CN109992481A - A kind of performance analysis tool, method and storage medium for distributed block storage - Google Patents
A kind of performance analysis tool, method and storage medium for distributed block storage Download PDFInfo
- Publication number
- CN109992481A CN109992481A CN201910290061.9A CN201910290061A CN109992481A CN 109992481 A CN109992481 A CN 109992481A CN 201910290061 A CN201910290061 A CN 201910290061A CN 109992481 A CN109992481 A CN 109992481A
- Authority
- CN
- China
- Prior art keywords
- time
- consuming
- timestamp
- stack
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3006—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3034—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a storage system, e.g. DASD based or network based
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3041—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is an input/output interface
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
- G06F11/3419—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment by assessing time
Abstract
This application discloses a kind of performance analysis tool, method and storage mediums for distributed block storage, comprising: timestamp module, for setup time to stab on each key node of IO stack in distributed block storage system;Time-consuming statistical module, for when IO passes through corresponding timestamp, counting and calculating the corresponding time-consuming information of each timestamp;Abnormal IO stack output module, for when the total time-consuming of entire IO stack is more than preset threshold, the corresponding time-consuming information of timestamp each in the entire IO stack of this exception IO to be exported;Performance management module resets interface for providing time-consuming statistical information printing interface and time-consuming statistical information.The application passes through the interaction of this four modules, time-consuming management and analysis are carried out to IO stack each stage, the bottleneck point that can be apparent from IO stack, simultaneously, abnormal IO information is exported, convenient for carrying out trace analysis to non-lethal abnormal performance, to help to advanced optimize performance of storage system.
Description
Technical field
The present invention relates to technical field of distributed memory, more particularly to a kind of performance evaluation for distributed block storage
Tool, method and storage medium.
Background technique
Currently, the team that goes out of transmitting-receiving, message pretreatment, queue including protocol message enters in distributed block storage system
Team, affairs modules, the entire IO stack process such as processing be related to the transmitting-receiving of many modules and message.And existing distributed block storage
System cannot can only accurately obtain the time-consuming situation of entire IO stack, also, when out to time-consuming statistics is carried out within the scope of single function
When existing non-lethal abnormal performance, it is often difficult to the problem of carrying out positioning analysis, causing to abnormal IO analysis and is helped without obvious.
Therefore, how to realize to the time-consuming statistical function in each stage in entire IO stack process and exception IO analytic function, be
Those skilled in the art's technical problem urgently to be resolved.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of performance analysis tools for distributed block storage, method
And storage medium, time-consuming management and performance evaluation can be carried out to IO stack each stage, and export to abnormal IO information, just
It is tracked in abnormal IO.Its concrete scheme is as follows:
A kind of performance analysis tool for distributed block storage, comprising:
Timestamp module, for setup time to stab on each key node of IO stack in distributed block storage system;
Time-consuming statistical module, for it is corresponding to count and calculate each timestamp when IO passes through corresponding timestamp
Time-consuming information;
Abnormal IO stack output module, for when the total time-consuming of entire IO stack be more than preset threshold when, by this exception IO's
The corresponding time-consuming information output of each timestamp in entire IO stack;
Performance management module resets interface for providing time-consuming statistical information printing interface and time-consuming statistical information, with complete
At the printing of time-consuming statistical information and the clearing of time-consuming statistical information.
Preferably, described in the above-mentioned performance analysis tool for distributed block storage provided in an embodiment of the present invention
Timestamp includes the upstream moment for executing moment and upstream timestamp when IO passes through each key node.
Preferably, in the above-mentioned performance analysis tool for distributed block storage provided in an embodiment of the present invention, first
The corresponding upstream timestamp of a timestamp upstream timestamp corresponding with timestamp described in the last one is first institute
State timestamp.
Preferably, in the above-mentioned performance analysis tool for distributed block storage provided in an embodiment of the present invention, each institute
Stating the corresponding time-consuming information of timestamp includes corresponding total degree of entire IO stack each stage, total time-consuming, average time-consuming and maximum consumption
When;Wherein,
The average time-consuming ratio for the total time-consuming and total degree;When the total degree is 0, the average time-consuming
It is 0.
Preferably, described in the above-mentioned performance analysis tool for distributed block storage provided in an embodiment of the present invention
Abnormal IO stack output module includes:
Abnormal deciding means, when for judging whether the total time-consuming of entire IO stack is more than preset threshold, if then determining this
Secondary IO is abnormal;
Information output unit is used for after the completion of this exception IO, successively will be in this entire IO stack according to time sequencing
The time-consuming information of each timestamp is output in log, by analyzing the log, is blocked in acquisition IO stack
Part.
The embodiment of the invention also provides a kind of method for analyzing performance for distributed block storage, comprising:
By timestamp module, setup time is stabbed on each key node of IO stack in distributed block storage system;
It is corresponding with each timestamp is calculated by time-consuming statistical module counts when IO passes through corresponding timestamp
Time-consuming information;
When entire IO stack total time-consuming is more than preset threshold, by abnormal IO stack output module by the entire of this exception IO
The corresponding time-consuming information output of each timestamp in IO stack;
The time-consuming statistical information printing interface of passage capacity management module and time-consuming statistical information reset interface, complete time-consuming
The clearing of the printing of statistical information and time-consuming statistical information.
Preferably, described in the above-mentioned method for analyzing performance for distributed block storage provided in an embodiment of the present invention
Timestamp includes the upstream moment for executing moment and upstream timestamp when IO passes through each key node;
It is described in first by the corresponding upstream time stamp setting of first timestamp in the timestamp module
Timestamp;It is first timestamp by the corresponding upstream time stamp setting of the last one described timestamp.
Preferably, in the above-mentioned method for analyzing performance for distributed block storage provided in an embodiment of the present invention, each institute
Stating the corresponding time-consuming information of timestamp includes corresponding total degree of entire IO stack each stage, total time-consuming, average time-consuming and maximum consumption
When;Wherein,
The average time-consuming ratio for the total time-consuming and total degree;When the total degree is 0, the average time-consuming
It is 0.
Preferably, in the above-mentioned method for analyzing performance for distributed block storage provided in an embodiment of the present invention, when whole
It, will be each described in the entire IO stack of this exception IO by abnormal IO stack output module when a IO stack total time-consuming is more than preset threshold
The corresponding time-consuming information output of timestamp, specifically includes:
When judging whether the total time-consuming of entire IO stack is more than preset threshold by abnormal deciding means, if then determining this
IO is abnormal;
It, successively will be each in this entire IO stack according to time sequencing through information output unit after the completion of this exception IO
The time-consuming information of the timestamp is output in log, by analyzing the log, obtains the portion being blocked in IO stack
Point.
The embodiment of the invention also provides a kind of computer readable storage mediums, for storing computer program, wherein institute
It states and is realized when computer program is executed by processor such as the above-mentioned performance for distributed block storage provided in an embodiment of the present invention
Analysis method.
It can be seen from the above technical proposal that a kind of performance evaluation work for distributed block storage provided by the present invention
Tool, method and storage medium, comprising: timestamp module, each key node for the IO stack in distributed block storage system
Upper setup time stamp;Time-consuming statistical module, for counting and calculating each timestamp when IO passes through corresponding timestamp
Corresponding time-consuming information;Abnormal IO stack output module, for when the total time-consuming of entire IO stack be more than preset threshold when, this is different
The corresponding time-consuming information output of each timestamp in the entire IO stack of normal IO;Performance management module, for providing time-consuming statistics
Information printing interface resets interface with time-consuming statistical information, clear with time-consuming statistical information to complete printing for time-consuming statistical information
Zero.
Timestamp module provides the Method of Data Organization of timestamp in the present invention;Time-consuming statistical module provides time-consuming letter
The statistics and calculating process of breath;Abnormal IO stack output module is provided when IO stack total time-consuming is more than preset threshold, this is different
The way of output of the entire IO stack of normal IO information time-consuming in detail;Performance management module provides statistical information Protection Counter Functions and statistics
Information printing function;Interaction in this way Jing Guo this four modules carries out time-consuming pipe to IO stack in distributed block storage system
Reason, may be implemented the performance evaluation to IO stack, is analyzed by the time-consuming to each stage, can be apparent from IO stack
Bottleneck point, meanwhile, abnormal IO information is exported, convenient for being carried out to the non-lethal abnormal performance in distributed block storage system
Trace analysis advanced optimizes performance of storage system to facilitate research staff.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is the structural schematic diagram of the performance analysis tool provided in an embodiment of the present invention for distributed block storage;
Fig. 2 is the concrete structure schematic diagram of each module in performance analysis tool provided in an embodiment of the present invention;
Fig. 3 is the flow diagram of entire IO stack in distributed block storage system provided in an embodiment of the present invention;
Fig. 4 is the flow chart of the method for analyzing performance provided in an embodiment of the present invention for distributed block storage.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The present invention provides a kind of performance analysis tool for distributed block storage, as shown in Figure 1, comprising the following modules:
Timestamp module 11, for setup time to stab on each key node of IO stack in distributed block storage system;
Time-consuming statistical module 12, for it is corresponding to count and calculate each timestamp when IO passes through corresponding timestamp
Time-consuming information;
Abnormal IO stack output module 13, for when the total time-consuming of entire IO stack be more than preset threshold when, by this exception IO
Entire IO stack in the corresponding time-consuming information output of each timestamp;
Performance management module 14 resets interface for providing time-consuming statistical information printing interface and time-consuming statistical information, with
Complete the printing of time-consuming statistical information and the clearing of time-consuming statistical information.
In the above-mentioned performance analysis tool for distributed block storage provided in an embodiment of the present invention, timestamp module is mentioned
The Method of Data Organization of timestamp is supplied;Time-consuming statistical module provides the statistics and calculating process of time-consuming information;Abnormal IO stack
Output module is provided when IO stack total time-consuming is more than preset threshold, by the entire IO stack of this exception IO information time-consuming in detail
The way of output;Performance management module provides statistical information Protection Counter Functions and statistical information printing function;Pass through this four in this way
The interaction of module carries out time-consuming management to IO stack in distributed block storage system, realizes the performance evaluation to IO stack, lead to
Cross to each stage time-consuming it is for statistical analysis, the case propagation delays in available each stage can be apparent from IO stack
Bottleneck point, meanwhile, abnormal IO information is exported, convenient for being carried out to the non-lethal abnormal performance in distributed block storage system
Trace analysis advanced optimizes performance of storage system to facilitate research staff.That is, passing through the performance evaluation
Tool realizes the performance evaluation of IO stack and exception IO tracking to distributed block storage system.
In the specific implementation, in the above-mentioned performance analysis tool for distributed block storage provided in an embodiment of the present invention
In, as shown in Fig. 2, in timestamp module 11, timestamp may include IO by each key node when execution moment and upper
It swims the upstream moment of timestamp, shows corresponding execution moment in the stages such as A, B, C and upstream moment in Fig. 2.Specifically, for
It can be its own by its corresponding upstream time stamp setting for first timestamp (i.e. the time started stabs);For last
It can be that first timestamp (is opened by its corresponding upstream time stamp setting for one timestamp (i.e. ending time stamp)
Beginning timestamp), the time-consuming of the last one timestamp in this way is the total time-consuming of entire IO stack.
In the specific implementation, in the above-mentioned performance analysis tool for distributed block storage provided in an embodiment of the present invention
In, as shown in Fig. 2, the corresponding time-consuming information of each timestamp may include entire IO stack each stage in time-consuming statistical module 12
Corresponding total degree, total time-consuming, average time-consuming and maximum are time-consuming, and the stages such as A, B, C corresponding total degree, total is shown in Fig. 2
Time-consuming, average time-consuming and maximum is time-consuming.When IO passes through corresponding timestamp, the statistics movement of time-consuming statistical module 12 is triggered, more
The statistical information of the new timestamp, including total time-consuming and total degree, and average time-consuming is calculated, refresh maximum time-consuming.Wherein, average
Time-consuming is the ratio of total time-consuming and total degree, i.e. average time-consuming=total time-consuming/total degree;When total degree is 0, average time-consuming is set
It is 0.That is, above-mentioned time-consuming statistical module specifically provides total time-consuming, the statistical method of total degree, average time-consuming is provided
Calculation method, and provide the update method of maximum time-consuming.
In the specific implementation, in the above-mentioned performance analysis tool for distributed block storage provided in an embodiment of the present invention
In, abnormal IO stack output module 13 may include: abnormal deciding means, for judging whether the total time-consuming of entire IO stack is more than pre-
If when threshold value, if then determining this IO exception;Information output unit is used for after the completion of this exception IO, suitable according to the time
The time-consuming information of each timestamp in this entire IO stack is successively output in log by sequence, by analyzing log, is obtained
The part being blocked in IO stack.
Specifically, the corresponding threshold value of total time-consuming that entire IO stack is preset by abnormal deciding means, when entire IO stack
Total time-consuming be more than the preset threshold when, after the completion of the IO, the timing node of the IO is successively printed into day with time sequencing
In will, by analyzing log, it can be clearly observed which part the IO stack being blocked belongs to, consequently facilitating from generation
Make deep analysis in terms of code and process, finally solves the problems, such as.
In the specific implementation, in the above-mentioned performance analysis tool for distributed block storage provided in an embodiment of the present invention
In, in performance management module 14, time-consuming statistical information printing interface is provided, it, can be by each timestamp by the interface
The information such as corresponding average time-consuming, maximum time-consuming, total degree, total time-consuming successively print;Also, provide time-consuming statistics letter
Breath resets interface, can be by corresponding average time-consuming, the maximum time-consuming of each timestamp, total degree, total time-consuming etc. by the interface
Information is reset.
Below by taking the process of IO stack entire in the distributed memory system shown in Fig. 3 as an example, 7 times are added for the process
Stamp is with its corresponding upstream timestamp such as following table one:
Table one
For going out team's time stamp, when IO goes out team, team's moment is recorded out, team out is subtracted into its upstream time the moment
Team's moment is protruded into, can be calculated group time-consuming cost3 of time stamp;
In time-consuming statistical module, the total degree count for going to out the timestamp at team's moment is added 1;Total time-consuming sum_
Time is plus the time-consuming cost3 for going out team's time stamp;Calculate average time-consuming;The timestamp recorded before cost3 is more than
When maximum time-consuming cost3_max, the maximum time-consuming cost3_max=cost3 of team's timestamp is enabled out;
When the time-consuming statistical information printing interface of invocation performance management module, by above-mentioned sum_time, count,
Avgcost3, cost3_max printout.
For finish time, upstream timestamp is start time, so the time-consuming cost7 of finish time is entire IO
The total time-consuming of stack.When total time-consuming is more than preset threshold value, when such as 10 seconds, according to time sequencing successively by 7 timestamps in table one
Time-consuming be output in log, in case subsequent performance evaluation.
Based on the same inventive concept, the embodiment of the invention also provides a kind of performance evaluation sides for distributed block storage
Method, since the principle that this method solves the problems, such as is similar to a kind of aforementioned performance analysis tool for distributed block storage,
The implementation of this method may refer to the implementation of the performance analysis tool stored for distributed block, and overlaps will not be repeated.
In the specific implementation, the method for analyzing performance provided in an embodiment of the present invention for distributed block storage, such as Fig. 4 institute
Show, specifically includes the following steps:
S401, by timestamp module in distributed block storage system setup time on each key node of IO stack
Stamp;
S402, when IO pass through corresponding timestamp when, it is corresponding with each timestamp is calculated by time-consuming statistical module counts
Time-consuming information;
S403, when entire IO stack total time-consuming is more than preset threshold, by abnormal IO stack output module by this exception IO
Entire IO stack in the corresponding time-consuming information output of each timestamp;
S404, the time-consuming statistical information printing interface of passage capacity management module and time-consuming statistical information reset interface, complete
At the printing of time-consuming statistical information and the clearing of time-consuming statistical information.
In the above-mentioned method for analyzing performance for distributed block storage provided in an embodiment of the present invention, pass through above-mentioned step
Suddenly, the performance evaluation of IO stack and exception IO tracking to distributed block storage system may be implemented.
In the specific implementation, in the above-mentioned method for analyzing performance for distributed block storage provided in an embodiment of the present invention
In, timestamp may include the upstream moment for executing moment and upstream timestamp when IO passes through each key node;In timestamp
It can be first timestamp by the corresponding upstream time stamp setting of first timestamp in module;Alternatively, it is also possible to will be last
The corresponding upstream time stamp setting of one timestamp is first timestamp, and the time-consuming of the last one timestamp in this way is as entire
The total time-consuming of IO stack.
In the specific implementation, in the above-mentioned method for analyzing performance for distributed block storage provided in an embodiment of the present invention
In, the corresponding time-consuming information of each timestamp may include corresponding total degree of entire IO stack each stage, total time-consuming, it is average time-consuming and
It is maximum time-consuming;Wherein, it is average time-consuming be total time-consuming and total degree ratio, i.e. average time-consuming=total time-consuming/total degree;When total time
When number is 0, average time-consuming is set to 0.
In the specific implementation, in the above-mentioned method for analyzing performance for distributed block storage provided in an embodiment of the present invention
In, step S403 is when entire IO stack total time-consuming is more than preset threshold, by abnormal IO stack output module by this exception IO's
The corresponding time-consuming information output of each timestamp, can specifically include: being judged first by abnormal deciding means whole in entire IO stack
When whether the total time-consuming of a IO stack is more than preset threshold, if then determining this IO exception;Then existed by information output unit
After the completion of this exception IO, the time-consuming information of each timestamp in this entire IO stack is successively output to by log according to time sequencing
In;It is analyzed finally by log, obtains the part being blocked in IO stack.
Further, the invention also discloses a kind of computer readable storage mediums, for storing computer program;It calculates
Machine program realizes the aforementioned disclosed performance analysis tool for distributed block storage when being executed by processor.
It can be with reference to corresponding contents disclosed in previous embodiment, herein no longer about the more specifical process of the above method
It is repeated.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For side disclosed in embodiment
For method, storage medium, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place ginseng
See method part illustration.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond scope of the present application.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
A kind of performance analysis tool, method and storage medium for distributed block storage provided in an embodiment of the present invention,
It include: timestamp module, for setup time to stab on each key node of IO stack in distributed block storage system;Time-consuming system
Module is counted, for when IO passes through corresponding timestamp, counting and calculating the corresponding time-consuming information of each timestamp;Abnormal IO stack
Output module, for when the total time-consuming of entire IO stack be more than preset threshold when, by each time in the entire IO stack of this exception IO
Stab corresponding time-consuming information output;Performance management module, for providing time-consuming statistical information printing interface and time-consuming statistical information
Interface is reset, to complete the printing of time-consuming statistical information and the clearing of time-consuming statistical information.Phase in this way Jing Guo this four modules
Interaction carries out time-consuming management to IO stack in distributed block storage system, the performance evaluation to IO stack may be implemented, by each
The time-consuming in a stage is analyzed, the bottleneck point that can be apparent from IO stack, meanwhile, abnormal IO information is exported, just
In carrying out trace analysis to the non-lethal abnormal performance in distributed block storage system, to facilitate research staff to storage system
System performance advanced optimizes.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
It is carried out above to provided by the present invention for the performance analysis tool of distributed block storage, method and storage medium
It is discussed in detail, used herein a specific example illustrates the principle and implementation of the invention, above embodiments
Explanation be merely used to help understand method and its core concept of the invention;At the same time, for those skilled in the art,
According to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion in this specification
Appearance should not be construed as limiting the invention.
Claims (10)
1. a kind of performance analysis tool for distributed block storage characterized by comprising
Timestamp module, for setup time to stab on each key node of IO stack in distributed block storage system;
Time-consuming statistical module, for counting and calculating the corresponding consumption of each timestamp when IO passes through corresponding timestamp
When information;
Abnormal IO stack output module, for when the total time-consuming of entire IO stack be more than preset threshold when, by the entire of this exception IO
The corresponding time-consuming information output of each timestamp in IO stack;
Performance management module resets interface for providing time-consuming statistical information printing interface and time-consuming statistical information, to complete to consume
When the printing of the statistical information and clearing of time-consuming statistical information.
2. the performance analysis tool according to claim 1 for distributed block storage, which is characterized in that the timestamp
Pass through the upstream moment for executing moment and upstream timestamp when each key node including IO.
3. the performance analysis tool according to claim 2 for distributed block storage, which is characterized in that described in first
The corresponding upstream timestamp of timestamp upstream timestamp corresponding with timestamp described in the last one is first time
Stamp.
4. the performance analysis tool according to claim 3 for distributed block storage, which is characterized in that each time
Stabbing corresponding time-consuming information includes corresponding total degree of entire IO stack each stage, total time-consuming, average time-consuming and maximum time-consuming;Its
In,
The average time-consuming ratio for the total time-consuming and total degree;When the total degree is 0, the average time-consuming is 0.
5. the performance analysis tool according to claim 1 for distributed block storage, which is characterized in that the exception IO
Stack output module includes:
Abnormal deciding means, when for judging whether the total time-consuming of entire IO stack is more than preset threshold, if then determining this IO
It is abnormal;
Information output unit is used for after the completion of this exception IO, according to time sequencing successively by each institute in this entire IO stack
The time-consuming information for stating timestamp is output in log, by analyzing the log, obtains the part being blocked in IO stack.
6. a kind of method for analyzing performance for distributed block storage characterized by comprising
By timestamp module, setup time is stabbed on each key node of IO stack in distributed block storage system;
When IO passes through corresponding timestamp, pass through time-consuming statistical module counts time-consuming corresponding with each timestamp is calculated
Information;
When entire IO stack total time-consuming is more than preset threshold, by abnormal IO stack output module by the entire IO stack of this exception IO
In the corresponding time-consuming information output of each timestamp;
The time-consuming statistical information printing interface of passage capacity management module and time-consuming statistical information reset interface, complete time-consuming statistics
The clearing of the printing of information and time-consuming statistical information.
7. the method for analyzing performance according to claim 6 for distributed block storage, which is characterized in that the timestamp
Pass through the upstream moment for executing moment and upstream timestamp when each key node including IO;
It is first time by the corresponding upstream time stamp setting of first timestamp in the timestamp module
Stamp;It is first timestamp by the corresponding upstream time stamp setting of the last one described timestamp.
8. the method for analyzing performance according to claim 6 for distributed block storage, which is characterized in that each time
Stabbing corresponding time-consuming information includes corresponding total degree of entire IO stack each stage, total time-consuming, average time-consuming and maximum time-consuming;Its
In,
The average time-consuming ratio for the total time-consuming and total degree;When the total degree is 0, the average time-consuming is 0.
9. the method for analyzing performance according to claim 6 for distributed block storage, which is characterized in that when entire IO stack
When total time-consuming is more than preset threshold, by abnormal IO stack output module by each timestamp in the entire IO stack of this exception IO
Corresponding time-consuming information output, specifically includes:
When judging whether the total time-consuming of entire IO stack is more than preset threshold by abnormal deciding means, if then determining that this IO is different
Often;
It, successively will be each described in this entire IO stack according to time sequencing through information output unit after the completion of this exception IO
The time-consuming information of timestamp is output in log, by analyzing the log, obtains the part being blocked in IO stack.
10. a kind of computer readable storage medium, which is characterized in that for storing computer program, wherein the computer journey
It is realized when sequence is executed by processor such as the described in any item method for analyzing performance for distributed block storage of claim 6 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910290061.9A CN109992481A (en) | 2019-04-11 | 2019-04-11 | A kind of performance analysis tool, method and storage medium for distributed block storage |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910290061.9A CN109992481A (en) | 2019-04-11 | 2019-04-11 | A kind of performance analysis tool, method and storage medium for distributed block storage |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109992481A true CN109992481A (en) | 2019-07-09 |
Family
ID=67133315
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910290061.9A Withdrawn CN109992481A (en) | 2019-04-11 | 2019-04-11 | A kind of performance analysis tool, method and storage medium for distributed block storage |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109992481A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110750432A (en) * | 2019-10-25 | 2020-02-04 | 北京浪潮数据技术有限公司 | IO performance analysis method and system of distributed storage system and related components |
CN111124308A (en) * | 2019-12-22 | 2020-05-08 | 浪潮电子信息产业股份有限公司 | Performance analysis method and device, electronic equipment and storage medium |
CN112000543A (en) * | 2020-07-29 | 2020-11-27 | 北京浪潮数据技术有限公司 | Method, device and equipment for detecting time delay performance of storage system |
-
2019
- 2019-04-11 CN CN201910290061.9A patent/CN109992481A/en not_active Withdrawn
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110750432A (en) * | 2019-10-25 | 2020-02-04 | 北京浪潮数据技术有限公司 | IO performance analysis method and system of distributed storage system and related components |
CN110750432B (en) * | 2019-10-25 | 2022-04-22 | 北京浪潮数据技术有限公司 | IO performance analysis method and system of distributed storage system and related components |
CN111124308A (en) * | 2019-12-22 | 2020-05-08 | 浪潮电子信息产业股份有限公司 | Performance analysis method and device, electronic equipment and storage medium |
CN112000543A (en) * | 2020-07-29 | 2020-11-27 | 北京浪潮数据技术有限公司 | Method, device and equipment for detecting time delay performance of storage system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109992481A (en) | A kind of performance analysis tool, method and storage medium for distributed block storage | |
CN108847994A (en) | Alarm localization method, device, equipment and storage medium based on data analysis | |
CN107302547A (en) | A kind of web service exceptions detection method and device | |
Naksinehaboon et al. | Reliability-aware approach: An incremental checkpoint/restart model in hpc environments | |
CN103365965B (en) | A kind of aggregation process method and apparatus of data | |
CN104598551B (en) | A kind of data statistical approach and device | |
CN103714004A (en) | JVM online memory leak analysis method and system | |
CN106649239A (en) | Method and device for generating report in cloud monitoring system based on visualization | |
CN109359160A (en) | Method of data synchronization, device, computer equipment and storage medium | |
CN106651344A (en) | Basic electricity fee error detection method and system based on power grid charging system | |
JP2007094850A (en) | Communication analyzing device and method | |
CN110225037A (en) | A kind of ddos attack detection method and device | |
CN101976394B (en) | Data acquiring and counting system and method | |
CN110262959A (en) | Underlying services method for testing pressure, device, electronic equipment and storage medium | |
CN107423110A (en) | A kind of virtual machine method of real-time and its device based on libvirt | |
US20100296411A1 (en) | Transaction Topology Discovery Using Constraints | |
WO2019029149A1 (en) | Insurance policy-type status information counting method, terminal apparatus and storage medium | |
CN115222259A (en) | Strategic target-based comprehensive performance management method and system | |
GB2492899A (en) | Method of modelling to-be virtualized infrastructures | |
KR101278135B1 (en) | Apparatus and method for assessing future impacts of patents based on stochastic patent citation analysis | |
CN105931142A (en) | Method for determining electric charge collection ranks | |
CN106201841A (en) | The statistical method of a kind of application program service data and device | |
CN110324211A (en) | A kind of data capture method and device | |
CN105302691B (en) | A kind of metadata method for monitoring performance and system | |
US20060174218A1 (en) | Method for controlling software modules |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190709 |