CN112579405A - Method and system for selecting performance evaluation index of block chain benchmark test program - Google Patents
Method and system for selecting performance evaluation index of block chain benchmark test program Download PDFInfo
- Publication number
- CN112579405A CN112579405A CN201910933759.8A CN201910933759A CN112579405A CN 112579405 A CN112579405 A CN 112579405A CN 201910933759 A CN201910933759 A CN 201910933759A CN 112579405 A CN112579405 A CN 112579405A
- Authority
- CN
- China
- Prior art keywords
- micro
- events
- architecture
- importance
- performance
- 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
Classifications
-
- 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
-
- 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
- G06F11/302—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a software system
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for selecting a performance evaluation index of a block chain benchmark test program. The method comprises the following steps: ordering microarchitectural events based on importance to performance of the blockchain system to form a first set of microarchitectural events, the microarchitectural events reflecting performance of interaction with the microprocessor architecture; and searching a fuzzy set from the first micro-architecture event set as an evaluation index of the blockchain benchmark test program, wherein the importance of the members of the fuzzy set on the performance of the blockchain system meets a preset target, and the fuzzy set forms a second micro-architecture event set. The invention can effectively select the results after the importance of the block chain micro-system structure events is sequenced, and accurately carry out similarity analysis on the block chain reference test program by utilizing a small number of selected events.
Description
Technical Field
The invention relates to the technical field of block chains, in particular to a method and a system for selecting a block chain benchmark test program performance evaluation index.
Background
In recent years, blockchain technology has spread worldwide, and more companies and users have started using blockchain applications. The block chain is used as a subversive technology, is leading a new round of technology change and industry change in the world, is expected to become a 'source of strategy' for global technology innovation and mode innovation, and promotes 'information internet' to be shifted to 'value internet'. Blockchains as a leading edge technology are subject to constant innovation, experimentation and application.
However, current blockchain systems suffer from various performance problems. To measure the performance of a blockchain, the blockchain system needs to be characterized from the microarchitectural level. However, the CPUs in common use today have over 200 microarchitectural events, each of which is understood to be very time consuming and unnecessary. It is therefore important how to select the most representative micro-architectural events. The existing work load characterization methods are all intuitive to select micro-architecture events to characterize programs.
The workload characterization methods in the prior art are all intuitive methods for selecting microarchitectural events to characterize a program, and do not select more representative events by an accurate measurement method. This results in intuitively chosen events that are not important to the program and thus do not accurately and comprehensively characterize the program. Meanwhile, some current benchmark test program kits are also measured by intuitively selected events to avoid the over-high similarity, which can lead to the condition that the result is not accurate, and actually have the benchmark test programs with the over-high similarity, thus increasing the test expense.
Therefore, there is a need to improve the prior art and provide a more accurate method for selecting the performance evaluation index of the blockchain benchmark test program.
Disclosure of Invention
The present invention is directed to overcome the above-mentioned drawbacks of the prior art, and provides a method and a system for selecting a performance evaluation index of a block chain benchmark test program, in which a micro-architecture event is selected based on a fuzzy mathematical method, and then a similarity analysis is performed on the block chain benchmark test program according to the selection result.
According to a first aspect of the present invention, a method for selecting a performance evaluation index of a block chain benchmark test program is provided. The method comprises the following steps:
ordering microarchitectural events based on importance to performance of the blockchain system to form a first set of microarchitectural events, the microarchitectural events reflecting performance of interaction with the microprocessor architecture;
and searching a fuzzy set from the first micro-architecture event set as an evaluation index of the blockchain benchmark test program, wherein the importance of the members of the fuzzy set on the performance of the blockchain system meets a preset target, and the fuzzy set forms a second micro-architecture event set.
In one embodiment, finding a fuzzy set from the first set of micro-architectural events as an evaluation index for a blockchain benchmark comprises:
performing multiple rounds of training on the investigated block chain reference test program, and determining the importance score of each micro-architecture event in each round of training;
and mapping the importance scores of the micro-architectural events to an interval [0,1], setting a score threshold, and forming the micro-architectural events larger than the score threshold into the second micro-architectural event set.
