CN111752704A - Distributed storage file system MDS log disk-dropping method and device - Google Patents
Distributed storage file system MDS log disk-dropping method and device Download PDFInfo
- Publication number
- CN111752704A CN111752704A CN202010444961.7A CN202010444961A CN111752704A CN 111752704 A CN111752704 A CN 111752704A CN 202010444961 A CN202010444961 A CN 202010444961A CN 111752704 A CN111752704 A CN 111752704A
- Authority
- CN
- China
- Prior art keywords
- log
- processed
- mds
- data
- packaging
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000004806 packaging method and process Methods 0.000 claims abstract description 57
- 238000012856 packing Methods 0.000 claims description 35
- 230000005540 biological transmission Effects 0.000 claims description 2
- 238000004904 shortening Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 101000878595 Arabidopsis thaliana Squalene synthase 1 Proteins 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- 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
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
- G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a distributed storage file system MDS log destaging method and a device, wherein the method comprises the steps of obtaining the service idle state of a node where MDS is located, and determining the number of log sections needing to be destaged according to the service idle state of the node; and in the process of executing the tray dropping of each log segment, packaging and tray dropping processing are carried out on a plurality of log events to be processed in the log segment in a multithread parallel processing mode. The invention determines the number of log segments needing to be landed according to the idle state of the node service, reduces landing operation when the service is busy, realizes dynamic control of log landing according to the busy degree of the node service, and reduces the time delay influence of the log landing on normal service. And packaging and destaging a plurality of to-be-processed log events in the log segment in a multithread parallel processing mode to achieve the purpose of shortening the processing time of a single directory, thereby further reducing the time delay influence of the MDS log segment disk on normal services.
Description
Technical Field
The invention relates to the field of distributed storage file systems, in particular to an MDS log disk-dropping method and device of a distributed storage file system.
Background
In a current distributed file storage system, all operations of an MDS (Metadata Server, Metadata service in the distributed file storage system) are recorded in log segments, each log segment can record 1024 log events, each Metadata operation is recorded as one log event, a directory entry object of the corresponding Metadata operation is marked to be processed, and the log event is waited to be updated to a disk after being landed.
The MDS starts a corresponding thread at regular time to check the number of log segments of the current MDS, if the number of the log segments of the current MDS exceeds a limit value, the MDS writes the log segments exceeding the limit value into a disk one by one in sequence, and in the writing process, the MDS is carried out by taking a directory corresponding to metadata operation as a unit (after log events related to a directory in the log segments are sequentially landed, the landing operation is carried out on the log events related to the next directory, namely all directory items to be processed under the directory are sequentially packed into the disk, the landing operation is carried out on the next log segment after the landing of the log events of the log segments is finished), and then the log segments are deleted from a memory.
When the MDS performs log segment disk operation, all business threads of the MDS can be suspended, and other business threads can continue to perform only after the log segment disk-falling task is completed; if a log destaging operation involves more log segments to be processed, and log events related to each log segment are excessive (in a large directory scene, the number of directories related to a single log segment is small, but directory entries needing to be processed under a single directory are more, and a log segment only relates to one directory under an extreme condition, and 1024 directory entries to be processed under the directory need to be processed in sequence for destaging), the log destaging time is too long, and further, service delay fluctuation corresponding to the MDS is caused, and blocking is caused.
Disclosure of Invention
In order to solve the problems, the invention provides a distributed storage file system MDS log destaging method which dynamically manages log destaging according to business busyness, shortens single directory processing time by a parallel processing destaging mode and reduces time delay influence of log destaging on normal business.
The technical scheme of the invention is as follows: a distributed storage file system MDS log disk-dropping method comprises the following steps:
acquiring the service idle state of a node where the MDS is located, and determining the number of log segments needing to be landed according to the service idle state of the node;
and in the process of executing the tray dropping of each log segment, packaging and tray dropping processing are carried out on a plurality of log events to be processed in the log segment in a multithread parallel processing mode.
Further, the service idle state of the node refers to a CPU occupation ratio and a memory occupation ratio of the node.
Further, determining the number of logs needing to be landed according to the service idle state of the node specifically comprises:
calculating the heat value of MDS, wherein the heat value of MDS is the CPU occupancy ratio multiplied by 100+ the memory occupancy ratio multiplied by 100;
determining the number of log segments needing to be landed according to the heat value of the MDS; the smaller the heat value is, the more the number of log segments required to be landed is.
