CN106506282A - A kind of monitoring method for improving cloud platform monitoring performance and scale - Google Patents
A kind of monitoring method for improving cloud platform monitoring performance and scale Download PDFInfo
- Publication number
- CN106506282A CN106506282A CN201611083976.5A CN201611083976A CN106506282A CN 106506282 A CN106506282 A CN 106506282A CN 201611083976 A CN201611083976 A CN 201611083976A CN 106506282 A CN106506282 A CN 106506282A
- Authority
- CN
- China
- Prior art keywords
- monitoring
- virtual machine
- calculate node
- refers
- dislocation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
Abstract
The present invention relates to cloud computing administrative skill field, particularly relates to a kind of monitoring method for improving cloud platform monitoring performance and scale.The method of the present invention is within the monitoring frequency cycle, and the multiple batches of the monitoring data collection task of single batch point are carried out；Realize dislocation monitoring.The inventive method O＆M simple and flexible, without the need for using distributed deployment structure, making full use of the existing monitoring cycle that extensive monitoring collection is realized dislocation collection.
Description
Technical field
The present invention relates to cloud computing administrative skill field, particularly relates to a kind of prison for improving cloud platform monitoring performance and scale
Prosecutor method.
Background technology
With the development of cloud computing, the scale of single cloud platform gradually expanding, for the ease of management and O＆M, generally
Capital exploitation is monitored for every resource of virtual machine, the such as items such as cpu busy percentage, memory usage, I/O utilization
Resource uses monitored item.Each monitoring data can recorded database or file server to facilitate O＆M people by cloud platform
Member consults, and these history monitoring datas can generally allow operation maintenance personnel to find the abnormal conditions of virtual machine in time.But with
The increase of virtual machine quantity, history monitoring data present outburst when growth, particularly monitoring frequency arrange comparison high when
Wait, such as when cloud platform scale reaches 10,000, monitoring frequency 5s once, during every virtual machine monitoring index 5, one minute will
Produce 600,000 datas, one day just 864,000,000 data of collection, and the concurrent collecting quantity that database and server can bear
It is limited.
Typically solve the problems, such as concurrently to gather by the way of distributed deployment or raising hardware configuration at present, this
Mode often consumes substantial amounts of cost and O＆M complexity.
Content of the invention
Present invention solves the technical problem that being to provide a kind of monitoring method for improving cloud platform monitoring performance and scale, solution
The deficiency of certainly present monitoring method, provides one kind for virtual machine items monitor control index under cloud computing environment and is ensureing monitoring data not
Impacted, and in the case of not increasing cost, improve the solution of cloud platform monitoring performance and scale.
The present invention solves the technical scheme of above-mentioned technical problem,
The method of methods described be within the monitoring frequency cycle, will be multiple for the monitoring data collection task of single batch point
Batch is carried out；Realize dislocation monitoring.
Described method specifically includes following steps：
Step 1：Dislocation monitoring value is set, and the twice of the monitoring value that misplaces is less than monitoring frequency；
Step 2：According to dislocation monitoring value and monitoring frequency, the wanted number of times for gathering in batches in a monitoring frequency is calculated
n；
Step 3：The virtual machine of every calculate node is grouped according to batch, grouping strategy is calculate node virtual machine quantity
After divided by batch, carry out being grouped and serial number according to presetting；
Step 4：The same virtual machine combination of each calculate node numbering is chosen, dislocation monitored object set, numbering is formed
Constant；
Step 5：According to monitoring batch n for calculating, in same monitoring cycle, the virtual machine of each monitoring set is completed
Monitoring data collection.
Described method may also include step 6：When new virtual machine is created, by the virtual machine for newly creating according to warp knit
Good numbering and batch inverted order add monitoring set；
Described inverted order adds monitoring set, refers to and newly create a virtual machine, this virtual machine when calculate node N1
Should be in last packet of this calculate node.
Described monitoring frequency, refers to the time cycle of each clocked flip data acquisition；
Described times of collection n in batches, equal to monitoring frequency divided by dislocation monitoring value, is as a result rounded downwards；
Described rounds downwards, refers to and ignores decimal.
Described monitoring frequency, refers to the time cycle of each clocked flip data acquisition；
Described times of collection n in batches, equal to monitoring frequency divided by dislocation monitoring value, is as a result rounded downwards；
Described rounds downwards, refers to and ignores decimal.
Described calculate node, refers to virtualization node, provides the user virtual machine；
Described carry out being grouped and serial number according to presetting, refer to that virtual machine quantity rounds up less than batch,
Remainder is placed on last group more than batch by virtual machine, and when batch is 10, calculate node N1 has 20, and calculate node N2 has 6
Platform, then by calculate node N1 per one group of two virtual machines serial number, one group of virtual machine of each of calculate node N2 is simultaneously sequentially
Numbering.