In one embodiment, the importance score for the micro-architectural event in each round of training is expressed as:
wherein:
wherein the function errorWeight (err)i) Wherein a, b, c are constants, erriIs the error rate of the ith round of training, len (err) represents the number of training rounds; function eventWeight (im)i,j) The method (a) in (1),b, c, d, k being constants, imi,jIs the importance of the jth micro-architectural event in the ith round of training.
In one embodiment, mapping the importance score of a microarchitectural event to the interval [0,1] is represented as:
wherein scorejIs the importance score, of the jth micro-architectural eventminAnd scoremaxThe minimum and maximum values of the j-th micro-architectural event score, respectively.
In one embodiment, the method further comprises performing similarity analysis on the plurality of blockchain benchmark test programs according to the second micro-architecture event set, and screening out the blockchain benchmark test programs with similarity higher than a preset target.
In one embodiment, performing a similarity analysis on a plurality of blockchain benchmark test programs according to the second set of micro-architectural events comprises:
normalizing the original data in the second micro-architecture event set;
the similarity of two blockchain benchmark test programs is measured by using the Euclidean distance, and is expressed as follows:
wherein the content of the first and second substances,
and
respectively, benchmark test program bpAnd bqIs the normalized ith micro-architectural event value of (a), n is the number of micro-architectural events.
In one embodiment, ordering micro-architectural events based on importance to blockchain system performance includes:
constructing a correlation model between the micro-architecture event and the performance of the block chain system, and expressing the correlation model as follows:
IPC＝pred(e1,e2,…en)
wherein the input e of the correlation modeliIs the value of the ith microarchitectural event, n is the total number of microarchitectural events input, and the output IPC of the correlation model is an index reflecting the performance of the blockchain system;
and obtaining a sequencing result of the input micro-architecture events of the association model on the performance importance of the blockchain system by using a machine learning algorithm.
In one embodiment, the set score threshold is 0.6.
According to a second aspect of the present invention, a system for selecting a performance evaluation index of a block chain benchmark test program is provided. The system comprises:
a sorting unit: the system comprises a first micro-architecture event set, a second micro-architecture event set and a third micro-architecture event set, wherein the micro-architecture events are used for sequencing micro-architecture events based on importance of performance of a blockchain system to form a first micro-architecture event set, and the micro-architecture events are used for reflecting performance of interaction with a microprocessor architecture;
screening unit: and the evaluation index is used for searching a fuzzy set from the first micro-architectural event set as an evaluation index of the blockchain benchmark test program, wherein the importance of the members of the fuzzy set on the performance of the blockchain system meets a preset target, and the fuzzy set forms a second micro-architectural event set.
Compared with the prior art, the invention has the advantages that: and selecting the results after sorting the importance of the block chain micro-architecture events by using fuzzy mathematics to select a proper number of events more important to the benchmark test program. Furthermore, when the micro-architecture level of the block chain is characterized, only a small number of events need to be observed, so that the characteristics of the block chain system can be comprehensively and accurately measured, and the performance can be optimized. Meanwhile, the selected event can also carry out similarity analysis on the existing block chain benchmark test program, and the block chain benchmark test program with high similarity is screened out, so that the test overhead is reduced.
Drawings
The invention is illustrated and described only by way of example and not by way of limitation in the scope of the accompanying drawings, in which:
FIG. 1 is a flow diagram of a method of selecting a block chain benchmark performance evaluation indicator according to one embodiment of the present invention;
FIG. 2 is a diagram illustrating the effects of a similarity matrix according to one embodiment of the present invention;
FIG. 3 is an effect diagram of a micro-architectural event chosen using fuzzy mathematics, according to one embodiment of the present invention;
FIG. 4 is a diagram illustrating the effect of similarity analysis of different blockchain benchmark programs using selected micro-architectural evaluation metrics, according to one embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions, design methods, and advantages of the present invention more apparent, the present invention will be further described in detail by specific embodiments with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not as a limitation. Thus, other examples of the exemplary embodiments may have different values.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
According to an embodiment of the invention, a method for selecting a performance evaluation index of a block chain benchmark test program is provided. The method is divided into two parts on the whole, one is to select a proper number of micro-architecture events which are more important to the performance of a block chain system by using a fuzzy mathematics method; and secondly, screening the similarity of the existing block chain system benchmark test programs, and selecting a more representative benchmark test program.
Specifically, referring to fig. 1, the method of the embodiment of the present invention includes the following steps:
step S110, according to the correlation degree of the micro-architecture event and the performance of the block chain system, the importance ranking of the micro-architecture event is obtained.
As used herein, microarchitectural events refer to performance indicators that reflect interactions with the microarchitecture (e.g., CPU). For example, micro-architectural events include, but are not limited to: BMMR, number of retirements of mispredicted macrobranch instructions; BIOT, calculating the retired number of unexecuted branch instructions; CA1P, number of cycles to suspend L1 data caching; CAS1, the number of execution stalls due to L1 data cache misses, etc. These microarchitectural events are related to the performance of the blockchain system, but with different degrees of correlation, the higher the degree of correlation, the more important it is.
In one embodiment, the results of the ranking of importance of microarchitectural events may be determined based on historical experience.
In a preferred embodiment, a correlation model between the micro-architecture events and the performance of the blockchain system is constructed, and the importance ranking result of the micro-architecture events is obtained by using a machine learning method.
For example, first, the association model is generally expressed as:
IPC＝pred(e1,e2,…en)
wherein the input e of the correlation modeliIs the value of the ith micro-architectural event, n is the total number of micro-architectural events input, and the output IPC of the correlation model is an index reflecting the performance of the blockchain system, e.g., IPC represents the number of instructions per cycle for measuring the node-level performance of the blockchain system. It should be understood that the values of these micro-architectural events (including the number of instructions per cycle) may be collected by running a benchmark program, and are not described herein.
Next, a machine learning algorithm is used to obtain a ranking result of the input micro-architecture events of the association model on the importance of the performance of the blockchain system. For example, the constructed association model is trained using a random gradient enhanced regression tree (SGBRT) in an ensemble learning algorithm. The key point is that SGBRT combines many tree models, each reflecting a portion of performance, in a phased manner, with the final model being called a collective model. By building an association model (or performance model), the importance of micro-architectural events and their interactions can be quantified using the model. Clearly, a more accurate performance model may result in a better quantification of event importance. To make the results more intuitive, the importance of the events can be normalized so that the sum of the importance of all events is 100%, and a higher percentage of the importance of each event indicates a greater impact of the event on the performance of the blockchain system.
Step S120, screening out the micro system structure events which are more important to the block chain system based on a fuzzy mathematical method.
In this step, a fuzzy mathematical approach is used to pick the appropriate number of more important blockchain system micro-architectural events.
In set theory, when a is a set, its membership function can take only two values: 1 or 0, respectively, indicates whether an element belongs to a. However, it is different for the fuzzy set A, which is characterized by a membership function μAEach element in the (u) domain maps to a real number interval [0,1]]I.e. muA:u→[0,1]。μA(u) a fuzzy set called "membership grade" u, which represents the degree of membership. If the value is close to 1, the u belongs to A and the degree is higher; conversely, if the value is close to 0, it indicates that u belongs to a lower degree a.
In an embodiment of the present invention, an appropriate number of more important micro-architectural events are screened out by training, with the goal of finding a fuzzy set A that represents a series of important micro-architectural events that are relevant to the performance of the blockchain system. And the membership degree of each event of A is obtained by a membership function.
For example, a Toronto training is performed, each round of training is performed for all of the benchmark test programs under investigation, each round has a model training error rate, and each round of training determines importance (e.g., in percentage) for each microarchitectural event.
Specifically, a training error rate weight function is first constructed for each round of training, and the training error rate is taken as an input, where the weight function errorwight is defined as:
wherein a, b, c are constants, erriIs the error rate of the ith round of training, len (err) indicates the number of rounds of training.
In equation (1), the exponential factor is used because the difference in error rates between different training processes is small, and the factor needs to be used to amplify the difference between them. The values of a, b and c can be determined empirically or experimentally, for example, a is 100, b is 180 and c is 45. Alternatively, other forms of weighting function errorwight may be used to measure the impact of the training error rate, e.g., not including exponential factors or not necessarily including all of the constants a, b, c, etc.
In addition, another weighting function is set to represent the importance of the micro-architecture event of each round of training (or called importance weighting function eventWeight), the input of the weighting function is the importance of each micro-architecture event, and the weighting function eventWeight is defined as:
where a, b, c, d, k are constants imi,jIs the importance of the j-th micro-architectural event (e.g., SGBRT) in the i-th round of training.
Equation (2) is an incremental function, i.e. more important events have higher weights, and a, b, c, d, k can be determined empirically or experimentally, e.g. a is 0, b is 0.2, c is 10000, d is 6, k is 0.55. The significance of the more accurate micro-architectural events is gained by performing a differential amplification with an exponential factor in equation (2). Alternatively, other forms of weighting function eventWeight may be used to measure the importance of microarchitectural events, e.g., not including the exponential factor k or not necessarily including all of the constants a, b, c, d, k, etc.
Next, an importance score for each microarchitectural event is calculated from the training error rate weight function and the importance weight function, and the result is normalized, expressed as:
wherein, scorejRepresents the score of microarchitectural event j (synthesizing all rounds of training),
is the normalized score, n is the number of training rounds, scoreminAnd scoremaxThe minimum and maximum score values, respectively.
Finally, the micro-architecture event scores of all the blockchain benchmark test programs are mapped to the interval [0,1], a threshold value is set, all the events larger than the threshold value are taken as the more important micro-architecture events which are selected, and the threshold value is set to be 0.6, 0.7 and the like.
In summary, in step S110, the importance ranking result is obtained first, and in step S120, it is determined how many microarchitectural events that the evaluation index is appropriate are selected based on a fuzzy mathematical method, and the selected microarchitectural events can be subsequently used for evaluating the blockchain benchmark test program.
Step S130, analyzing the similarity of the benchmark test programs of the block chain system according to the screening result, and selecting a more representative benchmark test program.
In this step, the similarity of the existing benchmark programs of the block chain system is screened, and more representative benchmark programs are selected. Based on the above-mentioned selection of a suitable number of micro-architectural events, a similarity analysis can be performed on the existing blockchain benchmark test program.
Specifically, first, prior to similarity analysis, the raw data in the important events (i.e., the raw values of the obtained micro-architectural events) are normalized, because the data values of the collected micro-architectural events are typically large and of different magnitudes. The normalization formula is expressed as:
wherein E isiAverage of the ith microarchitectural event for the benchmark test program, Ei,min、Ei,maxThe minimum and maximum values of the event in the selected benchmark test program, respectively
Is a normalized value.
Next, a benchmark similarity matrix is defined that represents the similarity of the selected benchmarks and that can be visualized. For example, the similarity matrix is based on the Euclidean distance between two significant vectors representing two benchmarks, i.e., the distance between two points in the n-dimensional space is measured by the Euclidean distance. Euclidean distance is expressed as:
wherein the content of the first and second substances,
and
respectively, benchmark test program bpAnd bqThe normalized ith micro-architectural event value of (c) can be obtained from equation (5) above, with n being the number of micro-architectural events.
FIG. 2 is a diagram illustrating the effect of benchmark similarity using a similarity matrix, wherein B1, B2, etc. represent different blockchain benchmarks, and each block corresponds to the direct similarity of two benchmarks. Darker squares indicate higher similarity between the two benchmarks. The similarity matrix can be used to observe the similarity of euclidean distances for all selected benchmarks.
Further, the feasibility of the invention was verified experimentally. First, for all selected blockchain systems, the results of the microarchitectural event importance ranking are selected by fuzzy mathematics, and the score threshold is set to 0.6. Fig. 3 shows all microarchitectural events that meet the conditions (the importance score is greater than 0.6), wherein the ordinate importance is the importance score of fuzzy mathematics, the abscissa events is the abbreviation of event name, and 17 microarchitectural events are selected as microarchitectural evaluation indexes of the block chain system benchmark test program.
Then, the micro-architecture evaluation index of the selected blockchain system benchmark test program is used to perform similarity analysis on different blockchain benchmark test programs, and a similarity matrix is drawn, as shown in fig. 4, wherein the horizontal and vertical coordinates represent different blockchain benchmark test programs, each block corresponds to direct similarity of the two benchmark test programs, and the deeper the block color indicates that the similarity between the two benchmark test programs is higher. Here, setting the normalized euclidean distance between the two benchmarks to less than 0.2 defines the similarity to be high.
Experimental results show that 17 more important events for a benchmark test program are selected by selecting results after importance sorting of block chain microarchitectural events by using a fuzzy mathematic method. Only these few events need to be observed to fully and accurately measure the characteristics of the blockchain system and optimize the performance. Meanwhile, similarity analysis is carried out on the existing block chain benchmark test programs by using the selected events, 9 of the 17 benchmark test programs are found to be redundant, and the test overhead can be reduced by screening the block chain benchmark test programs with high similarity.
The embodiment of the invention also provides a system for selecting the performance evaluation index of the block chain benchmark test program. The system can implement one or more aspects of the above method, for example, the system comprises: the ordering unit is used for ordering the micro-architecture events based on the importance of the performance of the block chain system to form a first micro-architecture event set, and the micro-architecture events are used for reflecting the performance of interaction with the microprocessor architecture; and the screening unit is used for searching a fuzzy set from the first micro-architecture event set as an evaluation index of the blockchain benchmark test program, wherein the importance of the members of the fuzzy set on the performance of the blockchain system meets a preset target, and the fuzzy set forms a second micro-architecture event set.
It should be understood that the fuzzy mathematical formula of the embodiments of the present invention may vary according to different application scenarios. In addition, the method of the embodiment of the present invention may be used not only in a blockchain system, but also in other systems, such as a big data system, and may also select the micro-architecture evaluation index of the benchmark test program set of different systems, and perform similarity analysis on the benchmark test programs. In summary, compared with the prior art, the invention has the following advantages: firstly, the micro-architecture evaluation indexes selected by the benchmark test program are intuitively selected by researchers through experience, which causes that the selected events are not relatively important to the program and cannot accurately and comprehensively characterize the system at the micro-architecture level. On the other hand, the existing block chain benchmark test programs may have the problem of high similarity, and the use of the benchmark test programs with high similarity increases the test and analysis overhead. The invention selects the events of the block chain system micro-architecture layer by using a fuzzy mathematics method, selects a proper number of more important micro-architecture events as the micro-architecture indexes of the block chain reference test program, can accurately and comprehensively characterize the block chain system on the micro-architecture layer only by analyzing the events, and reduces the analysis cost. Meanwhile, the indexes are used for carrying out similarity analysis on the block chain benchmark test programs so as to screen out more representative block chain benchmark test programs and reduce the test expense. The invention can solve the problem of how to select the more important micro-architecture events of the blockchain system with proper quantity and the problem of over-high similarity of the existing benchmark test programs of the blockchain system.
It should be noted that, although the steps are described in a specific order, it is not meant that the steps must be executed in the specific order, and in fact, some of the steps may be executed concurrently or even in a different order as long as the required functions are achieved.
The present invention may be a system, method and/or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied therewith for causing a processor to implement various aspects of the present invention.
The computer readable storage medium may be a tangible device that retains and stores instructions for use by an instruction execution device. The computer readable storage medium may include, for example, but is not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as punch cards or in-groove projection structures having instructions stored thereon, and any suitable combination of the foregoing.
Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (10)
1. A method for selecting performance evaluation indexes of a block chain benchmark test program comprises the following steps:
ordering microarchitectural events based on importance to performance of the blockchain system to form a first set of microarchitectural events, the microarchitectural events reflecting performance of interaction with the microprocessor architecture;
and searching a fuzzy set from the first micro-architecture event set as an evaluation index of the blockchain benchmark test program, wherein the importance of the members of the fuzzy set on the performance of the blockchain system meets a preset target, and the fuzzy set forms a second micro-architecture event set.
2. The method of claim 1, wherein finding a fuzzy set from the first set of micro-architectural events as an evaluation indicator of a blockchain benchmark test procedure comprises:
performing multiple rounds of training on the investigated block chain reference test program, and determining the importance score of each micro-architecture event in each round of training;
and mapping the importance scores of the micro-architectural events to an interval [0,1], setting a score threshold value, and forming the micro-architectural events larger than the score threshold value into the second micro-architectural event set.
3. The method of claim 2, wherein the importance score of the micro-architectural event in each training round is expressed as:
wherein:
wherein the function: errorWeight (err)i) Wherein a, b, c are constants, erriIs the error rate of the ith round of training, len (err) represents the number of training rounds; function eventWeight (im)i，j) Wherein a, b, c, d, k are constants imi，jIs the importance of the jth micro-architectural event in the ith round of training.
4. The method of claim 3, wherein mapping the importance score of a microarchitectural event to the interval [0,1] is represented as:
wherein scorejIs the importance score, of the jth micro-architectural eventminAnd scoremaxThe minimum and maximum values of the j-th micro-architectural event score, respectively.
5. The method of claim 1, further comprising performing a similarity analysis on the plurality of blockchain benchmark test programs according to the second set of micro-architectural events, and screening out blockchain benchmark test programs having a similarity higher than a predetermined target.
6. The method of claim 5, wherein performing similarity analysis on a plurality of blockchain benchmark test programs according to the second set of micro-architectural events comprises:
normalizing the original data in the second micro-architecture event set;
the similarity of two blockchain benchmark test programs is measured by using the Euclidean distance, and is expressed as follows:
wherein the content of the first and second substances,
and
respectively, benchmark test program bpAnd bqIs the normalized ith micro-architectural event value of (a), n is the number of micro-architectural events.
7. The method of claim 1, wherein ranking micro-architectural events based on importance to blockchain system performance comprises:
constructing a correlation model between the micro-architecture event and the performance of the block chain system, and expressing the correlation model as follows:
IPC＝pred(e1，e2，...en)
wherein the input e of the correlation modeliIs the value of the ith micro-architecture event, n is the total number of the micro-architecture events input, and the output IPC of the association model is an index reflecting the performance of the block chain system;
and obtaining a sequencing result of the input micro-architecture events of the association model on the importance of the performance of the blockchain system by utilizing a machine learning algorithm.
8. The method of claim 2, wherein the set score threshold is 0.6.
9. A system for selecting a performance evaluation index of a block chain benchmark test program comprises:
a sorting unit: the system comprises a first micro-architecture event set and a second micro-architecture event set, wherein the micro-architecture events are used for sequencing micro-architecture events based on the importance of the performance of a blockchain system to form the first micro-architecture event set, and the micro-architecture events are used for reflecting the performance of interaction with a microprocessor architecture;
screening unit: and the evaluation index is used for searching a fuzzy set from the first micro-architecture event set as an evaluation index of the blockchain benchmark test program, wherein the importance of the members of the fuzzy set on the performance of the blockchain system meets a preset target, and the fuzzy set forms a second micro-architecture event set.
10. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910933759.8A CN112579405B (en) | 2019-09-29 | 2019-09-29 | Method and system for selecting performance evaluation index of block chain benchmark test program |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910933759.8A CN112579405B (en) | 2019-09-29 | 2019-09-29 | Method and system for selecting performance evaluation index of block chain benchmark test program |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112579405A true CN112579405A (en) | 2021-03-30 |
| CN112579405B CN112579405B (en) | 2022-07-15 |
Family
ID=75110628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910933759.8A Active CN112579405B (en) | 2019-09-29 | 2019-09-29 | Method and system for selecting performance evaluation index of block chain benchmark test program |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112579405B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090276190A1 (en) * | 2008-04-30 | 2009-11-05 | International Business Machines Corporation | Method and Apparatus For Evaluating Integrated Circuit Design Performance Using Basic Block Vectors, Cycles Per Instruction (CPI) Information and Microarchitecture Dependent Information |
| US20170132635A1 (en) * | 2015-11-11 | 2017-05-11 | Idm Global, Inc. | Systems and methods for sanction screening |
| US20180181898A1 (en) * | 2016-12-22 | 2018-06-28 | Atlassian Pty Ltd | Method and apparatus for a benchmarking service |
| CN110061889A (en) * | 2019-04-01 | 2019-07-26 | 北京众享比特科技有限公司 | Block chain performance test methods, device, equipment and storage medium |
-
2019
- 2019-09-29 CN CN201910933759.8A patent/CN112579405B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090276190A1 (en) * | 2008-04-30 | 2009-11-05 | International Business Machines Corporation | Method and Apparatus For Evaluating Integrated Circuit Design Performance Using Basic Block Vectors, Cycles Per Instruction (CPI) Information and Microarchitecture Dependent Information |
| US20170132635A1 (en) * | 2015-11-11 | 2017-05-11 | Idm Global, Inc. | Systems and methods for sanction screening |
| US20180181898A1 (en) * | 2016-12-22 | 2018-06-28 | Atlassian Pty Ltd | Method and apparatus for a benchmarking service |
| CN110061889A (en) * | 2019-04-01 | 2019-07-26 | 北京众享比特科技有限公司 | Block chain performance test methods, device, equipment and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| YIRONG LV等: "CounterMiner: Mining Big Performance Data from Hardware Counters", 《2018 51ST ANNUAL IEEE/ACM INTERNATIONAL SYMPOSIUM ON MICROARCHITECTURE (MICRO)》 * |
| 吕超等: "计算机体系结构基准测试程序集的研究", 《计算机应用与软件》 * |
| 戴意瑜等: "基于概率信息不完备的群决策模型", 《计算机应用》 * |
| 王洋等: "区块链技术与传统软件技术对比分析", 《信息技术与标准化》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112579405B (en) | 2022-07-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shepperd | Software project economics: a roadmap | |
| RU2672394C1 (en) | Methods and systems for evaluation of training objects through a machine training algorithm | |
| JP6636071B2 (en) | Computer mounting method, computer system and computer device | |
| Yadav et al. | A fuzzy logic based approach for phase-wise software defects prediction using software metrics | |
| US20200019577A1 (en) | Candidate answers for speculative questions in a deep question answering system | |
| WO2017118435A1 (en) | System, device and method for releasing vehicle insurance surveying task, and readable storage medium | |
| US20190251458A1 (en) | System and method for particle swarm optimization and quantile regression based rule mining for regression techniques | |
| Antunes et al. | On the metrics for benchmarking vulnerability detection tools | |
| US20110313817A1 (en) | Key performance indicator weighting | |
| US8170894B2 (en) | Method of identifying innovations possessing business disrupting properties | |
| EP1835426A1 (en) | Estimating software power consumption | |
| JP2007035034A (en) | Method and system for detecting in real-time search terms whose popularity increase rapidly | |
| Abotaleb et al. | Construction bidding markup estimation using a multistage decision theory approach | |
| JP2010507146A (en) | Method and apparatus for capacity planning and resource optimization of distributed systems | |
| US20120197674A1 (en) | Estimating a future project characteristic based on the similarity of past projects | |
| Lee et al. | Software measurement and software metrics in software quality | |
| CN109726811A (en) | Use priority formation neural network | |
| WO2019234716A1 (en) | Methods and systems for blockchain testing | |
| Yuan et al. | Calibration of stochastic computer models using stochastic approximation methods | |
| Chen et al. | Bayesian estimation of realized GARCH-type models with application to financial tail risk management | |
| Yan et al. | Functional principal components analysis on moving time windows of longitudinal data: dynamic prediction of times to event | |
| CN112579405B (en) | Method and system for selecting performance evaluation index of block chain benchmark test program | |
| Sokolowski | Monte Carlo Simulation | |
| US20080126029A1 (en) | Run-Time Characterization of On-Demand Analytical Model Accuracy | |
| Jovanović et al. | How to measure Triple Helix performance? A fresh approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |