CN102346710A - Dynamic stubbing technology based time-delay analysis method for data packet processing - Google Patents
Dynamic stubbing technology based time-delay analysis method for data packet processing Download PDFInfo
- Publication number
- CN102346710A CN102346710A CN2011103092586A CN201110309258A CN102346710A CN 102346710 A CN102346710 A CN 102346710A CN 2011103092586 A CN2011103092586 A CN 2011103092586A CN 201110309258 A CN201110309258 A CN 201110309258A CN 102346710 A CN102346710 A CN 102346710A
- Authority
- CN
- China
- Prior art keywords
- machine
- function
- packet
- kernel
- pitching pile
- 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
Images
Abstract
The invention discloses a dynamic stubbing technology based time-delay analysis method for data packet processing. The method comprises the following five main steps of: 1, collecting a kernel symbol table and acquiring a function list capable of being stubbed; 2, determining a function list to be stubbed; 3, configuring a stubbing module, inserting a section of codes into each function to be stubbed, and adding function addresses and time stamps for function calling to a current data packet under processing; 4, starting a packet sending and receiving module; and 5, counting and analyzing the data packets stored in the step 4 to obtain the time consumed by each function of a kernel. The method disclosed by the invention has the advantages of less expense and very strong instantaneity and adaptability, can be used for elaborately analyzing the expense brought by each function and time delay for the data packets to wait in a queue in a process of treating the data packets by the Linux kernel, can provide a powerful basis for network analysis and optimization, and has a better practical value and a wide application prospect in the technical field of computers.
Description
(1) technical field
The present invention relates to a kind of processing data packets Delay Analysis based on dynamic pitching pile technology; Be specifically related to the method that a kind of utilization is analyzed the processing delay of each function of operating system nucleus in the packet processing procedure based on dynamic pitching pile technology, belong to field of computer technology.
(2) background technology
Along with the enriching constantly of network application, network distributes more and more wider in recent years, and network traffics are also increasing, has proposed new demand and challenge for the network performance of data center.In order to tackle this challenge, data center is necessary network structure is adjusted, and improves the processing power of server.Yet, present many data centers for the utilization factor that improves server, simplify server the availability that administers and maintains, improves application program, reduce cost, begin to adopt Intel Virtualization Technology mostly.Intel Virtualization Technology has a lot of advantages, but also can introduce additional overhead, especially I/O virtual aspect.Under virtualized environment, the processing procedure of packet is complicated more, and transmission path is longer, makes that overall network performance is very poor.
For to optimizing network performance, the bottleneck that just must know system where, which link expense of system is maximum, improves targetedly then.Have many people how to improve the network performance under the virtualized environment in research at present, these researchs are from virtualized principle mostly, attempt to reduce virtualized expense, improve the network bandwidth, reduce network delay.Yet the analysis that these researchs not do not quantize the processing delay of packet does not have instrument can analyze packet automatically from receiving the time delay of sending each function the entire process process at present yet.The present invention is based on dynamic pitching pile technology, designed the method for the time delay of a kind of each function of analysis operation system kernel automatically in packet transmission and receiving course.
The pitching pile technology is a kind of utilization widely that in software test, has, and the GCC compiler also provides function corresponding to be used for the test code coverage rate, and some operating system is also supported the pitching pile technology, like the DTrace of Solaris, the Kprobe of Linux etc.Dynamically a then nearer step of pitching pile technology, certain position that can in program run, dynamically one section code be inserted into program is used for the behavior of acquisition of information or reprogramming, and need not change to original program.The present invention just is based on that dynamic pitching pile technology analyzes operating system processing data packets time delay.
(3) summary of the invention
1, purpose
The purpose of this invention is to provide a kind of processing data packets Delay Analysis based on dynamic pitching pile technology; It can count the time that each function consumed of operating system nucleus in the processing data packets process and the time of in formation, waiting in line, for the further network optimization provides strong foundation.
2, technical scheme
Technical scheme of the present invention is following: a kind of processing data packets Delay Analysis based on dynamic pitching pile technology; Its overall framework as shown in Figure 1; Machine among Fig. 1 can be real computing machine; Also can be through the virtual virtual machine that comes out of virtualization software; Machine 1 among Fig. 1 is mainly divided from view of function with machine 2, and they can be deployed on the same physical machine simultaneously.The major function of machine 1 provides the control interface; Analyze also output result, machine 1 deploy give out a contract for a project module and statistical module, the major function of machine 2 is acquisition of informations; And give machine 1 with the information feedback collected, machine 2 deploy pitching pile module and packet receiving module.Idiographic flow as shown in Figure 2, a kind of processing data packets Delay Analysis of the present invention based on dynamic pitching pile technology, this method may further comprise the steps:
Wherein, Step 201 is collected the kernel symbol table on machine 2 be the symbol table that will collect the kernel that is moving; Kernel function and the address of variable in internal memory have been write down in the kernel symbol table; Under linux system; Through read/the boot catalogue under with the corresponding System.map of kernel version or directly read/proc/ksyms just can get access to the kernel symbol table; In addition, the kernel symbol table that can also obtain through analysis kernel file or kernel Debugging message.Because the address in internal memory of not only having write down kernel function in the kernel symbol table; Also write down the address of kernel variable; Therefore need analyze the kernel symbol table that reads; Filter out the interior nuclear symbol of those non-functions; Export the symbol table of all kernel function; These functions are exactly the function list that can carry out pitching pile, after confirming, it are passed to machine 1.
Wherein, step 202 according to the target of optimizing, need to confirm the function of analysis on machine 1, and output needs the function list of pitching pile.When machine 1 receive machine 2 can the function list of pitching pile after, can be manually or need to confirm the function list of pitching pile automatically.Manually confirming need be familiar for the kernel processes process, knows which function kernel can call and come handle packet, and the manual then function that will need pitching pile writes the function list that file formation needs pitching pile.Except manually confirming, also can write a little steps and come to confirm automatically through the method for name coupling.Certainly manually confirm to reduce the quantity of pitching pile, can reduce the influence of pitching pile network performance.
Wherein, Step 203 at first will be delivered to the determined function list of step 202 on the machine 2; The dynamic pitching pile technology of 2 utilizations is inserted a bit of code at the head of all functions that need monitor on machine then; Insert after the completion; Can at first carry out the code of insertion when kernel is carried out this function, then carry out original function.The function of the code that inserts is that the current packet of handling is detected, and then the packet back is added in the address of the current function of carrying out and current timestamp if satisfy some condition.
Wherein, step 204 can activate the packet receiving module of the give out a contract for a project module and the machine 2 of machine 1 after having disposed the pitching pile module.The packet receiving module of the give out a contract for a project module and the machine 2 of machine 1 all is the application program that operates under user's attitude.After being activated, the module of giving out a contract for a project of machine 1 can send packet to the packet receiving module of machine 2 according to certain speed that is configured to of user; The structure of packet as shown in Figure 3; Packet comprises two parts information, the metamessage partial record altogether size of packet, data message side-play amount and need the information such as side-play amount of insertion information at present; Data message is then simple relatively, is exactly the form of a function address and timestamp, the each end that all function address and timestamp is written to data message of pitching pile module.The data message of the packet that the module of giving out a contract for a project of machine 1 is sent is partly used zero padding; Packet arrives after the machine 2; Be arranged in the pitching pile module of the kernel timestamp can be with the address of the function of needs monitoring and this function of kernel calls the time and write the data message part of packet, and revise the metamessage of packet.After data packet delivery arrived the packet receiving module, the packet receiving module can not revised packet, but directly packet was returned to the module of giving out a contract for a project of machine 1.
Wherein, The module of giving out a contract for a project of step 205 machine 1 receives after the packet of machine 2 returns; Earlier packet is got up with the stored in form of file; The kernel symbol table collected of integrating step 201 again after the end; Through analyzing the data message of packet; Can calculate kernel and in the processing data packets process, call which function, and the time that each function consumed.If the processing data packets process for kernel is familiar,, can also analyze the queuing delay of packet in processing procedure through checking timestamp into formation and dequeue.
3, advantage and effect
A kind of processing data packets Delay Analysis of the present invention based on dynamic pitching pile technology; It compared with prior art; Its main advantage is: the processing procedure of the tracking data bag that (1) the present invention can be omnidistance; The time that each function consumed in reception of detail analysis packet and the process of transmitting, and the time of in formation, lining up and being consumed.(2) because the transmission rate of packet is that the user is controlled, can effectively control the expense that the pitching pile module is brought through the transmission rate of regulating packet, therefore total system overhead is very little.(3) the present invention has good real-time and adaptability, and real-time is meant that the result of packet time-delay analysis has reflected the situation of the processing data packets time delay under the current environment.Yet along with the variation of offered load, the time delay of packet also can change a lot, because the system overhead that the present invention brought is little, even therefore under the situation of heavy offered load, also can use, so have very strong adaptability.
(4) description of drawings
The overall framework synoptic diagram of Fig. 1 processing data packets Delay Analysis of the present invention
The schematic flow sheet of Fig. 2 processing data packets Delay Analysis of the present invention
The structural representation of the packet that Fig. 3 processing data packets Delay Analysis of the present invention is used
Symbol description is following among the figure:
The 201-205 step number.
(5) embodiment
For making the object of the invention, technical scheme and advantage express clearlyer, the present invention is remake further detailed explanation below in conjunction with accompanying drawing and specific embodiment.
Main thought of the present invention is address and the invoked timestamp information that function relevant with processing data packets in the kernel was collected and write down to the dynamic pitching pile technology of utilization; Again through the statistics and analysis to these information draw each function of kernel packet send with receiving course in time of being consumed reach the time of in formation, lining up and being consumed, thereby the detail analysis data are provided for the network optimization.
The present invention is provided with two machines, and promptly machine 1 all is real computing machine with 2, two machines of machine; Concrete model is the perfectly sound R510 server of association; As shown in Figure 1, machine 1 is mainly divided from view of function with machine 2, and they can be deployed on the same physical machine simultaneously.The major function of machine 1 provides the control interface; Analyze also output result, machine 1 deploy give out a contract for a project module and statistical module, the major function of machine 2 is acquisition of informations; And give machine 1 with the information feedback collected, machine 2 deploy pitching pile module and packet receiving module.
The present invention requires each machine to adopt (SuSE) Linux OS on software, and the kernel of the (SuSE) Linux OS on the machine 2 should be 2.6.18 or version thereafter, so that support the kprobe technology.Concrete enforcement roughly is divided into three phases to be accomplished, and first stage is initialization, and main work is to obtain the kernel symbol table, need to confirm the function of pitching pile; Second stage is in machine 2 deploy pitching pile modules, owing to collect and record function address and timestamp; Three phases is the bag that begins to transmit and receive data, and the result is carried out statistics and analysis, obtains a result.
Describe with an instance below, as shown in Figure 2, a kind of processing data packets Delay Analysis of the present invention based on dynamic pitching pile technology, these method concrete steps are following:
Step 201: collect the kernel symbol table of machine 2, the method for collection be check earlier/whether proc/ksyms exist, if exist, then its content copied to a file, pass to machine 1 then and get final product.If there is not then check/whether exists under the boot catalogue System.map file with the kernel coupling of moving; If have then it passed to machine 1; If not then need analysis kernel file or kernel Debugging message to obtain, if still can not obtain then collect the failure of kernel symbolic information.Get access to and also need filter out the kernel symbolic information of those non-functions after the kernel symbol table, thus obtain can pitching pile function list.
Step 202: if the kernel symbolic information of having collected machine 2 of step 201 success, then can obtain one can pitching pile function list, need to confirm the function list of pitching pile then through manual or automatic method.
Step 203: need to have confirmed after the function of pitching pile; Need the function passes of pitching pile to give machine 2 those; Machine 2 acquisitions need after the function list of pitching pile; Ensuing work is exactly to need the head of the function of pitching pile to add one section code at those, the timestamp when address and the function that is used for writing down function is called.This need be machine-processed by means of the kprobe that Linux provides, and kprobe mechanism can be inserted one section code in any position of kernel be used for collecting the relevant information of kernel.For the convenience of disposing can adopt SystemTap, SystemTap is based on a kind of script of kprobe, and it can translate into script a C file, and is compiled into kernel module, adds in the kernel dynamically.Only need provide function name and the code segment that needs to insert to get final product during use to it.Therefore, get access to after the function list that needs pitching pile, can accomplish the work of pitching pile through SystemTap easily.
Step 204: start the packet receiving module and give out a contract for a project module.Accomplish after the pitching pile work, just can start the machine 2 packet receiving module, its can be on certain port wait come from the packet of machine 1.Start after the packet receiving module; On machine 1, start the module of giving out a contract for a project, machine 1 can send packet for machine 2 with certain speed, also receives machine 2 return data bags simultaneously; And packet got up with the stored in form of file, so that statistics and analysis can be continued in the back.
Step 205: the packet that will receive machine 2 in the step 204 has stored with file, and this step mainly is exactly to read this document, and statistics and analysis goes out the time that is consumed of each function, and the output result.
Fig. 3 is the structural representation of the packet of processing data packets Delay Analysis use of the present invention
It should be noted last that: above embodiment is the unrestricted technical scheme of the present invention in order to explanation only; Although the present invention is described in detail with reference to the foregoing description; Those of ordinary skill in the art is to be understood that: still can make amendment or be equal to replacement the present invention; And replace any modification or the part that do not break away from the spirit and scope of the present invention, and it all should be encompassed in the middle of the claim scope of the present invention.
Claims (6)
1. processing data packets Delay Analysis based on dynamic pitching pile technology, it is characterized in that: these method concrete steps are following:
Step 1. on machine 2, collect the kernel symbol table, obtain the corresponding relation of kernel function and address, output can be carried out the function list of pitching pile, and passes to machine 1;
Step 2. on machine 1, according to the target of optimizing, need to confirm the function of analysis, output needs the function list of pitching pile, and passes to machine 2;
Step 3. the determined function list of step 2 is delivered on the machine 2, and the dynamic pitching pile technology of machine 2 utilizations is inserted one section code at the head of the function of needs monitoring, and this code can write packet with the address of function and current timestamp;
Step 4. start the packet receiving module and give out a contract for a project module; Machine 1 begins to send packet, and data packet delivery arrives after the kernel of machine 2, is the packet receiving module of machine 2 if detect the destination of packet, and just address and the timestamp with the current function of carrying out writes packet; When packet arrived the packet receiving module of machine 2, the packet receiving module was returned to machine 1 with packet, and the module of giving out a contract for a project of machine 1 will be from the packet of machine 2 stored in form with file;
Step 5. machine 1 all packets from machine 2 receiver modules of statistics, calculate and the output statistics;
Wherein, described machine 1 is real computing machines with machine 2, or through the virtual virtual machine that comes out of virtualization software, they are deployed on the same physical machine simultaneously; The function of machine 1 provides the control interface, analyzes also output result, machine 1 deploy give out a contract for a project module and statistical module, the function of machine 2 is acquisition of informations, and gives machine 1 with the information feedback of collecting, machine 2 deploy pitching pile module and packet receiving module.
(2) as claimed in claim 1, wherein the stub technique based on dynamic packet processing delay analysis method characterized in that: said step of one of the "2 in the machine to collect the kernel symbol table" means: collection is running kernel symbol table, the kernel symbol table records the kernel functions and variables in memory address, the Linux system, by reading the / boot directory with the corresponding System.map kernel version or direct read / proc / ksyms on access to the kernel symbol table, in addition, also by analyzing the core file or kernel debugging information to get the kernel symbol table; Since the kernel symbol table only records the kernel function in memory address, also recorded the kernel variable address, so the need for read kernel symbol table for analysis, filtering out those non-core function symbols, the output symbol table of all the kernel function, which is able to carry out these functions staked list of functions is determined, it is passed to the machine 1.
3. a kind of processing data packets Delay Analysis according to claim 1 based on dynamic pitching pile technology; It is characterized in that: described in the step 2 " on machine 1 according to the target of optimizing; need to confirm the function analyzed; output needs the function list of pitching pile; and pass to machine 2 " be meant: after machine 1 is received the function list of ability pitching pile of machine 2, manually or automatically need to confirm the function list of pitching pile; Manually confirm and to be familiar with for the kernel processes process, know which function kernel can call and come handle packet, and the manual then function that will need pitching pile writes the function list that file formation needs pitching pile; Except manually confirming, also can write a little steps and come to confirm automatically through the method for name coupling.
4. a kind of processing data packets Delay Analysis according to claim 1 based on dynamic pitching pile technology; It is characterized in that: " the determined function list of step 2 being delivered on the machine 2; the dynamic pitching pile technology of machine 2 utilizations is inserted one section code at the head of the function that needs are monitored; this code can write packet with the address of function and current timestamp " described in the step 3 is meant: need the head of the function of monitoring to insert after one section code completion; Can at first carry out the code of insertion when kernel is carried out this function, then carry out original function; The function of the code that inserts is that the current packet of handling is detected, and then the packet back is added in the address of the current function of carrying out and current timestamp if satisfy condition.
5. a kind of processing data packets Delay Analysis based on dynamic pitching pile technology according to claim 1 is characterized in that: " start the packet receiving module and give out a contract for a project module " described in the step 4 is meant: the packet receiving module of the give out a contract for a project module and the machine 2 of machine 1 all is the application program that operates under user's attitude; After being activated, the module of giving out a contract for a project of machine 1 can send packet to the packet receiving module of machine 2 according to certain speed that is configured to of user; Packet comprises two parts information, the metamessage partial record altogether size of packet, data message side-play amount and need the side-play amount of insertion information at present; Data message is exactly the form of a function address and timestamp, the each end that all function address and timestamp is written to data message of pitching pile module; The data message of the packet that the module of giving out a contract for a project of machine 1 is sent is partly used zero padding; Packet arrives after the machine 2; Be arranged in the pitching pile module of the kernel timestamp can be with the address of the function of needs monitoring and this function of kernel calls the time and write the data message part of packet, and revise the metamessage of packet; After data packet delivery arrived the packet receiving module, the packet receiving module can not revised packet, but directly packet was returned to the module of giving out a contract for a project of machine 1.
6. a kind of processing data packets Delay Analysis according to claim 1 based on dynamic pitching pile technology; It is characterized in that: " machine 1 all packets from machine 2 receiver modules of statistics; calculate and the output statistics " described in the step 5 are meant: the module of giving out a contract for a project of machine 1 receives after the packet of machine 2 returns; Earlier packet is got up with the stored in form of file; The kernel symbol table collected of integrating step two again after the end; By analyzing the data message of packet; Calculate kernel and in the processing data packets process, called which function; And the time that each function consumed; If to the familiar words of the processing data packets process of kernel; By checking timestamp, can also analyze the queuing delay of packet in processing procedure into formation and dequeue.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110309258.6A CN102346710B (en) | 2011-10-13 | 2011-10-13 | Dynamic stubbing technology based time-delay analysis method for data packet processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110309258.6A CN102346710B (en) | 2011-10-13 | 2011-10-13 | Dynamic stubbing technology based time-delay analysis method for data packet processing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102346710A true CN102346710A (en) | 2012-02-08 |
CN102346710B CN102346710B (en) | 2015-04-29 |
Family
ID=45545404
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201110309258.6A Expired - Fee Related CN102346710B (en) | 2011-10-13 | 2011-10-13 | Dynamic stubbing technology based time-delay analysis method for data packet processing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102346710B (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103049378A (en) * | 2012-12-18 | 2013-04-17 | 惠州市亿能电子有限公司 | Program debugging method without emulator |
WO2013144731A1 (en) * | 2012-03-28 | 2013-10-03 | International Business Machines Corporation | Optimizing subroutine calls based on architecture level of called subroutine |
CN103365775A (en) * | 2013-06-21 | 2013-10-23 | 北京大学 | Unit testing method based on internal state checking |
CN104504333A (en) * | 2014-11-25 | 2015-04-08 | 武汉安天信息技术有限责任公司 | Malicious code detection method and device of ELF (executable and linkable format) file |
CN104680042A (en) * | 2015-03-10 | 2015-06-03 | 北京深思数盾科技有限公司 | Method and system for analyzing performances of virtual machine |
CN105577480A (en) * | 2016-02-01 | 2016-05-11 | 新浪网技术(中国)有限公司 | Monitoring method and device of network connection performances |
CN106155897A (en) * | 2015-04-16 | 2016-11-23 | 腾讯科技(深圳)有限公司 | A kind of method for processing business and device |
CN106649063A (en) * | 2016-11-22 | 2017-05-10 | 腾讯科技(深圳)有限公司 | Method and system used for monitoring time consuming data when program runs |
CN108664372A (en) * | 2018-05-08 | 2018-10-16 | 平安科技(深圳)有限公司 | Monitoring device, method and the computer readable storage medium of test process |
CN109101416A (en) * | 2014-09-28 | 2018-12-28 | 华为技术有限公司 | A kind of kernel fault filling method and electronic equipment |
CN109558325A (en) * | 2018-12-03 | 2019-04-02 | 浪潮电子信息产业股份有限公司 | A kind of delay statistics method, apparatus and computer readable storage medium |
CN110109673A (en) * | 2019-04-25 | 2019-08-09 | 武汉大学 | A kind of compiler detection device and method based on Beidou subnanosecond grade high-precision time service |
CN112363900A (en) * | 2020-11-16 | 2021-02-12 | 中国科学院计算技术研究所 | Host inner container network performance analysis method and system based on pile insertion |
CN112416794A (en) * | 2020-12-03 | 2021-02-26 | 平安银行股份有限公司 | Method, device and equipment for processing code coverage rate and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004295162A (en) * | 2003-02-04 | 2004-10-21 | I-Net Device Co Ltd | Remote monitoring system |
CN101465863A (en) * | 2009-01-14 | 2009-06-24 | 北京航空航天大学 | Method for implementing high-efficiency network I/O in kernel virtual machine circumstance |
WO2010141010A1 (en) * | 2009-06-01 | 2010-12-09 | Hewlett-Packard Development Company, L.P. | System and method for collecting application performance data |
-
2011
- 2011-10-13 CN CN201110309258.6A patent/CN102346710B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004295162A (en) * | 2003-02-04 | 2004-10-21 | I-Net Device Co Ltd | Remote monitoring system |
CN101465863A (en) * | 2009-01-14 | 2009-06-24 | 北京航空航天大学 | Method for implementing high-efficiency network I/O in kernel virtual machine circumstance |
WO2010141010A1 (en) * | 2009-06-01 | 2010-12-09 | Hewlett-Packard Development Company, L.P. | System and method for collecting application performance data |
Non-Patent Citations (1)
Title |
---|
陶捷: "动态跟踪系统的性能模型研究及基于动态跟踪技术的机群监测软件的设计与实现", 《中国优秀博士学位论文全文数据库》 * |
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013144731A1 (en) * | 2012-03-28 | 2013-10-03 | International Business Machines Corporation | Optimizing subroutine calls based on architecture level of called subroutine |
GB2513813A (en) * | 2012-03-28 | 2014-11-05 | Ibm | Optimizing Subroutine Calls Based On Architecture Level Of Called Subroutine |
US9063759B2 (en) | 2012-03-28 | 2015-06-23 | International Business Machines Corporation | Optimizing subroutine calls based on architecture level of called subroutine |
US9075636B2 (en) | 2012-03-28 | 2015-07-07 | International Business Machines Corporation | Optimizing subroutine calls based on architecture level of called subroutine |
GB2513813B (en) * | 2012-03-28 | 2017-02-22 | Ibm | Optimizing Subroutine Calls Based On Architecture Level Of Called Subroutine |
CN103049378A (en) * | 2012-12-18 | 2013-04-17 | 惠州市亿能电子有限公司 | Program debugging method without emulator |
CN103365775A (en) * | 2013-06-21 | 2013-10-23 | 北京大学 | Unit testing method based on internal state checking |
CN103365775B (en) * | 2013-06-21 | 2016-04-06 | 北京大学 | Based on the unit test method that internal state checks |
CN109101416A (en) * | 2014-09-28 | 2018-12-28 | 华为技术有限公司 | A kind of kernel fault filling method and electronic equipment |
CN109101416B (en) * | 2014-09-28 | 2022-01-14 | 华为技术有限公司 | Kernel fault injection method and electronic equipment |
CN104504333A (en) * | 2014-11-25 | 2015-04-08 | 武汉安天信息技术有限责任公司 | Malicious code detection method and device of ELF (executable and linkable format) file |
CN104680042A (en) * | 2015-03-10 | 2015-06-03 | 北京深思数盾科技有限公司 | Method and system for analyzing performances of virtual machine |
CN106155897A (en) * | 2015-04-16 | 2016-11-23 | 腾讯科技(深圳)有限公司 | A kind of method for processing business and device |
CN106155897B (en) * | 2015-04-16 | 2020-12-08 | 腾讯科技(深圳)有限公司 | Service processing method and device |
CN105577480B (en) * | 2016-02-01 | 2019-04-12 | 新浪网技术(中国)有限公司 | A kind of monitoring method and device being connected to the network performance |
CN105577480A (en) * | 2016-02-01 | 2016-05-11 | 新浪网技术(中国)有限公司 | Monitoring method and device of network connection performances |
CN106649063A (en) * | 2016-11-22 | 2017-05-10 | 腾讯科技(深圳)有限公司 | Method and system used for monitoring time consuming data when program runs |
CN108664372A (en) * | 2018-05-08 | 2018-10-16 | 平安科技(深圳)有限公司 | Monitoring device, method and the computer readable storage medium of test process |
CN109558325A (en) * | 2018-12-03 | 2019-04-02 | 浪潮电子信息产业股份有限公司 | A kind of delay statistics method, apparatus and computer readable storage medium |
CN110109673A (en) * | 2019-04-25 | 2019-08-09 | 武汉大学 | A kind of compiler detection device and method based on Beidou subnanosecond grade high-precision time service |
CN112363900A (en) * | 2020-11-16 | 2021-02-12 | 中国科学院计算技术研究所 | Host inner container network performance analysis method and system based on pile insertion |
CN112416794A (en) * | 2020-12-03 | 2021-02-26 | 平安银行股份有限公司 | Method, device and equipment for processing code coverage rate and storage medium |
CN112416794B (en) * | 2020-12-03 | 2023-11-21 | 平安银行股份有限公司 | Code coverage rate processing method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN102346710B (en) | 2015-04-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102346710A (en) | Dynamic stubbing technology based time-delay analysis method for data packet processing | |
CN108776934B (en) | Distributed data calculation method and device, computer equipment and readable storage medium | |
CN101465863B (en) | Method for implementing high-efficiency network I/O in kernel virtual machine circumstance | |
CN100477643C (en) | Method for realizing data packet catching based on sharing internal memory | |
CN105577567A (en) | Network data packet parallel processing method based on Intel DPDK | |
CN100451989C (en) | Software testing system and testing method | |
García-Dorado et al. | High-performance network traffic processing systems using commodity hardware | |
CN101630285A (en) | Software performance testing method applied in embedded system | |
CN107046508A (en) | Message method of reseptance and the network equipment | |
CN104506379A (en) | Method and system for capturing network data | |
CN104102543A (en) | Load regulation method and load regulation device in cloud computing environment | |
CN105376270A (en) | Method, device and system for improving uploading efficiency | |
CN101901167A (en) | Multi-network security auditing method and system under multi-CPU architecture | |
CN111431757A (en) | Virtual network flow acquisition method and device | |
CN102831046A (en) | Method and device for collecting software performance in communication system | |
CN103856337A (en) | Resource occupation rate acquiring method, providing method, system and server thereof | |
CN110290190A (en) | A kind of method, monitored device and monitoring server acquiring data | |
CN111970151A (en) | Flow fault positioning method and system for virtual and container network | |
CN110489242A (en) | Distributed data calculation method, device, terminal device and storage medium | |
CN110442819A (en) | Data processing method, device, storage medium and terminal | |
JP5642725B2 (en) | Performance analysis apparatus, performance analysis method, and performance analysis program | |
CN101707622A (en) | Method for realizing remote data monitoring | |
CN106528411A (en) | Method and device for coverage rate detection and equipment | |
CN101425034A (en) | Log recording method and log system | |
CN109213566B (en) | Virtual machine migration method, device and equipment |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150429 Termination date: 20181013 |