CN102945196B - Method and device for server IO (Input Output) diagnosis - Google Patents

Method and device for server IO (Input Output) diagnosis Download PDF

Info

Publication number
CN102945196B
CN102945196B CN201210422249.2A CN201210422249A CN102945196B CN 102945196 B CN102945196 B CN 102945196B CN 201210422249 A CN201210422249 A CN 201210422249A CN 102945196 B CN102945196 B CN 102945196B
Authority
CN
China
Prior art keywords
time
bio
time period
self
variance
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.)
Expired - Fee Related
Application number
CN201210422249.2A
Other languages
Chinese (zh)
Other versions
CN102945196A (en
Inventor
钱天进
晏臣
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Huawei Digital Technologies Co Ltd
Original Assignee
Hangzhou Huawei Digital Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hangzhou Huawei Digital Technologies Co Ltd filed Critical Hangzhou Huawei Digital Technologies Co Ltd
Priority to CN201210422249.2A priority Critical patent/CN102945196B/en
Publication of CN102945196A publication Critical patent/CN102945196A/en
Application granted granted Critical
Publication of CN102945196B publication Critical patent/CN102945196B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

The invention discloses a method and a device for server IO (Input Output) diagnosis, and relates to the technical field of servers, which achieves tracking of each bio processing procedure and can directly determine an IO path with a problem without removing influence factors one by one so as to improve the efficiency of IO diagnosis. The embodiment of the invention comprises the steps as follows: obtaining variances of scheduling time ranges, storage equipment processing time ranges and callback time ranges, which correspond to multiple bios by calculating; and comparing the obtained variances and determining the IO path with the problem according to a comparison result. The method and the device are mainly applied to the server IO diagnosis process.

Description

The method of server I/O diagnosis and device
Technical field
The present invention relates to server technology field, particularly relate to a kind of server I/O (English full name is: Input/Output Chinese full name is: I/O) method of diagnosing and device.
Background technology
IO subsystem in server system comprises: the memory device such as hardware and software relevant with IO in server apparatus and the disk array that is connected with server apparatus.When server I/O performance has problems, technical scheme that prior art provides needs may affecting CPU, internal memory, service application and the memory device is connected with server of server I/O performance and storing the factors such as networking and analyze, and verifies one by one and get rid of the impact of each factor until the IO path that goes wrong, location.A large amount of time need be spent, because this reducing IO diagnosis efficiency owing to verifying one by one and getting rid of each factor.
Summary of the invention
A kind of method that embodiments of the invention provide server I/O to diagnose and device, achieve to each bio that (English full name is: block input output, Chinese full name is: the input and output of block) tracking of processing procedure, the IO path gone wrong directly can be determined according to the variance that time period each in bio processing procedure is corresponding, need not go to get rid of influence factor one by one, thus improve the efficiency of IO diagnosis.
For achieving the above object, embodiments of the invention adopt following technical scheme:
A method for server I/O diagnosis, comprising:
Obtain the input and output bio of several blocks section of each self-corresponding scheduling time, storage device processes time period and readjustment time period;
Calculate according to several bio described section of each self-corresponding scheduling time, storage device processes time period and readjustment time period, obtain scheduling time section variance, storage device processes time period variance and readjustment time period variance;
By scheduling time section variance, storage device processes time period variance and readjustment time period variance compare, determine the IO path gone wrong according to comparative result.
A device for server I/O diagnosis, comprising:
Acquiring unit, the section of each self-corresponding scheduling time of the input and output bio for obtaining several blocks, storage device processes time period and readjustment time period;
Computing unit, several bio for obtaining according to described acquiring unit section of each self-corresponding scheduling time, storage device processes time period and readjustment time period calculate, and obtain scheduling time section variance, storage device processes time period variance and readjustment time period variance;
Comparing unit, compares for scheduling time section variance, storage device processes time period variance and the readjustment time period variance obtained by described computing unit, determines the IO path gone wrong according to comparative result.
The method of server I/O that the embodiment of the present invention provides diagnosis and device, by calculating the variance of several bio obtained section of each self-corresponding scheduling time, storage device processes time period and readjustment time period; And compare according to the variance value obtained, the IO path gone wrong is determined according to comparative result.Prior art, must verify and get rid of the impact of each factors such as CPU, internal memory, service application one by one until locate the IO path gone wrong.A large amount of time will be spent like this, because this reducing IO diagnosis efficiency.And technical scheme provided by the invention achieves the tracking to each bio processing procedure, the IO path gone wrong directly can be determined according to the variance that time period each in bio processing procedure is corresponding, need not go to get rid of influence factor one by one, thus improve the efficiency of IO diagnosis.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is the process flow diagram of a kind of server I/O diagnostic method in the embodiment of the present invention 1;
Fig. 2 is the process flow diagram of a kind of server I/O diagnostic method in the embodiment of the present invention 2;
Fig. 3 is the process flow diagram of a kind of server I/O diagnostic method in the embodiment of the present invention 3;
Fig. 4 is the composition frame chart of a kind of server I/O diagnostic device in the embodiment of the present invention 4;
Fig. 5 is the composition frame chart of the another kind of server I/O diagnostic device in the embodiment of the present invention 4;
Fig. 6 is the composition frame chart of a kind of server I/O diagnostic device in the embodiment of the present invention 5.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
Embodiment 1
Embodiments provide the method for a kind of server I/O diagnosis, as shown in Figure 1, the method comprises:
101, the input and output bio of several blocks section of each self-corresponding scheduling time, storage device processes time period and readjustment time period is obtained.
What deserves to be explained is, scheduling time section be bio from IO dispatch layer to the time of main this flow process of driving, the storage device processes time period is the processing time of bio in whole main driving, and the readjustment time period is bio from the main time being driven into the flow process of IO dispatch layer.
102, calculate according to several bio described section of each self-corresponding scheduling time, storage device processes time period and readjustment time period, obtain scheduling time section variance, storage device processes time period variance and readjustment time period variance.
103, by scheduling time section variance, storage device processes time period variance and readjustment time period variance compare, determine the IO path gone wrong according to comparative result.
Wherein, compare the variance of time period, namely compare the size of its numerical value, the IO path corresponding to the time period that comparative result determination maximum variance value is corresponding is the IO path gone wrong.
The method of server I/O that the embodiment of the present invention provides diagnosis, by calculating the variance of several bio obtained section of each self-corresponding scheduling time, storage device processes time period and readjustment time period; And the variance value obtained compares, determine the IO path gone wrong according to comparative result.Prior art, must verify and get rid of the impact of each factors such as CPU, internal memory, service application one by one until locate the IO path gone wrong.A large amount of time will be spent like this, because this reducing IO diagnosis efficiency.And technical scheme provided by the invention achieves the tracking to each bio processing procedure, the IO path gone wrong directly can be determined according to the variance that time period each in bio processing procedure is corresponding, need not go to get rid of influence factor one by one, thus improve the efficiency of IO diagnosis.
Embodiment 2
The method of the server I/O diagnosis that the embodiment of the present invention provides, as shown in Figure 2, the method comprises:
201, obtain the time that IO dispatch layer receives each bio, and the time described IO dispatch layer being received each bio is as each self-corresponding very first time value of each bio.
Wherein, it is realized by the dependent instruction specified in " submit_bio " function that bio is handed down to IO dispatch layer by generic block layer, what deserves to be explained is, the embodiment of the present invention sets up printing function in this function, print the system time that IO dispatch layer receives each bio respectively, be worth as the very first time using this time.
202, obtain the time that main driving starts to process each bio, and the time starting described main driving to process each bio is as each self-corresponding second time value of each bio.
What deserves to be explained is, in this step, by the dependent instruction specified in " scsi_dispatch_cmd " function, bio is converted into scsi command, and scsi command is handed down to main driving.What deserves to be explained is, the embodiment of the present invention sets up printing function in this function, prints system time scsi command being handed down to main driving respectively, and using this as each self-corresponding second time value of each bio.
203, obtain the time that main driving processes each bio, and the time described main driving being processed each bio is as each self-corresponding 3rd time value of each bio.
What deserves to be explained is, in this step, main driving performs the dependent instruction of " scsi_io_completion " function after completing bio operation.The present invention sets up printing function in this function, respectively the system time of " scsi_io_completion " function described in print execution, as each self-corresponding 3rd time value of each bio.
204, obtain the time that IO dispatch layer sends each bio, and the time described IO dispatch layer being sent each bio is as each self-corresponding 4th time value of each bio.
Wherein, IO dispatch layer sends bio and namely represents that whole bio completes, and after whole bio completes, performs " bio_endio " function.The present invention sets up printing function in this function, respectively the system time of " bio_endio " function described in print execution, as each self-corresponding 4th time value of each bio.
205, each for each bio self-corresponding very first time value and described second time value are made additive operation respectively, obtain each bio section of each self-corresponding scheduling time.
206, each for each bio self-corresponding second time value and described 3rd time value are made additive operation respectively, obtain each bio each self-corresponding storage device processes time period.
207, each for each bio self-corresponding 3rd time value and described 4th time value are made additive operation respectively, obtain each bio each self-corresponding readjustment time period.
What deserves to be explained is, the scheduling time section mentioned in above-mentioned steps 205-207, storage device processes time period, readjustment time period, its each self-corresponding IO path is identical with the associated description in step 101 in embodiment 1, is not repeated.
208, calculate according to several bio described section of each self-corresponding scheduling time, storage device processes time period and readjustment time period, obtain scheduling time section variance, storage device processes time period variance and readjustment time period variance.
209, more described scheduling time section variance, storage device processes time period variance and readjustment time period side extent, according to comparative result determination maximum variance value.
210, determine the time period that described maximum variance value is corresponding, and determine that path corresponding to time period corresponding to described maximum variance value is the IO path gone wrong.
What deserves to be explained is, wherein the time period is scheduling time section, the storage device processes time period in step 208 and adjusts back the time period.Time period is calculated by the difference of two time values to get, and the concrete acquisition of time period is originally being that in embodiment, step 205-207 has made related description, in this no longer repeated description.
In addition, the IO path that the time period in this step is corresponding, to be described in detail in step 101 in embodiment 1, is not repeated.
In addition, what deserves to be explained is, in main driving, user can arrange time delay by postponing interface function, thus can modify to normal bio treatment scheme, to get rid of the impact that memory device is diagnosed server I/O.
Further, the method to set up postponing interface function specifically comprises:
1. define global variable delay_time, this variable represents time delay.
2. defining interface function, the content of definition is: the interface function name being supplied to user; The title of function is set.The interface that user arranges time delay is supplied to by definition.
3. definition arranges function, arranges function and the time delay obtained by interface function is sent to global variable delay_time.
Particularly, time delay is for simulating the time delay of actual bio through memory device.Delay function postpones the time delay of setting, waits delay to terminate rear " scsi_io_completion " function that directly calls to change bio treatment scheme, makes the memory device that bio does not need through reality.Thus the impact of storer on IO path can be got rid of.
The method of server I/O that the embodiment of the present invention provides diagnosis, by calculating the variance of several bio obtained section of each self-corresponding scheduling time, storage device processes time period and readjustment time period; And the variance value obtained compares, determine the IO path gone wrong according to comparative result.Prior art, must verify and get rid of the impact of each factors such as CPU, internal memory, service application one by one until locate the IO path gone wrong.A large amount of time will be spent like this, because this reducing IO diagnosis efficiency.And technical scheme provided by the invention achieves the tracking to each bio processing procedure, the IO path gone wrong directly can be determined according to the variance that time period each in bio processing procedure is corresponding, need not go to get rid of influence factor one by one, thus improve the efficiency of IO diagnosis.
Embodiment 3
The invention provides the method for a kind of server I/O diagnosis, the method may be used in following example.This example is to obtain the concrete method flow of 5 bio introductions.Very first time value can be expressed as sub-time in this example, and the second time value can be expressed as dis-time, and the 3rd time value can be expressed as com-time, and the 4th time value can be expressed as end-time.Scheduling time, section was expressed as sub-dis-time, and the storage device processes time period is expressed as dis--com-time, and the readjustment time period is expressed as com-end-time.What deserves to be explained is, just change has been done to the method for expressing of time value and time period title herein, there is no other practical significance.In addition, very first time value, the second time value, the 3rd time value, the 4th time value are identical with the associated description of step 201-204 in embodiment 2; Scheduling time section, the storage device processes time period, readjustment the time period identical with the associated description of step 205-207 in embodiment 2, then no longer repeat specification in the present embodiment.As shown in Figure 3, the method comprises:
301,5 each self-corresponding very first time values of bio, the second time value, the 3rd time value, the 4th time values are obtained.
What deserves to be explained is, the implementation method of this step is corresponding with step 201-204 in embodiment 2, is just converted into by the uncertain number in embodiment 2 and determines number 5, so be not repeated.In addition, as following table 1, this example is described for the concrete numerical value of four kinds of times.
Table 1
302, additive operation is carried out to the very first time value in step 301, the second time value, the 3rd time value, the 4th time value.
Wherein, this step additive operation is corresponding with step 205-207 in embodiment 2, and concrete methods of realizing is identical, no longer describes in detail at this.
In addition, the concrete time value of provide for step 301 5 bio is obtained its corresponding difference, as shown in following table table 2:
Table 2
Remember that the sub-dis-time of bio1-bio5 is s1-s5 respectively, remember that the dis--com-time of bio1-bio5 is d1-d5 respectively, remember that the com-end-time of bio1-bio5 is c1-c5 respectively.
303, calculate according to described 5 bio each self-corresponding sub-dis-time, dis--com-time and com-end-time, obtain respective variance.
Wherein, calculating variance needs first to calculate above-mentioned three kinds of time periods each self-corresponding mean value, and the mean value that sub-dis-time is corresponding is designated as, and the mean value that dis--com-time is corresponding is designated as ad, and the mean value that com-end-time is corresponding is designated as ac.The occurrence of three kinds of time periods of 5 bio obtained according to step 302 in the present embodiment obtains the mean value of its correspondence, as shown in table 3 below:
Table 3
As follows according to mean value calculation variance process:
Scheduling time, section variance was:
1/5 [(s1-as) 2+ (s2-as) 2+ (s3-as) 2+ (s4-as) 2+ (s5-as) 2]=27.6 storage device processes time period variance is:
1/5 [(d1-ad) 2+ (d2-ad) 2+ (d3-ad) 2+ (d4-ad) 2+ (d5-ad) 2time period variance is adjusted back in]=0.8:
1/5 [(c1-ac) 2+ (c2-ac) 2+ (c 3-ac) 2+ (c4-ac) 2+ (c5-ac) 2]=0.8304, determine maximum variance value.
Wherein, maximum variance value position scheduling time section variance, namely by 27.6 maximum in three variance yields obtained in above-mentioned steps 303, corresponding scheduling time section variance.
305, determine that the IO path gone wrong is that IO dispatch layer is to main driving according to comparative result.
The method of server I/O that the embodiment of the present invention provides diagnosis, by calculating the variance of acquisition 5 bio section of each self-corresponding scheduling time, storage device processes time period and readjustment time period; And the variance value obtained compares, determine the IO path gone wrong according to comparative result.Prior art, must verify and get rid of the impact of each factors such as CPU, internal memory, service application one by one until locate the IO path gone wrong.A large amount of time will be spent like this, because this reducing IO diagnosis efficiency.And technical scheme provided by the invention achieves the tracking to each bio processing procedure, directly can determine the IO path that goes wrong, need not go to get rid of influence factor one by one according to comparative result, thus improve the efficiency of IO diagnosis.
Embodiment 4
The invention provides the device of a kind of server I/O diagnosis, as shown in Figure 4, this device comprises: acquiring unit 41, computing unit 42, comparing unit 43.
Acquiring unit 41, the section of each self-corresponding scheduling time of the input and output bio for obtaining several blocks, storage device processes time period and readjustment time period.
Computing unit 42, several bio for obtaining according to described acquiring unit 41 section of each self-corresponding scheduling time, storage device processes time period and readjustment time period calculate, and obtain scheduling time section variance, storage device processes time period variance and readjustment time period variance.
Comparing unit 43, compares for scheduling time section variance, storage device processes time period variance and the readjustment time period variance obtained by described computing unit 42, determines the IO path gone wrong according to comparative result.
Optionally, as shown in Figure 5, described acquiring unit 41 also comprises: acquisition module 411, computing module 412; Described comparing unit 43 also comprises: comparison module 431, determination module 432.
Acquisition module 411, receives the time of each bio for obtaining IO dispatch layer, and the time described IO dispatch layer being received each bio is as each self-corresponding very first time value of each bio; Start to process the time of each bio for obtaining main driving, and the time starting described main driving to process each bio is as each self-corresponding second time value of each bio; Process the time of each bio for obtaining main driving, and the time described main driving being processed each bio is as each self-corresponding 3rd time value of each bio; Obtain the time that IO dispatch layer sends each bio, and the time described IO dispatch layer being sent each bio is as each self-corresponding 4th time value of each bio.
Computing module 412, each self-corresponding very first time value of each bio for being got by described acquisition module 411 and described second time value make additive operation respectively, obtain each bio section of each self-corresponding described scheduling time; For each for each bio self-corresponding second time value and described 3rd time value are made additive operation respectively, obtain each bio each self-corresponding storage device processes time period; Also for each for each bio self-corresponding 3rd time value and described 4th time value are made additive operation respectively, obtain each bio each self-corresponding readjustment time period.
Comparison module 431, for more described scheduling time section variance, storage device processes time period variance and readjustment time period side extent.
Determination module 432, for the comparative result determination maximum variance value obtained according to described comparison module 431; Determine the time period that described maximum variance value is corresponding, and determine that path corresponding to time period corresponding to described maximum variance value is the IO path gone wrong.
The device of server I/O that the embodiment of the present invention provides diagnosis, by calculating the variance of several bio obtained section of each self-corresponding scheduling time, storage device processes time period and readjustment time period; And the variance value obtained compares, determine the IO path gone wrong according to comparative result.Prior art, must verify and get rid of the impact of each factors such as CPU, internal memory, service application one by one until locate the IO path gone wrong.A large amount of time will be spent like this, because this reducing IO diagnosis efficiency.And technical scheme provided by the invention achieves the tracking to each bio processing procedure, directly can determine the IO path that goes wrong, need not go to get rid of influence factor one by one according to comparative result, thus improve the efficiency of IO diagnosis.