Further, the method for performing packing and tray dropping processing on a plurality of log events to be processed in the log segment in a multithread parallel processing mode specifically comprises the following steps:
storing all log events to be processed of the log segment into a data pool to be processed;
data in a data pool to be processed is transmitted into a packing thread pool; the packaging thread pool comprises a plurality of packaging threads, and data in the to-be-processed data pool is assigned to each packaging thread in a balanced manner;
and after the data assignment is completed, starting all the packaging threads and simultaneously performing packaging and tray dropping operation.
Further, in the process of storing all to-be-processed log events of the log segment into the to-be-processed data pool, grouping all to-be-processed log events of the log segment according to the affiliated directory, and storing the to-be-processed log events into the to-be-processed data pool by taking the directory as a unit.
Furthermore, when the data in the data pool to be processed is transmitted into the packing thread pool, the data in the data pool to be processed is transmitted into the packing thread pool by taking the directory as a unit, and after the log event of one directory is processed by packing and dropping, the log event of the next directory is transmitted into the packing thread pool to be processed by packing and dropping.
Further, the balanced assignment of the data in the to-be-processed data pool to each packing thread specifically includes:
assigning the first log event to be processed to a first packing thread, assigning the second log event to be processed to a second packing thread, and performing data assignment again after all the packing threads assign values once.
The technical scheme of the invention also comprises a distributed storage file system MDS log disk-dropping device, which comprises,
a node service idle state acquisition module: acquiring a service idle state of a node where MDS is located;
a number of log segments of a landing disk determination module: determining the number of log segments needing to be landed according to the idle state of the node service;
a packaging and tray-dropping processing module: and in the process of executing the tray dropping of each log segment, packaging and tray dropping processing are carried out on a plurality of log events to be processed in the log segment in a multithread parallel processing mode.
Further, the service idle state of the node refers to a CPU occupation ratio and a memory occupation ratio of the node;
the number determining module of log segments of the landed disk determines the number of log segments needing to be landed according to the idle state of the node service, and specifically comprises the following steps:
calculating the heat value of MDS, wherein the heat value of MDS is the CPU occupancy ratio multiplied by 100+ the memory occupancy ratio multiplied by 100;
determining the number of log segments needing to be landed according to the heat value of the MDS; the smaller the heat value is, the more the number of log segments required to be landed is.
Further, the package tray-dropping processing module comprises,
a data saving unit: grouping all log events to be processed of the log segments by the belonged directories, and storing the log events to a data pool to be processed by taking the directories as units;
the data transmission assignment unit: transmitting data in a data pool to be processed into a packing thread pool by taking a directory as a unit; the packaging thread pool comprises a plurality of packaging threads, data in the data pool to be processed is assigned to each packaging thread in a balanced mode, namely a first log event to be processed is assigned to a first packaging thread, a second log event to be processed is assigned to a second packaging thread, and after all the packaging threads are assigned once, a round of data assignment is carried out again;
a tray falling operation unit: and after the data assignment of one directory is completed, starting all packaging threads and simultaneously performing packaging and tray dropping operation.
The method and the device for logging of the MDS of the distributed storage file system, provided by the invention, have the advantages that the business idle state of the node where the MDS is located is obtained, the number of log sections needing to be logged is determined according to the business idle state of the node, the logging operation is reduced when the business is busy, the logging is dynamically controlled according to the busy degree of the node business, and the time delay influence of the logging on normal business is reduced. In the process of executing the tray-dropping of each log segment, a plurality of log events to be processed in the log segment are packed and subjected to tray-dropping in a multithreading parallel processing mode, namely, a multithreading method is used for concurrently processing a plurality of log events under a single directory related to a certain log segment, and after all the log events under the directory are completely processed, a tray-dropping operation is executed, so that the purpose of shortening the processing time of the single directory is achieved, and the time delay influence of an MDS log segment tray on normal services is further reduced.
Drawings
FIG. 1 is a schematic flow chart of a method according to an embodiment of the present invention.
Fig. 2 is a flow chart of a parallel packaging and tray-dropping processing method according to an embodiment of the present invention.
Fig. 3 is a flowchart of a method of a specific implementation manner according to an embodiment of the present invention.
Fig. 4 is a schematic block diagram of a second structure according to an embodiment of the present invention.
Fig. 5 is a block diagram illustrating a structure of a second package tray-dropping processing module according to an embodiment of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings by way of specific examples, which are illustrative of the present invention and are not limited to the following embodiments.
Example one
The embodiment provides a distributed storage file system MDS log destaging method, wherein the busy degree of a service is dynamically changed during the operation of MDS, when the service is idle, the method performs destaging operation on log segments exceeding the limit as much as possible, and reduces the operation when the service is busy, thereby realizing the dynamic management of log destaging and reducing the time delay influence of the MDS log segment on normal service; the method comprises the steps of concurrently processing a plurality of log events (each log event corresponds to one directory entry) under a single directory related in a log segment by using a multithreading method, and executing one-time tray-dropping operation after all the log events under the directory are completely processed, so that the purpose of shortening the processing time of the single directory is achieved, and the time delay influence of an MDS log segment tray on normal service is reduced.
As shown in fig. 1, the method comprises the following steps:
SS 1: acquiring the service idle state of a node where the MDS is located, and determining the number of log segments needing to be landed according to the service idle state of the node;
and SS2, in the process of performing tray dropping on each log segment, performing packaging tray dropping on a plurality of log events to be processed in the log segment in a multithread parallel processing mode.
Specifically, in this embodiment, the service idle state of the node is represented by the CPU occupancy ratio and the memory occupancy ratio of the node. Determining the number of logs needing to be landed according to the service idle state of the node, specifically comprising:
calculating the heat value of MDS, wherein the heat value of MDS is the CPU occupancy ratio multiplied by 100+ the memory occupancy ratio multiplied by 100;
determining the number of log segments needing to be landed according to the heat value of the MDS; the smaller the heat value is, the more the number of log segments required to be landed is.
For example, 10% of log segments are processed when the heat value of the MDS exceeds a first threshold, 20% of log segments are processed when the heat value is between the first threshold and a second threshold, and so on. The first threshold is greater than the second threshold.
It should be noted that, a timing task may be added to the MDS, a CPU occupancy ratio and a memory occupancy ratio under the node may be collected at a timing, and a heat value may be calculated and stored at the same time. When the log disk-falling processing thread is started, checking the stored latest MDS heat value, and determining the number of log segments which can be processed by the log disk-falling operation. The current MDS heat value is updated each time a log paragraph disk operation is completed. Or when the log landing processing thread is started, acquiring the CPU occupation ratio and the memory occupation ratio under the current node, calculating the heat value, and determining the number of the landing log segments.
The embodiment performs the packaging and disk-dropping processing in a multi-thread parallel processing mode, shortens the processing time of a single directory, and reduces the influence of the log disk-dropping on normal services. As shown in fig. 2, in step SS2, performing packed destaging processing on a plurality of to-be-processed log events in a log segment by means of multi-thread parallel processing is implemented by the following steps:
SS 2-1, storing all to-be-processed log events of the log segment into a to-be-processed data pool;
SS 2-2, data in the data pool to be processed is transmitted to a packing thread pool; the packaging thread pool comprises a plurality of packaging threads, and data in the to-be-processed data pool is assigned to each packaging thread in a balanced manner;
and SS 2-3, after data assignment is completed, starting all packing threads to simultaneously carry out packing and tray dropping operations.
In specific implementation, all log events in each log segment are processed by taking a directory as a unit, namely, after the log events in a directory are packed and landed, the operation of the directory is performed. Therefore, in the present embodiment, in the process of saving all to-be-processed log events of the log segment in the to-be-processed data pool in step S201, all to-be-processed log events of the log segment are grouped by the directory to which they belong, and the to-be-processed log events are saved in the to-be-processed data pool by taking the directory as a unit. Meanwhile, when the data in the data pool to be processed is transferred into the packing thread pool in step S202, the data in the data pool to be processed is transferred into the packing thread pool in units of directories, and after the log event of one directory is processed by packing and destaging, the log event of the next directory is transferred into the packing thread pool to be processed by packing and destaging. In addition, the data assignment in step S203 is completed, which means that the assignment of the log event in one directory is completed in the packaging thread.
When there are a plurality of log segments to be landed, the data saving operation is performed in units of log segments. Firstly, storing the log event to be processed in a log segment into a data pool to be processed by taking a directory as a unit, after all directories of the log segment are processed in sequence, storing the log event to be processed in the next log segment into the data pool to be processed by taking the directory as a unit, and so on until all log segments needing to be landed are processed. In addition, operations may be controlled and triggered by a log-off-disk primary thread.
To further explain the method in detail, a specific implementation is provided as follows, as shown in fig. 3, including the following steps:
s101, starting a log disk-dropping main thread;
s102, detecting the heat value of the node where the current MDS is located, and determining the number of log segments which can be processed by the log destaging operation;
s103, sequencing the log segments which can be processed;
s104, grouping the to-be-processed log events of the first log segment according to a directory, and storing the to-be-processed log events of the log segment into a to-be-processed data pool by taking the directory as a unit;
s105, sequencing the to-be-processed log events in the to-be-processed data pool by taking a directory as a unit;
s106, transmitting the to-be-processed log event contained in the first directory into a packing thread pool; the packaging thread pool comprises a plurality of packaging threads, and the log events to be processed in the data pool to be processed are assigned to each packaging thread in a balanced manner; meanwhile, the log off-disc main thread is dormant;
s107, after assignment is completed, starting all packaging threads and simultaneously performing packaging and tray dropping operation;
s108, awakening a log destaging main thread after the packaging destaging operation is finished, detecting whether the data in the data pool to be processed is processed or not by the log destaging main thread, and triggering a next directory in the data pool to be processed to repeatedly execute the steps S106-S107 if the data in the data pool to be processed is not processed; if the log segment is processed, triggering the next log segment to repeat S104-S107 until all the log segments needing to be processed are processed.
Example two
The embodiment provides an MDS log dropping device of a distributed storage file system, and the method in the first embodiment can be implemented.
As shown in fig. 4, the apparatus includes the following functional modules.
Node service idle state acquisition module 1: acquiring a service idle state of a node where MDS is located;
and a disk-drop log segment number determining module 2: determining the number of log segments needing to be landed according to the idle state of the node service;
packing and tray-dropping processing module 3: and in the process of executing the tray dropping of each log segment, packaging and tray dropping processing are carried out on a plurality of log events to be processed in the log segment in a multithread parallel processing mode.
The service idle state of the node refers to a CPU occupation ratio and a memory occupation ratio of the node.
The number determining module 2 for log segments to be landed determines the number of log segments to be landed according to the idle state of the node service, specifically:
calculating the heat value of MDS, wherein the heat value of MDS is the CPU occupancy ratio multiplied by 100+ the memory occupancy ratio multiplied by 100;
determining the number of log segments needing to be landed according to the heat value of the MDS; the smaller the heat value is, the more the number of log segments required to be landed is.
It should be noted that, a timing task may be added to the MDS, a CPU occupancy ratio and a memory occupancy ratio under the node may be collected at a timing, and a heat value may be calculated and stored at the same time. When the log disk-falling processing thread is started, checking the stored latest MDS heat value, and determining the number of log segments which can be processed by the log disk-falling operation. The current MDS heat value is updated each time a log paragraph disk operation is completed. Or when the log landing processing thread is started, acquiring the CPU occupation ratio and the memory occupation ratio under the current node, calculating the heat value, and determining the number of the landing log segments.
The embodiment performs the packaging and disk-dropping processing in a multi-thread parallel processing mode, shortens the processing time of a single directory, and reduces the influence of the log disk-dropping on normal services. As shown in fig. 5, the package-and-tray-dropping processing module 3 includes the following functional units to implement multi-thread parallel processing.
The data holding unit 31: grouping all log events to be processed of the log segments by the belonged directories, and storing the log events to a data pool to be processed by taking the directories as units;
data-in assignment unit 32: transmitting data in a data pool to be processed into a packing thread pool by taking a directory as a unit; the packaging thread pool comprises a plurality of packaging threads, data in the data pool to be processed is assigned to each packaging thread in a balanced mode, namely a first log event to be processed is assigned to a first packaging thread, a second log event to be processed is assigned to a second packaging thread, and after all the packaging threads are assigned once, a round of data assignment is carried out again;
the tray falling operation unit 33: and after the data assignment of one directory is completed, starting all packaging threads and simultaneously performing packaging and tray dropping operation.
The above disclosure is only for the preferred embodiments of the present invention, but the present invention is not limited thereto, and any non-inventive changes that can be made by those skilled in the art and several modifications and amendments made without departing from the principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. A distributed storage file system MDS log disk-dropping method is characterized by comprising the following steps:
acquiring the service idle state of a node where the MDS is located, and determining the number of log segments needing to be landed according to the service idle state of the node;
and in the process of executing the tray dropping of each log segment, packaging and tray dropping processing are carried out on a plurality of log events to be processed in the log segment in a multithread parallel processing mode.
2. The method for MDS log destaging of a distributed storage file system of claim 1, wherein the business idle state of the node is a CPU occupancy ratio and a memory occupancy ratio of the node.
3. The method for MDS log destaging of the distributed storage file system of claim 2, wherein the determining the number of logs to be destaged according to the service idle state of the node specifically comprises:
calculating the heat value of MDS, wherein the heat value of MDS is the CPU occupancy ratio multiplied by 100+ the memory occupancy ratio multiplied by 100;
determining the number of log segments needing to be landed according to the heat value of the MDS; the smaller the heat value is, the more the number of log segments required to be landed is.
4. The MDS log destaging method of the distributed storage file system according to claim 1, 2 or 3, wherein the packing and destaging are performed on a plurality of to-be-processed log events in the log segment in a multi-thread parallel processing manner, specifically comprising:
storing all log events to be processed of the log segment into a data pool to be processed;
data in a data pool to be processed is transmitted into a packing thread pool; the packaging thread pool comprises a plurality of packaging threads, and data in the to-be-processed data pool is assigned to each packaging thread in a balanced manner;
and after the data assignment is completed, starting all the packaging threads and simultaneously performing packaging and tray dropping operation.
5. The MDS log destaging method of the distributed storage file system of claim 4, wherein in the process of saving all to-be-processed log events of the log segment into the to-be-processed data pool, all to-be-processed log events of the log segment are grouped by the directory to which the to-be-processed log events belong, and the to-be-processed log events are saved into the to-be-processed data pool in units of the directory.
6. The MDS log destaging method of the distributed storage file system of claim 5, wherein when the data in the to-be-processed data pool is transferred into the packing thread pool, the data in the to-be-processed data pool is transferred into the packing thread pool in units of directories, and after a log event of one directory is processed by packing, a log event of a next directory is transferred into the packing thread pool to be processed by packing.
7. The method for MDS log destaging of a distributed storage file system of claim 6, wherein the balanced assignment of data in the to-be-processed data pool to each packing thread comprises:
assigning the first log event to be processed to a first packing thread, assigning the second log event to be processed to a second packing thread, and performing data assignment again after all the packing threads assign values once.
8. The device for MDS log disk-dropping of the distributed storage file system is characterized by comprising a node service idle state acquisition module: acquiring a service idle state of a node where MDS is located; a number of log segments of a landing disk determination module: determining the number of log segments needing to be landed according to the idle state of the node service;
a packaging and tray-dropping processing module: and in the process of executing the tray dropping of each log segment, packaging and tray dropping processing are carried out on a plurality of log events to be processed in the log segment in a multithread parallel processing mode.
9. The MDS log destaging device of claim 8, wherein the business idle state of the node is a CPU occupancy ratio and a memory occupancy ratio of the node;
the number determining module of log segments of the landed disk determines the number of log segments needing to be landed according to the idle state of the node service, and specifically comprises the following steps:
calculating the heat value of MDS, wherein the heat value of MDS is the CPU occupancy ratio multiplied by 100+ the memory occupancy ratio multiplied by 100;
determining the number of log segments needing to be landed according to the heat value of the MDS; the smaller the heat value is, the more the number of log segments required to be landed is.
10. The MDS log destaging device of claim 9, wherein the package destaging processing module comprises,
a data saving unit: grouping all log events to be processed of the log segments by the belonged directories, and storing the log events to a data pool to be processed by taking the directories as units;
the data transmission assignment unit: transmitting data in a data pool to be processed into a packing thread pool by taking a directory as a unit; the packaging thread pool comprises a plurality of packaging threads, data in the data pool to be processed is assigned to each packaging thread in a balanced mode, namely a first log event to be processed is assigned to a first packaging thread, a second log event to be processed is assigned to a second packaging thread, and after all the packaging threads are assigned once, a round of data assignment is carried out again;
a tray falling operation unit: and after the data assignment of one directory is completed, starting all packaging threads and simultaneously performing packaging and tray dropping operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010444961.7A CN111752704B (en) | 2020-05-23 | 2020-05-23 | Distributed storage file system MDS log disk-dropping method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010444961.7A CN111752704B (en) | 2020-05-23 | 2020-05-23 | Distributed storage file system MDS log disk-dropping method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111752704A true CN111752704A (en) | 2020-10-09 |
| CN111752704B CN111752704B (en) | 2023-01-10 |
Family
ID=72673775
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010444961.7A Active CN111752704B (en) | 2020-05-23 | 2020-05-23 | Distributed storage file system MDS log disk-dropping method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111752704B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118012732A (en) * | 2024-04-08 | 2024-05-10 | 合众新能源汽车股份有限公司 | Log management method and device and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104239133A (en) * | 2014-09-26 | 2014-12-24 | 北京国双科技有限公司 | Log processing method, device and server |
| CN109298990A (en) * | 2018-10-17 | 2019-02-01 | 平安科技（深圳）有限公司 | Log storing method, device, computer equipment and storage medium |
| CN109582244A (en) * | 2018-12-05 | 2019-04-05 | 广东浪潮大数据研究有限公司 | A kind of metadata rule method, apparatus, terminal and computer readable storage medium |
| CN110442646A (en) * | 2019-07-29 | 2019-11-12 | 北京易捷思达科技发展有限公司 | A kind of ceph data simultaneous module main side write performance optimization system and method |
-
2020
- 2020-05-23 CN CN202010444961.7A patent/CN111752704B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104239133A (en) * | 2014-09-26 | 2014-12-24 | 北京国双科技有限公司 | Log processing method, device and server |
| CN109298990A (en) * | 2018-10-17 | 2019-02-01 | 平安科技（深圳）有限公司 | Log storing method, device, computer equipment and storage medium |
| CN109582244A (en) * | 2018-12-05 | 2019-04-05 | 广东浪潮大数据研究有限公司 | A kind of metadata rule method, apparatus, terminal and computer readable storage medium |
| CN110442646A (en) * | 2019-07-29 | 2019-11-12 | 北京易捷思达科技发展有限公司 | A kind of ceph data simultaneous module main side write performance optimization system and method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118012732A (en) * | 2024-04-08 | 2024-05-10 | 合众新能源汽车股份有限公司 | Log management method and device and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111752704B (en) | 2023-01-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10776263B2 (en) | Non-deterministic window scheduling for data storage systems | |
| US10817217B2 (en) | Data storage system with improved time-to-ready | |
| JP6014925B2 (en) | Memory recovery method and apparatus | |
| EP2889776B1 (en) | Data arrangement control program, data arrangement control method and data arrangment control apparatus | |
| CN104137081A (en) | Memory reorder queue biasing preceding high latency operations | |
| US11314689B2 (en) | Method, apparatus, and computer program product for indexing a file | |
| CN111324303B (en) | SSD garbage recycling method, SSD garbage recycling device, computer equipment and storage medium | |
| US9569381B2 (en) | Scheduler for memory | |
| US5758137A (en) | Method and system for processing timer requests within a computer | |
| US20190286582A1 (en) | Method for processing client requests in a cluster system, a method and an apparatus for processing i/o according to the client requests | |
| JP2010231502A (en) | Job processing method, computer-readable recording medium having stored job processing program, and job processing system | |
| CN111752704B (en) | Distributed storage file system MDS log disk-dropping method and device | |
| CN110597879B (en) | Method and device for processing time series data | |
| CN106844417B (en) | Hot migration method and device based on file directory | |
| US10783982B2 (en) | Probation bit for data storage memory | |
| EP1970815A1 (en) | Data transfering apparatus and information processing system | |
| CN106537321B (en) | Method, device and storage system for accessing file | |
| US20100185829A1 (en) | Extent consolidation and storage group allocation | |
| CN110837428B (en) | Storage device management method and device | |
| US20200004636A1 (en) | Data Storage System with Strategic Contention Avoidance | |
| US9749577B1 (en) | Host video recording by baseboard management controller (BMC) | |
| CN115756955A (en) | Data backup and data recovery method and device and computer equipment | |
| CN112597112B (en) | Data processing method and device, electronic equipment and storage medium | |
| CN114116790A (en) | Data processing method and device | |
| CN116149897B (en) | Chip functional safety fault processing 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |