CN116501712A - Fine-grained storage service quality control implementation method - Google Patents
Fine-grained storage service quality control implementation method Download PDFInfo
- Publication number
- CN116501712A CN116501712A CN202310498183.3A CN202310498183A CN116501712A CN 116501712 A CN116501712 A CN 116501712A CN 202310498183 A CN202310498183 A CN 202310498183A CN 116501712 A CN116501712 A CN 116501712A
- Authority
- CN
- China
- Prior art keywords
- sliding window
- quality
- service
- qos
- directory
- 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
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000003908 quality control method Methods 0.000 title claims abstract description 14
- 230000008569 process Effects 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 4
- 239000010410 layer Substances 0.000 description 16
- 239000012792 core layer Substances 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 241001362551 Samba Species 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
- G06F16/1824—Distributed file systems implemented using Network-attached Storage [NAS] architecture
- G06F16/183—Provision of network file services by network file servers, e.g. by using NFS, CIFS
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention provides a fine-grained storage service quality control implementation method, which comprises the following steps: the QoS control module captures the catalog operation requests, searches the catalog ID of each catalog operation request, searches the QoS strategy of the corresponding catalog, determines the sliding window related to the searched QoS strategy, enables the catalog operation requests to sequentially pass through the related sliding window, and then executes the catalog operation requests. The invention provides a fine-grained storage service quality control realization method, which combines a bottom file system, utilizes a sliding window technology to realize the bandwidth and iops precise control of directory resources, avoids the excessively high use of a certain directory resource, simultaneously avoids high concurrent instantaneous pressure and improves the use experience of the file system.
Description
Technical Field
The invention belongs to the technical field of network file storage, and particularly relates to a fine-grained storage service quality control implementation method.
Background
With the continuous development of network file systems, the demands for network file storage are continuously increased, multiple network file systems are isolated from each other, and the resource occupation of one directory is often too high, so that the access to other directories is blocked.
Disclosure of Invention
Aiming at the defects existing in the prior art, the invention provides a fine-grained storage service quality control implementation method which can effectively solve the problems.
The technical scheme adopted by the invention is as follows:
the invention provides a fine-grained storage service quality control implementation method, which comprises the following steps:
step 1, setting a service quality strategy file in a file system layer; the service quality strategy file is used for storing the service quality strategy of each catalogue in the file system; the quality of service policy includes a maximum limit of n quality of service items; wherein, n kinds of service quality items are respectively expressed as: a 1 st quality of service item, a 2 nd quality of service item, …, an n-th quality of service item;
step 2, in the file system layer, sliding windows corresponding to each service quality item are respectively set, so n sliding windows are respectively set: a 1 st sliding window corresponding to the 1 st quality of service item, a 2 nd sliding window corresponding to the 2 nd quality of service item, …, an nth sliding window corresponding to the nth quality of service item;
for any i-th sliding window, i=1, 2, …, n, a timer is used to control the total storage resource of the sliding window, namely: when the timing time of the timer is reached, namely, a sliding window time period is reached, the total storage resource of the sliding window is allocated to the ith sliding window, and the total storage resource of the sliding window is the same as the maximum limit value of the corresponding ith service quality item;
step 3, setting a quality of service Qos control module at a file system layer, wherein the quality of service Qos control module captures directory operation requests for all directories in the file system, and each directory operation request carries a directory ID, a service quality item requested at the time and a storage resource request value at the time;
the QoS control module queues the captured catalogue operation requests according to time sequence to form a total waiting queue;
step 4, the QoS control module judges whether to turn on the QoS switch currently, if not, the flow is ended; if the device is started, executing the step 5;
step 5, the QoS control module reads the catalog operation request of the queue head from the total waiting queue according to a set time interval to obtain catalog ID; step 6 is executed for each directory ID read;
step 6, the QoS control module executes step 6.1-step 6.2 for each directory ID read;
step 6.1, the QoS control module searches the QoS strategy file according to the catalog ID, judges whether the QoS strategy of the catalog corresponding to the catalog ID exists in the QoS strategy file, if not, directly executes the catalog operation request, and ends the flow; if so, executing the step 6.2;
step 6.2, the QoS control module determines sliding windows related to the searched QoS policies according to the searched QoS policies of the catalogs corresponding to the catalog IDs, and presumes m related sliding windows, wherein m is less than or equal to n;
the QoS control module enables the catalogue operation request to sequentially pass through m related sliding windows, then executes the catalogue operation request, and ends the flow;
when the directory operation request passes through any one of m related sliding windows, the sliding window is set to be a j-th sliding window, wherein j=1, 2, …, m, and the method comprises the following steps:
the QoS control module judges whether the residual storage resources of the jth sliding window in the current sliding window time period are larger than or equal to the storage resource request value of the directory operation request, if so, the directory operation request is executed, the storage resource request value of the directory operation request is subtracted by the residual storage resources of the jth sliding window in the current sliding window time period, the residual storage resources of the jth sliding window in the current sliding window time period are used as the residual storage resources of the jth sliding window in the current sliding window time period, and the process is ended; if not, adding the catalog operation request into a j sliding window queue corresponding to the j sliding window, and waiting for a timer to time to the next sliding window time period;
when the time period of the next sliding window is reached, the jth sliding window reacquires the total storage resources of the sliding window, the jth sliding window queue is activated, and all the directory operation requests in the jth sliding window queue are sequentially processed until the directory operation request is processed until the current directory operation request is processed.
Preferably, the service quality policy file adopts a hash table form, a key of the hash table is a directory id, and a value is a service quality policy.
Preferably, the quality of service item includes a number of reads per second, a number of writes per second, a number of reads per second, a read bandwidth value per second, a write bandwidth value per second, and a read write bandwidth value per second.
Preferably, when the directory operation request is a read request, the sliding window corresponding to the number of times of reading per second, the sliding window corresponding to the number of times of reading and writing per second, the sliding window corresponding to the read bandwidth value per second and the sliding window corresponding to the read bandwidth value per second are needed to pass through;
when the directory operation request is a write request, the sliding window corresponding to the write times per second, the sliding window corresponding to the read/write times per second, the sliding window corresponding to the write bandwidth value per second and the sliding window corresponding to the read/write bandwidth value per second are needed to pass through.
Preferably, when the timing time of the timer is reached, that is, when a sliding window time period is reached, the total storage resource of the sliding window is allocated to the ith sliding window, specifically:
when the time starting point of the sliding window time period is reached, the total storage resources of the sliding window are distributed to the ith sliding window at once;
or alternatively
And when the sliding window time period is reached, gradually distributing the total storage resources of the sliding window to the ith sliding window in the sliding window time period.
Preferably, the quality of service Qos control module is disposed in a kernel layer.
The fine-grained storage service quality control implementation method provided by the invention has the following advantages:
the invention provides a fine-grained storage service quality control realization method, which combines a bottom file system, utilizes a sliding window technology to realize the bandwidth and iops precise control of directory resources, avoids the excessively high use of a certain directory resource, simultaneously avoids high concurrent instantaneous pressure and improves the use experience of the file system.
Drawings
Fig. 1 is a schematic flow chart of a fine-grained storage service quality control implementation method provided by the invention;
FIG. 2 is a block diagram of a fine-grained storage QoS control implementation provided by the present invention;
fig. 3 is a flowchart of updating a qos policy file provided in the present invention.
Detailed Description
In order to make the technical problems, technical schemes and beneficial effects solved by the invention more clear, the invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The invention provides a fine-grained storage service quality control realization method, which combines a bottom file system, utilizes a sliding window technology to realize the bandwidth and iops precise control of directory resources, avoids the excessively high use of a certain directory resource, simultaneously avoids high concurrent instantaneous pressure and improves the use experience of the file system.
Noun interpretation:
iops: number of reads/writes per second.
qos: quality of service for bandwidth and iops control.
Quality of service policy file: a qos rule file is recorded, including bandwidth and iops set rules.
Bandwidth: the size of the data traffic per unit time.
proc file: the memory file does not occupy the space of a hard disk, and is a way for the communication between the application layer and the kernel layer.
The invention provides a fine-grained storage service quality control implementation method, the architecture of which is shown in figure 2, comprising the following steps:
1) A plurality of clients: the invention supports various clients such as mac, windows, linux and various protocols such as nfs, cifs and the like.
2) Network: the client and the Nas Server interact through a network.
3) Nas Server Server: the entire shared back-end storage server.
4) Nfs, samba server: and analyzing the read-write operation request sent by the client to perform corresponding read-write operation.
5) Quality of service Qos control module: under the condition of guaranteeing service bandwidth and iops, storage resources are reasonably allocated, preemption of application services on the resources is effectively relieved or controlled, flow monitoring and reasonable resource allocation are realized, and main functions are realized in a specific file system layer.
6) Web interface: qos configuration and presentation are carried out through a web interface, and the qos configuration and presentation mainly comprise read-write iops and read-write bandwidth.
Referring to fig. 3, the invention uses the ioctl interface function to implement the notification mechanism of the kernel and the application layer, and uses the Qos control module of the kernel layer to analyze the Qos policy file to implement the communication between the kernel and the application layer. The QoS control module of the inner core layer analyzes the QoS strategy file, so that the occupation of the application layer caused by the fact that the application layer program analyzes the file and then transmits the file to the inner core layer can be avoided, the overall complexity of the program is reduced, meanwhile, the QoS strategy file is directly analyzed by the QoS control module of the inner core layer, the transmission path can be shortened, and the analysis efficiency is improved. The invention can automatically load the service quality strategy file by the zfs file system when the system is restarted, can control the read-write operation of the directory earlier, and also avoids the condition of control failure caused by the failure of starting the intermediate agent program.
In order to improve the usability, the invention adds interfaces such as strategy deletion, strategy updating, strategy addition and the like in the service quality strategy file in the zfs user mode and the zfs kernel mode. In order to determine whether the service quality policy file is successfully loaded, a proc file display is added in zfs, and a user state program of the proc file can check the loading condition of the kernel state service quality policy file in real time, so that the interaction problem of the user state and the kernel state is solved, and the consistency of loading and setting is ensured.
The invention utilizes the sliding window technology and combines the QoS strategy file in the form of a hash table to quickly search the catalogue, and simultaneously uniformly controls the bandwidth and the iops in the underlying file system, thereby realizing the precise control of the bandwidth and the iops catalogue level.
Referring to fig. 1, the present invention provides a fine-grained storage quality of service control implementation method, which includes the following steps:
step 1, setting a service quality strategy file in a file system layer; the service quality strategy file is used for storing the service quality strategy of each catalogue in the file system; the quality of service policy includes a maximum limit of n quality of service items; wherein, n kinds of service quality items are respectively expressed as: a 1 st quality of service item, a 2 nd quality of service item, …, an n-th quality of service item;
in practical application, the service quality policy file adopts a hash table form, a key of the hash table is a directory id, and a value is a service quality policy.
As a specific embodiment, the quality of service items include, but are not limited to, read times per second, write times per second, read/write bandwidth values per second, write/bandwidth values per second, and read/write bandwidth values per second.
Step 2, in the file system layer, sliding windows corresponding to each service quality item are respectively set, so n sliding windows are respectively set: a 1 st sliding window corresponding to the 1 st quality of service item, a 2 nd sliding window corresponding to the 2 nd quality of service item, …, an nth sliding window corresponding to the nth quality of service item;
for any i-th sliding window, i=1, 2, …, n, a timer is used to control the total storage resource of the sliding window, namely: when the timing time of the timer is reached, namely, a sliding window time period is reached, the total storage resource of the sliding window is allocated to the ith sliding window, and the total storage resource of the sliding window is the same as the maximum limit value of the corresponding ith service quality item;
step 3, setting a quality of service Qos control module at a file system layer, wherein the quality of service Qos control module captures directory operation requests for all directories in the file system, and each directory operation request carries a directory ID, a service quality item requested at the time and a storage resource request value at the time;
the QoS control module queues the captured catalogue operation requests according to time sequence to form a total waiting queue;
step 4, the QoS control module judges whether to turn on the QoS switch currently, if not, the flow is ended; if the device is started, executing the step 5;
step 5, the QoS control module reads the catalog operation request of the queue head from the total waiting queue according to a set time interval to obtain catalog ID; step 6 is executed for each directory ID read;
step 6, the QoS control module executes step 6.1-step 6.2 for each directory ID read;
step 6.1, the QoS control module searches the QoS strategy file according to the catalog ID, judges whether the QoS strategy of the catalog corresponding to the catalog ID exists in the QoS strategy file, if not, directly executes the catalog operation request, and ends the flow; if so, executing the step 6.2;
step 6.2, the QoS control module determines sliding windows related to the searched QoS policies according to the searched QoS policies of the catalogs corresponding to the catalog IDs, and presumes m related sliding windows, wherein m is less than or equal to n;
for example, when the directory operation request is a read request, a sliding window corresponding to the number of reads per second, a sliding window corresponding to the read bandwidth value per second, and a sliding window corresponding to the read bandwidth value per second are needed to pass through;
when the directory operation request is a write request, the sliding window corresponding to the write times per second, the sliding window corresponding to the read/write times per second, the sliding window corresponding to the write bandwidth value per second and the sliding window corresponding to the read/write bandwidth value per second are needed to pass through.
The QoS control module enables the catalogue operation request to sequentially pass through m related sliding windows, then executes the catalogue operation request, and ends the flow;
when the directory operation request passes through any one of m related sliding windows, the sliding window is set to be a j-th sliding window, wherein j=1, 2, …, m, and the method comprises the following steps:
the QoS control module judges whether the residual storage resources of the jth sliding window in the current sliding window time period are larger than or equal to the storage resource request value of the directory operation request, if so, the directory operation request is executed, the storage resource request value of the directory operation request is subtracted by the residual storage resources of the jth sliding window in the current sliding window time period, the residual storage resources of the jth sliding window in the current sliding window time period are used as the residual storage resources of the jth sliding window in the current sliding window time period, and the process is ended; if not, adding the catalog operation request into a j sliding window queue corresponding to the j sliding window, and waiting for a timer to time to the next sliding window time period;
when the time period of the next sliding window is reached, the jth sliding window reacquires the total storage resources of the sliding window, the jth sliding window queue is activated, and all the directory operation requests in the jth sliding window queue are sequentially processed until the directory operation request is processed until the current directory operation request is processed.
In the invention, when the timing time of the timer is reached, namely, when a sliding window time period is reached, the total storage resource of the sliding window is allocated to the ith sliding window, and the following two methods can be adopted specifically:
when the time starting point of the sliding window time period is reached, the total storage resources of the sliding window are distributed to the ith sliding window at once;
or when the sliding window time period is reached, gradually distributing the sliding window total storage resources to the ith sliding window in the sliding window time period.
One specific embodiment is described below:
step 1, setting a service quality strategy file in a file system layer; the service quality strategy file is used for storing the service quality strategy of each catalogue in the file system; the service quality policy includes 6 items, which are respectively: a maximum limit for the number of reads per second, a maximum limit for the number of writes per second, a maximum limit for the number of reads per second, a maximum limit for the read bandwidth value per second, a maximum limit for the write bandwidth value per second, and a maximum limit for the read bandwidth value per second;
step 2, setting 6 sliding windows in a file system layer, wherein the sliding windows are respectively as follows: a 1 st sliding window corresponding to the reading times per second, a 2 nd sliding window corresponding to the writing times per second, a 3 rd sliding window corresponding to the reading and writing times per second, a 4 th sliding window corresponding to the reading bandwidth value per second, a 5 th sliding window corresponding to the writing bandwidth value per second and a 6 th sliding window corresponding to the reading and writing bandwidth value per second;
taking the 4 th sliding window corresponding to the reading bandwidth value per second as an example: assuming that the maximum limit of the read bandwidth value per second is 2M, the principle of the 4 th sliding window is: setting the timing time of the timer to be 1 second, and distributing 2M total storage resources to a 4 th sliding window in a sliding window time period of 0 to 1 second in the following distribution mode: dividing 0 to 1 second into a plurality of time slices, and dividing 2M total storage resources into each time slice; the time slicing is realized through the kernel timer, the default slicing size is available, and meanwhile, the time slicing size can be designated through zfs command lines to adapt to different service scenes, so that the flow control is more accurate, and the problem of flow doubling caused by a fixed window algorithm is solved.
In this time period, the 4 th sliding window processes the current read request, assuming that the current read request queue has three read requests, the 1 st read request is: reading 1.5M storage resources; the 2 nd read request is: reading a 0.5M storage resource; the 3 rd read request is: reading 1M storage resources; then: 1) Initially, the 4 th sliding window has 2M total memory resources, which is greater than 1.5M memory resources of the 1 st read request, so the 1 st read request passes through the 4 th sliding window; at this time, the 4 th sliding window remains 0.5M storage resources; 2) The remaining 0.5M memory resources of the 4 th sliding window are greater than or equal to the 0.5M memory resources of the 2 nd read request, so the 2 nd read request passes through the 4 th sliding window; at this time, the 4 th sliding window remains 0M storage resources; 3) The 4 th sliding window remains 0M storage resources, 1M storage resources of the 3 rd read request are not satisfied, and therefore, the 3 rd read request is continuously located in the waiting queue; 4) When the next timing time is reached, that is, the sliding window time period of 1 to 2 seconds is reached, reallocating 2M total memory resources to the 4 th sliding window, wherein at the moment, 1M memory resources of the 3 rd reading request are met, and therefore the 3 rd reading request passes through the 4 th sliding window;
for each read request, the type and number of the sliding windows are specifically set flexibly according to the requirement, for example, if the sliding windows are set, the sliding windows need to pass through four, namely: the sliding window corresponding to the reading times per second, the sliding window corresponding to the reading and writing times per second, the sliding window corresponding to the reading bandwidth value per second and the sliding window corresponding to the reading and writing bandwidth value per second are executed only when the reading request passes through the 4 sliding windows successfully, so that the fine granularity control of the reading request is realized, the excessively high use of a certain directory resource is avoided, and meanwhile, the high concurrent instantaneous pressure is also avoided.
According to the invention, under various sharing protocols, the iops and bandwidth of the directory level of the network file system can be uniformly controlled, and the implementation of specific protocols is not needed. According to the technical scheme, the control of the bandwidth and iops of the appointed directory can be realized more smoothly by realizing the sliding window on the underlying file system, so that the excessively high use of a certain directory resource is avoided, the high concurrent instant pressure is avoided, and the use experience of the file system is improved.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which is also intended to be covered by the present invention.
Claims (6)
1. The fine-grained storage service quality control implementation method is characterized by comprising the following steps of:
step 1, setting a service quality strategy file in a file system layer; the service quality strategy file is used for storing the service quality strategy of each catalogue in the file system; the quality of service policy includes a maximum limit of n quality of service items; wherein, n kinds of service quality items are respectively expressed as: a 1 st quality of service item, a 2 nd quality of service item, …, an n-th quality of service item;
step 2, in the file system layer, sliding windows corresponding to each service quality item are respectively set, so n sliding windows are respectively set: a 1 st sliding window corresponding to the 1 st quality of service item, a 2 nd sliding window corresponding to the 2 nd quality of service item, …, an nth sliding window corresponding to the nth quality of service item;
for any i-th sliding window, i=1, 2, …, n, a timer is used to control the total storage resource of the sliding window, namely: when the timing time of the timer is reached, namely, a sliding window time period is reached, the total storage resource of the sliding window is allocated to the ith sliding window, and the total storage resource of the sliding window is the same as the maximum limit value of the corresponding ith service quality item;
step 3, setting a quality of service Qos control module at a file system layer, wherein the quality of service Qos control module captures directory operation requests for all directories in the file system, and each directory operation request carries a directory ID, a service quality item requested at the time and a storage resource request value at the time;
the QoS control module queues the captured catalogue operation requests according to time sequence to form a total waiting queue;
step 4, the QoS control module judges whether to turn on the QoS switch currently, if not, the flow is ended; if the device is started, executing the step 5;
step 5, the QoS control module reads the catalog operation request of the queue head from the total waiting queue according to a set time interval to obtain catalog ID; step 6 is executed for each directory ID read;
step 6, the QoS control module executes step 6.1-step 6.2 for each directory ID read;
step 6.1, the QoS control module searches the QoS strategy file according to the catalog ID, judges whether the QoS strategy of the catalog corresponding to the catalog ID exists in the QoS strategy file, if not, directly executes the catalog operation request, and ends the flow; if so, executing the step 6.2;
step 6.2, the QoS control module determines sliding windows related to the searched QoS policies according to the searched QoS policies of the catalogs corresponding to the catalog IDs, and presumes m related sliding windows, wherein m is less than or equal to n;
the QoS control module enables the catalogue operation request to sequentially pass through m related sliding windows, then executes the catalogue operation request, and ends the flow;
when the directory operation request passes through any one of m related sliding windows, the sliding window is set to be a j-th sliding window, wherein j=1, 2, …, m, and the method comprises the following steps:
the QoS control module judges whether the residual storage resources of the jth sliding window in the current sliding window time period are larger than or equal to the storage resource request value of the directory operation request, if so, the directory operation request is executed, the storage resource request value of the directory operation request is subtracted by the residual storage resources of the jth sliding window in the current sliding window time period, the residual storage resources of the jth sliding window in the current sliding window time period are used as the residual storage resources of the jth sliding window in the current sliding window time period, and the process is ended; if not, adding the catalog operation request into a j sliding window queue corresponding to the j sliding window, and waiting for a timer to time to the next sliding window time period;
when the time period of the next sliding window is reached, the jth sliding window reacquires the total storage resources of the sliding window, the jth sliding window queue is activated, and all the directory operation requests in the jth sliding window queue are sequentially processed until the directory operation request is processed until the current directory operation request is processed.
2. The fine-grained storage quality of service control implementation method according to claim 1, wherein the quality of service policy file is in a form of a hash table, a key of the hash table is a directory id, and a value is a quality of service policy.
3. The fine grained storage quality of service control implementation of claim 1, wherein the quality of service term comprises a read per second number, a write per second number, a read per second bandwidth value, a write per second bandwidth value, and a read per second bandwidth value.
4. The fine-grained storage service quality control implementation method according to claim 3, wherein when the directory operation request is a read request, a sliding window corresponding to a number of reads per second, a sliding window corresponding to a read bandwidth value per second, and a sliding window corresponding to a read bandwidth value per second are needed;
when the directory operation request is a write request, the sliding window corresponding to the write times per second, the sliding window corresponding to the read/write times per second, the sliding window corresponding to the write bandwidth value per second and the sliding window corresponding to the read/write bandwidth value per second are needed to pass through.
5. The method for implementing fine-grained storage quality of service control according to claim 1, wherein when a timing time of a timer is reached, i.e. a sliding window time period is reached, a sliding window total storage resource is allocated to an i-th sliding window, specifically:
when the time starting point of the sliding window time period is reached, the total storage resources of the sliding window are distributed to the ith sliding window at once;
or alternatively
And when the sliding window time period is reached, gradually distributing the total storage resources of the sliding window to the ith sliding window in the sliding window time period.
6. The fine-grained storage Qos control implementation method according to claim 1, wherein the Qos control module is disposed in a kernel layer.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310498183.3A CN116501712A (en) | 2023-05-05 | 2023-05-05 | Fine-grained storage service quality control implementation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310498183.3A CN116501712A (en) | 2023-05-05 | 2023-05-05 | Fine-grained storage service quality control implementation method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116501712A true CN116501712A (en) | 2023-07-28 |
Family
ID=87316261
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310498183.3A Pending CN116501712A (en) | 2023-05-05 | 2023-05-05 | Fine-grained storage service quality control implementation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116501712A (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6047322A (en) * | 1997-05-27 | 2000-04-04 | Ukiah Software, Inc. | Method and apparatus for quality of service management |
US6640248B1 (en) * | 1998-07-10 | 2003-10-28 | Malibu Networks, Inc. | Application-aware, quality of service (QoS) sensitive, media access control (MAC) layer |
US20060069665A1 (en) * | 2004-09-24 | 2006-03-30 | Nec Corporation | File access service system, switch apparatus, quota management method and program |
US20120269053A1 (en) * | 2010-10-15 | 2012-10-25 | Brookhaven Science Associates, Llc | Co-Scheduling of Network Resource Provisioning and Host-to-Host Bandwidth Reservation on High-Performance Network and Storage Systems |
US20180307712A1 (en) * | 2017-04-25 | 2018-10-25 | Citrix Systems, Inc. | Detecting outliers in server transaction time as a form of time series data |
US10127234B1 (en) * | 2015-03-27 | 2018-11-13 | Amazon Technologies, Inc. | Proactive optimizations at multi-tier file systems |
US10585802B1 (en) * | 2017-07-13 | 2020-03-10 | EMC IP Holding Company LLC | Method and system for caching directories in a storage system |
CN114301891A (en) * | 2021-12-27 | 2022-04-08 | 北京睿芯高通量科技有限公司 | Python-based Web service and Socket client interaction method and system |
US20220334876A1 (en) * | 2021-04-16 | 2022-10-20 | EMC IP Holding Company LLC | Using machine learning techniques to flow control clients in a deduplication file system |
CN115426374A (en) * | 2022-08-29 | 2022-12-02 | 济南浪潮数据技术有限公司 | Service quality control method, device, equipment and storage medium |
-
2023
- 2023-05-05 CN CN202310498183.3A patent/CN116501712A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6047322A (en) * | 1997-05-27 | 2000-04-04 | Ukiah Software, Inc. | Method and apparatus for quality of service management |
US6640248B1 (en) * | 1998-07-10 | 2003-10-28 | Malibu Networks, Inc. | Application-aware, quality of service (QoS) sensitive, media access control (MAC) layer |
US20060069665A1 (en) * | 2004-09-24 | 2006-03-30 | Nec Corporation | File access service system, switch apparatus, quota management method and program |
US20120269053A1 (en) * | 2010-10-15 | 2012-10-25 | Brookhaven Science Associates, Llc | Co-Scheduling of Network Resource Provisioning and Host-to-Host Bandwidth Reservation on High-Performance Network and Storage Systems |
US10127234B1 (en) * | 2015-03-27 | 2018-11-13 | Amazon Technologies, Inc. | Proactive optimizations at multi-tier file systems |
US20180307712A1 (en) * | 2017-04-25 | 2018-10-25 | Citrix Systems, Inc. | Detecting outliers in server transaction time as a form of time series data |
US10585802B1 (en) * | 2017-07-13 | 2020-03-10 | EMC IP Holding Company LLC | Method and system for caching directories in a storage system |
US20220334876A1 (en) * | 2021-04-16 | 2022-10-20 | EMC IP Holding Company LLC | Using machine learning techniques to flow control clients in a deduplication file system |
CN114301891A (en) * | 2021-12-27 | 2022-04-08 | 北京睿芯高通量科技有限公司 | Python-based Web service and Socket client interaction method and system |
CN115426374A (en) * | 2022-08-29 | 2022-12-02 | 济南浪潮数据技术有限公司 | Service quality control method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US5630067A (en) | System for the management of multiple time-critical data streams | |
US20200089624A1 (en) | Apparatus and method for managing storage of data blocks | |
US7281032B2 (en) | File sharing system with data mirroring by storage systems | |
KR100449981B1 (en) | How video file servers work and video file servers for video on demand | |
US7260656B2 (en) | Storage system having a plurality of controllers | |
US7287134B2 (en) | Methods and systems of managing I/O operations in data storage systems | |
US6442706B1 (en) | Resource allocation throttle for remote data mirroring system | |
US8204863B2 (en) | Multi-phase file system restore with selective on-demand data availability | |
US20190197028A1 (en) | Database management system with database hibernation and bursting | |
US7370142B2 (en) | Command control method in network storage system | |
US9424314B2 (en) | Method and apparatus for joining read requests | |
EP2618261B1 (en) | Qos control method and system for storage system | |
EP3285187B1 (en) | Optimized merge-sorting of data retrieved from parallel storage units | |
CN112565774B (en) | Video transcoding resource scheduling method and device | |
US10523743B2 (en) | Dynamic load-based merging | |
JPH0950667A (en) | Method for control of disk drive | |
JPH0793205A (en) | Distributed filing system | |
Golubchik et al. | Analysis of striping techniques in robotic storage libraries | |
JP3876100B2 (en) | A system that manages access to devices | |
CN113553346B (en) | Large-scale real-time data stream integrated processing, forwarding and storing method and system | |
JP2005165852A (en) | Storage system, storage control device, and control method of storage system | |
CN111124302B (en) | SAN shared file storage and archiving method and system | |
US11500799B2 (en) | Managing access to a CPU on behalf of a block application and a non-block application | |
US7143250B1 (en) | Methods and systems for improving mirror performance via optimized multi-channel utilization | |
CN116501712A (en) | Fine-grained storage service quality control implementation method |
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 | ||
CB02 | Change of applicant information |
Address after: Room 711C, Floor 7, Building A, Yard 19, Ronghua Middle Road, Daxing District, Beijing Economic-Technological Development Area, 100176 Applicant after: Beijing Zhongke Flux Technology Co.,Ltd. Address before: Room 711C, Floor 7, Building A, Yard 19, Ronghua Middle Road, Daxing District, Beijing Economic-Technological Development Area, 100176 Applicant before: Beijing Ruixin high throughput technology Co.,Ltd. |
|
CB02 | Change of applicant information |