Described dislocation monitored object set, refers to the virtual machine set per secondary gathered data, mainly preserves virtual machine
Example ID；
Described virtual machine instance ID, refers to the unique mark of virtual machine.
Scheme of the invention has the beneficial effect that：
1st, the method for the present invention saves user cost, in the case where hardware configuration or quantity is increased without, improves cloud
Platform monitoring performance and scale；
2nd, the inventive method O＆M simple and flexible, without the need for using distributed deployment structure, making full use of existing monitoring week
Extensive monitoring collection is realized dislocation collection by the phase, there is provided a kind of unaffected in guarantee monitoring data, and does not increase cost
In the case of improve the method for cloud platform monitoring performance and scale.
Description of the drawings
The present invention is further described below in conjunction with the accompanying drawings：
Fig. 1 is the flow chart of the present invention；
Fig. 2 is the technical schematic diagram of the present invention.
Specific embodiment
As shown in figure 1, specific implementation process of the present invention is as follows：
Step 1：Dislocation monitoring value is set, and the twice of error monitoring value should be less than monitoring frequency.
Step 2：According to dislocation monitoring value and monitoring frequency, the wanted number of times for gathering in batches in a monitoring frequency is calculated
n.
Step 3：The virtual machine of every calculate node is grouped according to batch, grouping strategy is calculate node virtual machine quantity
Divided by batch, virtual machine quantity rounds up less than batch, and remainder is placed on last group, such as more than batch by virtual machine
Batch is 10, and calculate node N1 has 20, and calculate node N2 has 6, then calculate node N1 is per one group of two virtual machines and suitable
Sequence numbering, one group of virtual machine of each of calculate node N2 serial number.
Step 4：The same virtual machine combination of each calculate node numbering is chosen, dislocation monitored object set, numbering is formed
Constant.
Step 5：According to monitoring batch n for calculating, in same monitoring cycle, the virtual machine of each monitoring set is completed
Monitoring data collection.
Step 6：When new virtual machine is created, by the virtual machine for newly creating according to the numbering and batch inverted order that has finished
Monitoring set, such as calculate node N1 is added newly to create a virtual machine, this virtual machine should be in this calculate node most
A packet afterwards.
Fig. 2 is the technical schematic diagram of the present invention, and monitoring management end refers to that cloud platform is specifically used to do monitoring management, main
It is responsible for the work of monitoring data related management.
Monitoring agent, refers to the monitor client installed in every calculate node, is responsible for the collecting work of monitoring data, and
It is responsible for reporting monitoring data to monitoring management end.
Control node installs monitoring management end, and each one minute monitoring management module can call each calculate node
Monitoring agent interface gathers every virtual machine indices performance data, it is assumed that if not misplacing monitoring, and monitoring frequency is 1 point
Clock, i.e. each one minute monitoring management module can call monitoring agent interface to gather every virtual machine indices performance number
According to if there is 10,000 virtual machines this when, 5 indexs of every virtual machine, that data for gathering every time are up to 50,000, that is, supervise
Keyholed back plate reason client database insertion number of concurrent per minute is up to 50,000, if dislocation monitoring value is arranged to 12 seconds, can be at 1 point
By original 50,000 inquiries point 5 times in the monitoring cycle of clock, 10,000 insertions every time are recorded；If dislocation monitoring value is arranged 6 seconds,
Original 50,000 inquiries can be divided 10 times within the monitoring cycle of 1 minute, then the insertion per minute of monitoring management client database is simultaneously
Send out and just record for 5,000.
Then the virtual machine of every calculate node is grouped according to batch, grouping strategy is removed for calculate node virtual machine quantity
With batch, virtual machine quantity rounds up less than batch, and remainder is placed on last group, is such as criticized more than batch by virtual machine
Secondary is 10, and calculate node N1 has 20, and calculate node N2 has 6, then by calculate node N1 per one group of two virtual machines and order
Numbering, one group of virtual machine of each of calculate node N2 serial number.
Then according to dislocation monitoring value and monitoring frequency, the wanted frequency n for gathering in batches in a monitoring frequency is calculated,
Times of collection n, equal to monitoring frequency divided by dislocation monitoring value, as a result rounds downwards, ignores decimal in batches.
Choose the same virtual machine combination of each calculate node numbering afterwards, form dislocation monitored object set, numbering is not
Become, misplace monitoring set, refers to the virtual machine set per secondary gathered data, main preservation virtual machine instance ID, virtual machine reality
Example ID, refers to the unique mark of virtual machine, finally according to monitoring batch n for calculating, in same monitoring cycle, completes each
The virtual machine monitoring data acquisition of monitoring set.
Claims (8)
1. a kind of monitoring method for improving cloud platform monitoring performance and scale, it is characterised in that the method for methods described is
In the individual monitoring frequency cycle, the multiple batches of the monitoring data collection task of single batch point are carried out；Realize dislocation monitoring.
2. a kind of monitoring method for improving cloud platform monitoring performance and scale according to claim 1, it is characterised in that institute
The method that states specifically includes following steps：
Step 1：Dislocation monitoring value is set, and the twice of the monitoring value that misplaces is less than monitoring frequency；
Step 2：According to dislocation monitoring value and monitoring frequency, the wanted frequency n for gathering in batches in a monitoring frequency is calculated；
Step 3：By the virtual machine of every calculate node according to batch be grouped, grouping strategy be calculate node virtual machine quantity divided by
After batch, carry out being grouped and serial number according to presetting；
Step 4：The same virtual machine combination of each calculate node numbering is chosen, dislocation monitored object set is formed, is numbered constant；
Step 5：According to monitoring batch n for calculating, in same monitoring cycle, the virtual machine monitoring of each monitoring set is completed
Data acquisition.
3. a kind of monitoring method for improving cloud platform monitoring performance and scale according to claim 2, it is characterised in that institute
The method that states may also include step 6：When new virtual machine is created, by the virtual machine for newly creating according to the numbering that has finished with
Batch inverted order adds monitoring set；
Described inverted order adds monitoring set, refers to and newly create a virtual machine when calculate node N1, and this virtual machine should
In last packet of this calculate node.
4. a kind of monitoring method for improving cloud platform monitoring performance and scale according to claim 2, it is characterised in that institute
The monitoring frequency that states, refers to the time cycle of each clocked flip data acquisition；
Described times of collection n in batches, equal to monitoring frequency divided by dislocation monitoring value, is as a result rounded downwards；
Described rounds downwards, refers to and ignores decimal.
5. a kind of monitoring method for improving cloud platform monitoring performance and scale according to claim 3, it is characterised in that institute
The monitoring frequency that states, refers to the time cycle of each clocked flip data acquisition；
Described times of collection n in batches, equal to monitoring frequency divided by dislocation monitoring value, is as a result rounded downwards；
Described rounds downwards, refers to and ignores decimal.
6. a kind of raising cloud platform monitoring performance according to any one of claim 2-5 and the monitoring method of scale, its are special
Levy and be, described calculate node refers to virtualization node, provides the user virtual machine；
Described carry out being grouped and serial number according to presetting, refer to that virtual machine quantity rounds up less than batch, virtual
Remainder is placed on last group more than batch by machine, and when batch is 10, calculate node N1 has 20, and calculate node N2 has 6,
Then by calculate node N1 per one group of two virtual machines serial number, one group of virtual machine of each of calculate node N2 is simultaneously sequentially compiled
Number.
7. a kind of raising cloud platform monitoring performance according to any one of claim 2-5 and the monitoring method of scale, its are special
Levy and be, described dislocation monitored object set refers to the virtual machine set per secondary gathered data, main preserve virtual machine reality
Example ID；
Described virtual machine instance ID, refers to the unique mark of virtual machine.
8. a kind of monitoring method for improving cloud platform monitoring performance and scale according to claim 6, it is characterised in that institute
The dislocation monitored object set that states, refers to the virtual machine set per secondary gathered data, main preservation virtual machine instance ID；
Described virtual machine instance ID, refers to the unique mark of virtual machine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611083976.5A CN106506282A (en) | 2016-11-30 | 2016-11-30 | A kind of monitoring method for improving cloud platform monitoring performance and scale |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611083976.5A CN106506282A (en) | 2016-11-30 | 2016-11-30 | A kind of monitoring method for improving cloud platform monitoring performance and scale |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106506282A true CN106506282A (en) | 2017-03-15 |
Family
ID=58327804
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611083976.5A Withdrawn CN106506282A (en) | 2016-11-30 | 2016-11-30 | A kind of monitoring method for improving cloud platform monitoring performance and scale |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106506282A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107483292A (en) * | 2017-09-11 | 2017-12-15 | 电子科技大学 | Dynamic monitoring and controlling method for cloud platform |
| CN109101321A (en) * | 2018-08-14 | 2018-12-28 | 郑州云海信息技术有限公司 | A kind of message monitoring method and device based on cloud platform |
| CN111506480A (en) * | 2020-04-23 | 2020-08-07 | 上海达梦数据库有限公司 | State detection method, device and system for components in cluster |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103037019A (en) * | 2013-01-07 | 2013-04-10 | 北京华胜天成科技股份有限公司 | Distributed data acquisition system and method based on cloud computing |
| CN103384206A (en) * | 2012-05-02 | 2013-11-06 | 中国科学院计算机网络信息中心 | Concurrent processing method and system for mass data |
| CN103780696A (en) * | 2014-01-23 | 2014-05-07 | 北京荣之联科技股份有限公司 | Cloud monitoring method, device and system based on distributed pushing |
| CN105119769A (en) * | 2015-07-01 | 2015-12-02 | 北京梅泰诺通信技术股份有限公司 | System for time hashing of periodic data report |
-
2016
- 2016-11-30 CN CN201611083976.5A patent/CN106506282A/en not_active Withdrawn
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103384206A (en) * | 2012-05-02 | 2013-11-06 | 中国科学院计算机网络信息中心 | Concurrent processing method and system for mass data |
| CN103037019A (en) * | 2013-01-07 | 2013-04-10 | 北京华胜天成科技股份有限公司 | Distributed data acquisition system and method based on cloud computing |
| CN103780696A (en) * | 2014-01-23 | 2014-05-07 | 北京荣之联科技股份有限公司 | Cloud monitoring method, device and system based on distributed pushing |
| CN105119769A (en) * | 2015-07-01 | 2015-12-02 | 北京梅泰诺通信技术股份有限公司 | System for time hashing of periodic data report |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107483292A (en) * | 2017-09-11 | 2017-12-15 | 电子科技大学 | Dynamic monitoring and controlling method for cloud platform |
| CN107483292B (en) * | 2017-09-11 | 2020-10-16 | 电子科技大学 | Dynamic monitoring method for cloud platform |
| CN109101321A (en) * | 2018-08-14 | 2018-12-28 | 郑州云海信息技术有限公司 | A kind of message monitoring method and device based on cloud platform |
| CN111506480A (en) * | 2020-04-23 | 2020-08-07 | 上海达梦数据库有限公司 | State detection method, device and system for components in cluster |
| CN111506480B (en) * | 2020-04-23 | 2024-03-08 | 上海达梦数据库有限公司 | Method, device and system for detecting states of components in cluster |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104065741A (en) | Data collection system and method | |
| CN110222029A (en) | A kind of big data multidimensional analysis computational efficiency method for improving and system | |
| CN103885881B (en) | A kind of high Real-time and Concurrent method of testing based on VxWorks system multiplex mechanism | |
| CN104239144A (en) | Multilevel distributed task processing system | |
| CN108810115B (en) | Load balancing method and device suitable for distributed database and server | |
| CN106506282A (en) | A kind of monitoring method for improving cloud platform monitoring performance and scale | |
| CN105527948B (en) | A kind of large-scale distributed data collecting system and method based on industrial process | |
| CN103679388A (en) | Production scheduling method and system | |
| CN104035786A (en) | Optimization method and system of software timers | |
| CN104407688A (en) | Virtualized cloud platform energy consumption measurement method and system based on tree regression | |
| CN108108248A (en) | A kind of CPU+GPU cluster management methods, device and equipment for realizing target detection | |
| CN103605578A (en) | Load balance scheduling method based on virtual machine migration | |
| CN110046048A (en) | A kind of load-balancing method adaptively quickly reassigned based on workload | |
| CN108243012A (en) | Charging application processing system, method and device in online charging system OCS | |
| CN106293947B (en) | GPU-CPU (graphics processing Unit-Central processing Unit) mixed resource allocation system and method in virtualized cloud environment | |
| CN103095598A (en) | Monitoring data aggregate method under large-scale cluster environment | |
| CN105488134A (en) | Big data processing method and big data processing device | |
| CN106227641B (en) | A kind of hardware performance monitoring method and system | |
| CN110414569A (en) | Cluster realizing method and device | |
| CN107193649A (en) | A kind of method for scheduling task and device based on NUMA system | |
| CN110032444A (en) | A kind of distributed system and distributed task scheduling processing method | |
| CN104536808B (en) | A kind of method for parameter configuration and system of cloud application program | |
| US20230376397A1 (en) | Method and System for Determining Interval Time for Testing of Server, and Device and Medium | |
| CN105242873B (en) | The acquisition of the performance data of cloud computing system and storage method and device | |
| CN103532786A (en) | Server synchronization detector as well as synchronous detection method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20170315 |
|
| WW01 | Invention patent application withdrawn after publication |