Embodiment 5
The invention provides the device of a kind of server I/O diagnosis, as shown in Figure 6, this device comprises: processor 01, storer 02.
Processor 01, receives the time of each bio for obtaining IO dispatch layer, and the time described IO dispatch layer being received each bio is as each self-corresponding very first time value of each bio.
Storer 02, for storing each self-corresponding very first time value of described each bio.
Processor 01, starts to process the time of each bio for obtaining main driving, and the time starting described main driving to process each bio is as each self-corresponding second time value of each bio.
Storer 02, for storing each self-corresponding second time value of described each bio.
Processor 01, processes the time of each bio for obtaining main driving, and the time described main driving being processed each bio is as each self-corresponding 3rd time value of each bio.
Storer 02, for storing each self-corresponding 3rd time value of described each bio.
Processor 01, sends the time of each bio for obtaining IO dispatch layer, and the time described IO dispatch layer being sent each bio is as each self-corresponding 4th time value of each bio.
Storer 02, for storing each self-corresponding 4th time value of described each bio.
Processor 01, for each for each bio self-corresponding very first time value and described second time value are made additive operation respectively, obtains each bio section of each self-corresponding scheduling time.
Storer 02, for storing described each bio section of each self-corresponding scheduling time.
Processor 01, for each for each bio self-corresponding second time value and described 3rd time value are made additive operation respectively, obtains each bio each self-corresponding storage device processes time period.
Storer 02, for storing described each bio each self-corresponding storage device processes time period.
Processor 01, for each for each bio self-corresponding 3rd time value and described 4th time value are made additive operation respectively, obtains each bio each self-corresponding readjustment time period.
Storer 02, for storing described each bio each self-corresponding readjustment time period.
Processor 01, for calculating according to several bio described section of each self-corresponding scheduling time, storage device processes time period and readjustment time period, obtains scheduling time section variance, storage device processes time period variance and readjustment time period variance.
Storer 02, for storing described scheduling time section variance, storage device processes time period variance and readjustment time period variance.
Processor 01, for more described scheduling time section variance, storage device processes time period variance and readjustment time period side extent, according to comparative result determination maximum variance value.
Storer 02, for storing described maximum variance value.
Processor 01, for determining the time period that described maximum variance value is corresponding, and determines that path corresponding to time period corresponding to described maximum variance value is the IO path gone wrong.
Storer 02, for the IO path gone wrong described in storing.
Optionally, processor 01, also for arranging delay interface function.
Storer 02, also for storing described delay interface function.
The device of server I/O that the embodiment of the present invention provides diagnosis, by calculating the variance of several bio obtained section of each self-corresponding scheduling time, storage device processes time period and readjustment time period; And the variance value obtained compares, determine the IO path gone wrong according to comparative result.Prior art, must verify and get rid of the impact of each factors such as CPU, internal memory, service application one by one until locate the IO path gone wrong.A large amount of time will be spent like this, because this reducing IO diagnosis efficiency.And technical scheme provided by the invention achieves the tracking to each bio processing procedure, directly can determine the IO path that goes wrong, need not go to get rid of influence factor one by one according to comparative result, thus improve the efficiency of IO diagnosis.
Through the above description of the embodiments, those skilled in the art can be well understood to the mode that the present invention can add required common hardware by software and realize, and can certainly pass through hardware, but in a lot of situation, the former is better embodiment.Based on such understanding, technical scheme of the present invention can embody with the form of software product the part that prior art contributes in essence in other words, this computer software product is stored in the storage medium that can read, as the floppy disk of computing machine, hard disk or CD etc., comprise some instructions and perform method described in each embodiment of the present invention in order to make a computer equipment (can be personal computer, server, or the network equipment etc.).
The above; be only the specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, is anyly familiar with those skilled in the art in the technical scope that the present invention discloses; change can be expected easily or replace, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of described claim.

Claims (6)

1. a method for server I/O diagnosis, is characterized in that, comprising:
Obtain the input and output bio of several blocks section of each self-corresponding scheduling time, storage device processes time period and readjustment time period, described scheduling time, section was that bio is from IO dispatch layer to the time of main this flow process of driving, the described storage device processes time period is the processing time of bio in whole described main driving, and the described readjustment time period is bio is driven into the flow process of described IO dispatch layer time from described master;
Calculate according to the input and output bio of several blocks described section of each self-corresponding scheduling time, storage device processes time period and readjustment time period, obtain scheduling time section variance, storage device processes time period variance and readjustment time period variance;
By scheduling time section variance, storage device processes time period variance and readjustment time period variance compare, determine the IO path gone wrong according to comparative result.
2. the method for server I/O according to claim 1 diagnosis, is characterized in that, the input and output bio of described several blocks of acquisition section of each self-corresponding scheduling time, storage device processes time period and readjustment time period, specifically comprises:
Obtain the time that IO dispatch layer receives each bio, and the time described IO dispatch layer being received each bio is as each self-corresponding very first time value of each bio;
Obtain the time that main driving starts to process each bio, and the time starting described main driving to process each bio is as each self-corresponding second time value of each bio;
Obtain the time that main driving processes each bio, and the time described main driving being processed each bio is as each self-corresponding 3rd time value of each bio;
Obtain the time that IO dispatch layer sends each bio, and the time described IO dispatch layer being sent each bio is as each self-corresponding 4th time value of each bio;
Each for each bio self-corresponding very first time value and described second time value are made additive operation respectively, obtains each bio section of each self-corresponding scheduling time;
Each for each bio self-corresponding second time value and described 3rd time value are made additive operation respectively, obtains each bio each self-corresponding storage device processes time period;
Each for each bio self-corresponding 3rd time value and described 4th time value are made additive operation respectively, obtains each bio each self-corresponding readjustment time period.
3. the method for server I/O diagnosis according to claim 1, it is characterized in that, described by scheduling time section variance, storage device processes time period variance and readjustment time period variance compare, determine the IO path gone wrong according to comparative result, specifically comprise:
More described scheduling time section variance, storage device processes time period variance and readjustment time period side extent, according to comparative result determination maximum variance value;
Determine the time period that described maximum variance value is corresponding, and determine that path corresponding to time period corresponding to described maximum variance value is the IO path gone wrong.
4. a device for server I/O diagnosis, is characterized in that, comprising:
Acquiring unit, input and output bio for obtaining several blocks section of each self-corresponding scheduling time, storage device processes time period and readjustment time period, described scheduling time, section was that bio is from IO dispatch layer to the time of main this flow process of driving, the described storage device processes time period is the processing time of bio in whole described main driving, and the described readjustment time period is bio is driven into the flow process of described IO dispatch layer time from described master;
Computing unit, input and output bio for several blocks obtained according to described acquiring unit section of each self-corresponding scheduling time, storage device processes time period and readjustment time period calculate, and obtain scheduling time section variance, storage device processes time period variance and readjustment time period variance;
Comparing unit, compares for scheduling time section variance, storage device processes time period variance and the readjustment time period variance obtained by described computing unit, determines the IO path gone wrong according to comparative result.
5. device according to claim 4, is characterized in that, described acquiring unit comprises:
Acquisition module, receives the time of each bio for obtaining IO dispatch layer, and the time described IO dispatch layer being received each bio is as each self-corresponding very first time value of each bio; Start to process the time of each bio for obtaining main driving, and the time starting described main driving to process each bio is as each self-corresponding second time value of each bio; Process the time of each bio for obtaining main driving, and the time described main driving being processed each bio is as each self-corresponding 3rd time value of each bio; Obtain the time that IO dispatch layer sends each bio, and the time described IO dispatch layer being sent each bio is as each self-corresponding 4th time value of each bio;
Computing module, each self-corresponding very first time value of each bio for being got by described acquisition module and described second time value make additive operation respectively, obtain each bio section of each self-corresponding described scheduling time; For each for each bio self-corresponding second time value and described 3rd time value are made additive operation respectively, obtain each bio each self-corresponding storage device processes time period; Also for each for each bio self-corresponding 3rd time value and described 4th time value are made additive operation respectively, obtain each bio each self-corresponding readjustment time period.
6. the device of server I/O diagnosis according to claim 4, it is characterized in that, described comparing unit comprises:
Comparison module, for more described scheduling time section variance, storage device processes time period variance and readjustment time period side extent;
Determination module, for the comparative result determination maximum variance value obtained according to described comparison module; Determine the time period that described maximum variance value is corresponding, and determine that path corresponding to time period corresponding to described maximum variance value is the IO path gone wrong.
CN201210422249.2A 2012-10-29 2012-10-29 Method and device for server IO (Input Output) diagnosis Expired - Fee Related CN102945196B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210422249.2A CN102945196B (en) 2012-10-29 2012-10-29 Method and device for server IO (Input Output) diagnosis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210422249.2A CN102945196B (en) 2012-10-29 2012-10-29 Method and device for server IO (Input Output) diagnosis

