记录载体中。
存储系统会根据该记录载体中的信息,对所收到的 10请求做精细化的数据管理。 具体的, 存储系统通过相应的查找算法, 在记录载体中快速获取到 10请求的相应文 件的的 QoS策略进行差异化处理, 达到对优先保证重要业务的服务质量的目标。
存储系统也可以根据记录载体中该相应文件的 QoS策略, 结合存储系统中预设 的相关策略得出存储系统最终的控制策略,然后再根据最终的控制策略进行差异化处 理。
另外, 主机可通过代理软件获取存储系统中的记录载体, 以查询相应的 QoS策 略配置, 从而实现主机与存储系统间的交互。
存储系统也可以监控主机的性能情况, 当系统资源出现紧张时,存储系统可以根 据接收到的 QoS策略进行存储资源调度, 例如: 通过遏制低优先级的流量和 Cache 下盘的并发来保证重要业务服务质量不受影响。
本发明实施例不仅通过主机向存储系统下发数据对象的 QoS策略, 使得存储系 统依据该策略进行资源调度, 实现了主机与存储系统之间的联动, 而且, 通过在文件 存储属性菜单中添加 Tab属性页描述文件或目录的优先级属性,主机代理程序可以识 别该优先级属性, 并通过将该优先级属性的 QoS策略下发至存储系统, 可以让存储 系统实现对单个文件或目录的进行 QoS控制, 以达到精细化 QoS控制的目的, 相对 于现有技术中对整个 LUN设置优先级的方式, 本实施例方法消除了传统的存储系统 无法对 LI 的下层数据对象进行 QoS控制的缺点。
在本发明的另一实施例中,该主机还可以在设定文件或目录等数据对象的优先级 后, 监控数据对象的优先级的变化情况, 例如, 从中级升高为高级, 或从高级降为低 级等, 并根据优先级的变化情况更新该数据对象的 QoS策略, 并将更新后的数据对 象的 QoS策略下发至存储系统。
上述实施例中, 主机可以与存储系统进行实时交互, 达成存储系统针对文件或目 录粒度的精细化 QoS控制,例如,对 LUN热点区域而不是整个 LUN进行精细化 QoS 控制。
以上各实施例中的数据对象包括并不限于应用程序、 文件、 LUN的热点区域等, 其中, 应用程序可以是指该应用程序对应的存储文件。
以上是对本发明方法实施例的描述, 下面对实现上述方法的装置进行介绍。 参见图 4, 为本发明实施例一种存储资源调度装置的结构示意图。
该终端可以包括:
优先级确定模块 401, 用于获取用户设置的数据对象的优先级。
查找模块 402, 用于查找优先级与 QoS策略的对应关系。
分析模块 403,用于根据优先级与服务质量 QoS策略的对应关系以及所述数据对 象的优先级, 确定所述数据对象的 QoS策略。
传输模块 404,用于将所述分析模块 403获得的数据对象的 QoS策略发送至存储 系统, 以使所述存储系统根据所述数据对象的 QoS策略对所述数据对象进行存储资 源调度。
本发明实施例通过上述单元获取用户设置的数据对象的优先级,确定该数据对象 的 QoS策略, 并将该数据对象的 QoS策略发送至存储系统, 使得存储系统可以依据 该数据对象的 QoS策略对该数据对象进行资源调度, 实现了与存储系统之间的联动, 该装置根据获取的用户设置的数据对象优先级来进行资源调度,更准确地体现了用户 的意愿,相比较现有技术中存储系统自行设定优先级并进行资源调度的方法,减少了 对数据对象优先级的误判, 也减小了因优先级误判而产生的对存储资源调度的影响, 提升了存储系统的处理性能。
参见图 5, 为本发明实施例另一种存储资源调度装置的结构示意图。
该终端也可以包括优先级确定模块 501、 查找模块 502, 分析模块 503和传输模 块 504。 查找模块 502、 分析模块 503与前述实施例中的查找模块 402、 分析模块 403 类似, 此处不再赘述。
其中, 优先级确定模块 501可以进一步包括:
信息查找单元 5011, 用于查找用户在所述数据对象的存储属性的 Tab属性页中 的设置信息;
信息确定单元 5012, 用于根据所述 Tab属性页中的设置信息确定所述数据对象 的优先级。 其中, 所述数据对象可以为文件系统中的文件或目录等。
传输模块 504,具体用于将分析模块 503确定的 QoS策略发送至存储系统的记录 载体。
在另一实施例中, 该终端还可以包括监控模块,用于监控优先级确定模块中数据 对象的优先级的变化情况, 例如, 从中级升高为高级, 或从高级降为低级等, 在监控 模块监测到数据对象的优先级变化后,分析模块根据优先级的变化情况更新该数据对 象的优先级对应的 QoS策略, 并由传输模块将更新后的数据对象的 QoS策略下发至 存储系统。
本发明实施例不仅通过上述单元向存储系统下发数据对象的 QoS策略, 使得存 储系统依据该策略进行资源调度, 实现了与存储系统之间的联动, 而且, 通过在文件 存储属性菜单中添加 Tab属性页描述文件或目录的优先级属性,主机代理程序可以识 别该优先级属性, 并通过将该优先级属性对应的 QOS策略下发至存储系统, 可以让 存储系统实现对单个文件或目录的 QOS控制以达到精细化 QoS控制的目的, 相对于 现有技术中对整个 LUN设置优先级的方式, 本实施例装置消除了传统的存储系统无 法对 LUN的下层数据对象进行 QoS控制的缺点。
参见图 6, 为本发明实施例一种存储资源调度系统的结构示意图。
该网络设备可以包括主机 601和存储系统 602。
其中, 主机 601, 用于获取用户设置的数据对象的优先级; 查找优先级与服务质 量 QoS策略的对应关系; 根据优先级与 QoS策略的对应关系以及所述数据对象的优 先级, 确定所述数据对象的 QoS策略; 将所述数据对象的 QoS策略发送至所述存储 系统 602。