Publications (2)

Publication Number Publication Date
CN102945196A CN102945196A (en) 2013-02-27
CN102945196B true CN102945196B (en) 2015-02-04

Family

ID=47728144

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210422249.2A Expired - Fee Related CN102945196B (en) 2012-10-29 2012-10-29 Method and device for server IO (Input Output) diagnosis

Country Status (1)

Country Link
CN (1) CN102945196B (en)

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3844490B2 (en) * 1993-07-19 2006-11-15 ブリテイッシュ・テレコミュニケーションズ・パブリック・リミテッド・カンパニー Method and apparatus for detecting and detecting errors in video images
US7290180B2 (en) * 2004-09-02 2007-10-30 International Business Machines Corporation Method to use an alternate I/O debug path
CN101616174B (en) * 2009-07-09 2012-07-11 浪潮电子信息产业股份有限公司 Method for optimizing system performance by dynamically tracking IO processing path of storage system
CN201540532U (en) * 2009-11-27 2010-08-04 英业达股份有限公司 Server system
CN102073605B (en) * 2010-12-27 2012-11-21 深圳市创新科信息技术有限公司 Method for storage interface bypassing Bio layer to access disk drive
JP5549622B2 (en) * 2011-02-23 2014-07-16 富士通セミコンダクター株式会社 Failure diagnosis method, failure diagnosis apparatus, test system and program

Also Published As

Publication number Publication date
CN102945196A (en) 2013-02-27

Similar Documents

Publication Publication Date Title
CN104246767A (en) Telemetry system for a cloud synchronization system
CN111143128B (en) Openstack cloud host data protection method and system
CN104572422A (en) Memory monitoring achievement method based on startup and shutdown of Linux system
US20200379748A1 (en) Upgrading method and apparatus
CN107644063A (en) Time series analysis method and system based on data parallel
CN104731678A (en) RAID storage mode test system and method and electronic device
CN102243506B (en) Method and device for acquiring system time
CN110737594A (en) Database standard conformance testing method and device for automatically generating test cases
CN105488692A (en) Method and device for computing number of online users
CN109243518A (en) A kind of test method for solid state hard disk, device and computer storage medium
CN104991836A (en) Database recovery method and apparatus and database server
CN102945196B (en) Method and device for server IO (Input Output) diagnosis
US8892956B2 (en) Electronic device and method for managing test results
EP2942714A2 (en) Monitoring method, monitoring apparatus, and electronic device
US11443191B2 (en) Computing device and parameter synchronization method implemented by computing device
CN107423488B (en) Automatic generating system and method for plane jig frame diagram
CN111984624A (en) Method and system for data migration through correction migration model
CN114443356A (en) Container cluster processing method and system
CN110806981B (en) Application program testing method, device, equipment and storage medium
CN105761119A (en) On-line number of people distribution calculation method and device
CN106775451A (en) A kind of method and device for processing logical volume
CN104699572A (en) Computer hardware diagnosis method based on UNIX operating system
CN104699593A (en) Computer software diagnosis method based on UNIX operating system
CN114116244A (en) Data processing method, device and system, electronic equipment and storage medium
CN105242880A (en) Method for automatically creating Raid and executing pressure testing

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20150204

Termination date: 20191029

CF01 Termination of patent right due to non-payment of annual fee