存储系统 602, 用于根据数据对象的 QoS策略对所述数据对象进行存储资源调 度。
该主机 601, 具体用于查找用户在所述数据对象的存储属性的 Tab属性页中的设 置信息; 根据所述 Tab属性页中的设置信息确定所述数据对象的优先级。其中, 所述 数据对象为文件系统中的文件或目录。
该存储系统 602可以进一步包括接收模块、 性能监控模块和决策模块, 其中, 接 收模块用于接收主机侧发送的数据对象的 QoS策略, 该接收模块可以进一步将该数 据对象的 QoS策略存储至存储系统的记录载体中; 性能监控模块用于监控主机性能 情况, 以给决策模块提供实时数据; 决策模块用于根据性能监控模块监控到的情况, 根据接收模块接收到的 QoS策略决策当前系统的 QoS控制点和具体的控制策略, 例 如: 当系统资源出现紧张,通过遏制低优先级的流量和 Cache下盘的并发来保证重要 业务服务质量不受影响。
本发明实施例通过主机获取用户设置的数据对象的优先级, 确定该数据对象的 QoS策略, 并将该数据对象的 QoS策略发送至存储系统, 使得存储系统依据该 QoS 策略对该数据对象进行资源调度, 实现了主机与存储系统之间的联动。该网络设备根 据主机获取的用户设置的数据对象优先级来进行资源调度,更准确地体现了用户的意 愿,减少了对数据对象优先级的误判, 也减小了因优先级误判而产生的对存储资源调 度的影响, 提升了存储系统的处理性能。
参见图 7, 为本发明实例一种终端的结构示意图。
该终端可以包括存储器 701和处理器 702.
存储器 701用于存储计算机指令或代码,所述处理器 702和所述存储器 701耦合, 用于执行所述存储器 701中的计算机指令或代码, 实现以下方法:
获取用户设置的数据对象的优先级;
查找优先级与服务质量 QoS策略的对应关系;
根据优先级与 QoS策略的对应关系以及所述数据对象的优先级, 确定所述数据 对象的 QoS策略;
将所述数据对象的 QoS策略发送至存储系统, 以使所述存储系统根据所述数据 对象的 QoS策略对所述数据对象进行存储资源调度。
在本发明的另一实施例中,该处理器可以查找用户在所述数据对象的存储属性的 Tab属性页中的设置信息; 根据所述 Tab属性页中的设置信息确定所述数据对象的优 先级。 该数据对象可以为文件系统中的文件或目录。
在另一实施例中, 该处理器还可以将所述数据对象的 QoS策略发送至存储系统 的记录载体。
本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单 元及算法步骤, 能够以电子硬件、 或者计算机软件和电子硬件的结合来实现。这些功 能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专 业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实 现不应认为超出本发明的范围。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统、 装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本申请所提供的几个实施例中, 应该理解到, 所揭露的系统、 装置和方法, 可 以通过其它的方式实现。 例如, 以上所描述的装置实施例仅仅是示意性的, 例如, 所 述单元的划分, 仅仅为一种逻辑功能划分, 实际实现时可以有另外的划分方式, 例如 多个单元或组件可以结合或者可以集成到另一个系统, 或一些特征可以忽略, 或不执 行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些 接口, 装置或单元的间接耦合或通信连接, 可以是电性, 机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显 示的部件可以是或者也可以不是物理单元, 即可以位于一个地方, 或者也可以分布到 多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例 方案的目的。
另外,在本发明各个实施例中的各功能单元可以集成在一个处理单元中, 也可以 是各个单元单独物理存在, 也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以 存储在一个计算机可读取存储介质中。基于这样的理解,本发明的技术方案本质上或 者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现 出来, 该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机 设备 (可以是个人计算机, 服务器, 或者网络设备等) 或处理器 (processor)执行本 发明各个实施例所述方法的全部或部分步骤。 而前述的存储介质包括: U盘、移动硬 盘、只读存储器(ROM, Read-Only Memory ) 随机存取存储器(RAM, Random Access Memory), 磁碟或者光盘等各种可以存储程序代码的介质。
以上所述, 仅为本发明的具体实施方式, 但本发明的保护范围并不局限于此, 任 何熟悉本技术领域的技术人员在本发明揭露的技术范围内, 可轻易想到变化或替换, 都应涵盖在本发明的保护范围之内。因此,本发明的保护范围应所述以权利要求的保 护范围为准。

Claims

权 利 要 求
1、 一种存储资源调度方法, 其特征在于, 包括:
主机获取用户设置的数据对象的优先级;
查找优先级与服务质量 QoS策略的对应关系;
根据所述优先级与 QoS策略的对应关系以及所述数据对象的优先级, 确定 所述数据对象的 QoS策略;
将所述数据对象的 QoS策略发送至存储系统, 以使所述存储系统根据所述 数据对象的 QoS策略对所述数据对象进行存储资源调度。
2、 根据权利要求 1所述的方法, 其特征在于, 所述主机获取用户设置的数 据对象的优先级, 包括: 所述主机查找用户在所述数据对象的存储属性的 Tab属性页中的设置信息; 根据所述 Tab属性页中的设置信息确定所述数据对象的优先级。
3、 根据权利要求 1或 2所述的方法, 其特征在于, 所述数据对象为文件系 统中的文件或目录。
4、 根据权利要求 1至 3中任意一项所述的方法, 其特征在于, 所述将所述 数据对象的 QoS策略发送至存储系统, 具体为: 将所述数据对象的 QoS策略发送至存储系统的记录载体。
5、 一种存储资源调度装置, 其特征在于, 包括: 优先级确定模块, 用于获取用户设置的数据对象的优先级; 查找模块, 用于查找优先级与服务质量 QoS策略的对应关系; 分析模块, 用于根据所述优先级与 QoS策略的对应关系以及所述数据对象 的优先级, 确定所述数据对象的 QoS策略; 传输模块, 用于将所述数据对象的 QoS策略发送至存储系统, 以使所述存 储系统根据所述数据对象的 QoS策略对所述数据对象进行存储资源调度。
6、 根据权利要求 5所述的装置, 其特征在于, 所述优先级确定模块包括: 信息查找单元,用于查找用户在所述数据对象的存储属性的 Tab属性页中的 设置信息; 信息确定单元,用于根据所述 Tab属性页中的设置信息确定所述数据对象的 优先级。
7、 根据权利要求 5或 6所述的装置, 其特征在于, 所述数据对象为文件系 统中的文件或目录。
8、 根据权利要求 5至 7中任意一项所述的装置, 其特征在于, 所述传输模块, 具体用于将所述数据对象的 QoS策略发送至存储系统的记 录载体。
9、 一种存储资源调度系统, 其特征在于, 包括主机和存储系统, 所述主机, 用于获取用户设置的数据对象的优先级; 查找优先级与服务质量 QoS策略的对应关系; 根据所述优先级与 QoS策略的对应关系以及所述数据对 象的优先级, 确定所述数据对象的 QoS策略; 将所述数据对象的 QoS策略发送 至所述存储系统; 所述存储系统, 用于根据所述数据对象的 QoS策略对所述数据对象进行存 储资源调度。
10、 根据权利要求 9所述的系统, 其特征在于,
所述主机,具体用于查找用户在所述数据对象的存储属性的 Tab属性页中的 设置信息; 根据所述 Tab属性页中的设置信息确定所述数据对象的优先级。
11、 根据权利要求 9或 10所述的系统, 其特征在于, 所述数据对象为文件 系统中的文件或目录。
12、 一种终端, 其特征在于, 包括存储器和处理器, 所述存储器用于存储计 算机指令或代码, 所述处理器和所述存储器耦合, 用于执行所述存储器中的计算 机指令或代码, 实现以下方法:
获取用户设置的数据对象的优先级;
查找优先级与服务质量 QoS策略的对应关系;
根据所述优先级与 QoS策略的对应关系以及所述数据对象的优先级, 确定 所述数据对象的 QoS策略;
将所述数据对象的 QoS策略发送至存储系统, 以使所述存储系统根据所述 数据对象的 QoS策略对所述数据对象进行存储资源调度。
PCT/CN2012/088071 2012-12-31 2012-12-31 一种存储资源调度方法及终端 WO2014101197A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2012/088071 WO2014101197A1 (zh) 2012-12-31 2012-12-31 一种存储资源调度方法及终端
CN201280002837.9A CN104040477A (zh) 2012-12-31 2012-12-31 一种存储资源调度方法及终端

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2012/088071 WO2014101197A1 (zh) 2012-12-31 2012-12-31 一种存储资源调度方法及终端

Publications (1)

Publication Number Publication Date
WO2014101197A1 true WO2014101197A1 (zh) 2014-07-03

Family

ID=51019797

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/088071 WO2014101197A1 (zh) 2012-12-31 2012-12-31 一种存储资源调度方法及终端

Country Status (2)

Country Link
CN (1) CN104040477A (zh)
WO (1) WO2014101197A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106126121A (zh) * 2016-06-22 2016-11-16 广东睿江云计算股份有限公司 一种基于qos的云主机磁盘资源保障方法及系统
CN107577428A (zh) * 2017-08-31 2018-01-12 郑州云海信息技术有限公司 一种数据存储方法及装置
CN115469817A (zh) * 2022-11-03 2022-12-13 广东睿江云计算股份有限公司 提升磁盘QoS的方法、装置、计算机设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1833232A (zh) * 2003-04-23 2006-09-13 国际商业机器公司 用于运行时数据管理的存储系统类区别提示
US20060253621A1 (en) * 2005-05-04 2006-11-09 Brewer Michael A Quality of service for data storage volumes
CN102426552A (zh) * 2011-10-31 2012-04-25 成都市华为赛门铁克科技有限公司 存储系统服务质量控制方法、装置和系统
CN102799487A (zh) * 2012-06-29 2012-11-28 记忆科技(深圳)有限公司 基于阵列/lun的io调度方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1833232A (zh) * 2003-04-23 2006-09-13 国际商业机器公司 用于运行时数据管理的存储系统类区别提示
US20060253621A1 (en) * 2005-05-04 2006-11-09 Brewer Michael A Quality of service for data storage volumes
CN102426552A (zh) * 2011-10-31 2012-04-25 成都市华为赛门铁克科技有限公司 存储系统服务质量控制方法、装置和系统
CN102799487A (zh) * 2012-06-29 2012-11-28 记忆科技(深圳)有限公司 基于阵列/lun的io调度方法及装置

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106126121A (zh) * 2016-06-22 2016-11-16 广东睿江云计算股份有限公司 一种基于qos的云主机磁盘资源保障方法及系统
CN107577428A (zh) * 2017-08-31 2018-01-12 郑州云海信息技术有限公司 一种数据存储方法及装置
CN115469817A (zh) * 2022-11-03 2022-12-13 广东睿江云计算股份有限公司 提升磁盘QoS的方法、装置、计算机设备及存储介质

Also Published As

Publication number Publication date
CN104040477A (zh) 2014-09-10

Similar Documents

Publication Publication Date Title
JP7060724B2 (ja) タスクスケジューリング方法、リソース共有使用方法、スケジューラ、コンピュータ可読記憶媒体および装置
EP2618261B1 (en) Qos control method and system for storage system
US7493419B2 (en) Input/output workload fingerprinting for input/output schedulers
US20190163371A1 (en) Next generation storage controller in hybrid environments
US20190319844A1 (en) Service deployment method and apparatus
US8315992B1 (en) Affinity based allocation for storage implementations employing deduplicated data stores
US8832218B2 (en) Determining priorities for cached objects to order the transfer of modifications of cached objects based on measured network bandwidth
WO2013029487A1 (zh) 一种资源分配方法及资源管理平台
WO2015067117A1 (zh) 一种云上传方法及系统、调度设备、客户端
US20120303905A1 (en) Method and apparatus for implementing cache
JP4407431B2 (ja) 計算機システム及び計算機システムの帯域制御方法
US8180730B2 (en) Arbitration token for managing data integrity and data accuracy of information services that utilize distributed data replicas
US9424314B2 (en) Method and apparatus for joining read requests
JP6246923B2 (ja) 管理サーバ、計算機システム及び方法
WO2012103231A1 (en) Computing platform with resource constraint negotiation
WO2011144093A2 (zh) 队列调度的方法和装置
US8793334B1 (en) Network-attached storage (NAS) bandwidth manager
US20180026855A1 (en) Network management system, device, and method
KR20170046786A (ko) 리소스 관리 방법, 호스트 및 엔드포인트
WO2012109961A1 (zh) 一种分配浏览器进程的方法及其设备
WO2014101197A1 (zh) 一种存储资源调度方法及终端
WO2014180443A1 (zh) 系统资源均衡调整方法及装置
US10846265B2 (en) Method and apparatus for accessing file, and storage system
US20220124151A1 (en) Task allocation among devices in a distributed data storage system
CN114640630A (zh) 一种流量管控方法、装置、设备及可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12890672

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12890672

Country of ref document: EP

Kind code of ref document: A1