CN-114281869-B - Storage device capacity and performance acquisition method and system

CN114281869BCN 114281869 BCN114281869 BCN 114281869BCN-114281869-B

Abstract

The invention provides a storage device capacity and performance acquisition method and system, the method comprises the steps of utilizing a crawling program to crawl operation log information of the storage device, determining the capacity of the storage device according to the operation log information by the crawling program, analyzing and researching I/O performance evaluation on the storage device through a preset storage system I/O performance analysis model according to request service information in the operation log, and accordingly utilizing the crawling program to rapidly acquire operation data of the storage device, namely determining the capacity and the I/O performance evaluation of the storage device.

Inventors

WANG XINGKAI

Assignees

中国建设银行股份有限公司
中国建设银行股份有限公司

Dates

Publication Date: 20260421
Application Date: 20220104
Priority Date: 20220104

Claims (8)

1. A method for acquiring capacity and performance of a storage device, comprising: Crawling operation log information of the storage device by utilizing a crawling program; the crawling program determines the capacity of the storage equipment according to the operation log information; Analyzing and researching I/O performance evaluation of the storage equipment through a preset storage system I/O performance analysis model according to the request service information in the operation log, wherein the storage system I/O performance analysis model reflects the dynamic characteristic of node performance through state transition; The formula adopted by the average request number is as follows: ; the formula adopted by the average fault server number is as follows: ; the ratio at which the request encounters a queue full uses the formula: ; The formula adopted by the total request loss rate is as follows: ; Wherein, let the Indicating time of day The number of requests waiting in the cache and persisting in the I/O server is , , Indicating time of day I/O server failure number of (a) with front end The position of the two-dimensional space is determined, There is The I/O servers, lm is the average request number, ED is the average fault server number, PB is the ratio of the requests meeting the queue full, PL is the total request loss rate, v is the transfer rate of the system, and L Q is the average waiting request number in the front end.
2. The storage device capacity and performance collection method of claim 1, wherein crawling the travel log information of the storage device using a crawling program comprises: acquiring a preset query request according to the type of the storage equipment; sending a query request to the storage device by utilizing a crawling program; and the storage equipment acquires self running log information according to the identification information in the query request and feeds back the self running log information to the crawling program.
3. The method for collecting capacity and performance of a storage device according to claim 1, wherein a calculation formula adopted for determining the capacity of the storage device is: A formula of bit calculation, wherein the storage capacity=the number of storage units x storage word length; The formula of calculation in bytes is storage capacity=number of storage units x storage word length/8.
4. The storage device capacity and performance collection method according to claim 1, wherein the storage device performance is represented by the performance of a node, the node performance being represented by the number of requests waiting in a cache and in an I/O server and the number of failures of the I/O server.
5. The method for collecting capacity and performance of a storage device according to claim 4, wherein the process for establishing the storage system I/O performance analysis model comprises: Setting a queuing system of a computer; Wherein the queuing system is in the form of ; Representing a time difference distribution of successive arrival of requests; a distribution that provides service time on behalf of the server; representing the number of computer servers; representing the capacity of the computer; representing the number of requests; Rules for providing services on behalf of the computer.
6. The method for collecting capacity and performance of a storage device according to claim 5, wherein a preset negative index distribution and corresponding parameters are substituted into the queuing system; Wherein the parameters are Is a distribution of request arrival time intervals, And the front end is provided with The position of the two-dimensional space is determined, If there is The number of I/O servers that are connected, There is The number of cache spaces in the cache memory is, The request source is The service time obeying parameter is Is a negative-exponential distribution of (a), ; When the waiting queue length is greater than 0, the service time compliance parameter is used When the waiting queue length is less than or equal to 0, the service time compliance parameter is used The rules of first come first served are adhered to.
7. A storage device capacity and performance acquisition system, comprising: A crawling unit for crawling the operation log information of the storage device by using a crawling program; A capacity determining unit for determining the capacity of the storage device according to the operation log information by using the crawling program; The system comprises a performance evaluation unit, a storage system I/O performance analysis model, a total request loss rate and a request service unit, wherein the performance evaluation unit is used for analyzing and researching I/O performance evaluation on storage equipment through the storage system I/O performance analysis model according to request service information in the operation log, and the storage system I/O performance analysis model reflects the dynamic characteristics of node performance through state transition; The formula adopted by the average request number is as follows: ; the formula adopted by the average fault server number is as follows: ; the ratio at which the request encounters a queue full uses the formula: ; The formula adopted by the total request loss rate is as follows: ; Wherein, let the Indicating time of day The number of requests waiting in the cache and persisting in the I/O server is , , Indicating time of day I/O server failure number of (a) with front end The position of the two-dimensional space is determined, There is The I/O servers, lm is the average request number, ED is the average fault server number, PB is the ratio of the requests meeting the queue full, PL is the total request loss rate, v is the transfer rate of the system, and L Q is the average waiting request number in the front end.
8. The storage device capacity and performance collection system according to claim 7, wherein the crawling unit is configured to, when crawling the operation log information of the storage device using a crawling program: acquiring a preset query request according to the type of the storage equipment; sending a query request to the storage device by utilizing a crawling program; and the storage equipment acquires self running log information according to the identification information in the query request and feeds back the self running log information to the crawling program.

Description

Storage device capacity and performance acquisition method and system Technical Field The invention belongs to the technical field of storage capacity setting, and particularly relates to a method and a system for acquiring capacity and performance of storage equipment. Background With the increasing of storage devices, the maintenance of the storage devices consumes a great deal of manpower and funds, and the risk of human misoperation caused by different operation modes and methods for different devices is increased. In the prior storage capacity determining technology, a physical page query request of a target storage unit is sent to a target storage system, the physical page number and the physical page capacity corresponding to the target storage unit, which are determined based on query identification information and sent by the target storage system, are received, and the capacity corresponding to the target storage unit is determined according to the physical page number and the physical page capacity corresponding to the target storage unit. The acquisition method in the prior art mainly analyzes some local characteristics, and the runtime characteristics based on the I/O load comprise I/O request arrival rate, I/O request size, read-write percentage, access sequence degree and burstiness, and the storage system performance model extraction method comprises the steps of throughput, bandwidth and I/O response time, fitting the characteristic parameters of the I/O load and the average response time of the system by means of multiple regression theory, and extracting the performance model of the storage system on line. But the prior art is accurate and efficient in a storage system environment where the functionality and structure is relatively simple. But the implementation costs are high, and the scale and complexity of the currently mainstream storage systems have exposed significant limitations and disadvantages in terms of accuracy, practicality, effectiveness, etc. Disclosure of Invention In view of the above, the present invention aims to provide a method and a system for acquiring capacity and performance of a storage device, which are used for determining capacity and I/O performance evaluation of the storage device, and have wide adaptability and low cost. The first aspect of the application discloses a method for collecting capacity and performance of a storage device, which comprises the following steps: Crawling operation log information of the storage device by utilizing a crawling program; the crawling program determines the capacity of the storage equipment according to the operation log information; And analyzing and researching the I/O performance evaluation of the storage equipment through a preset storage system I/O performance analysis model according to the request service information in the operation log. Optionally, crawling the operation log information of the storage device by using a crawling program includes: acquiring a preset query request according to the type of the storage equipment; sending a query request to the storage device by utilizing a crawling program; and the storage equipment acquires self running log information according to the identification information in the query request and feeds back the self running log information to the crawling program. Optionally, a calculation formula adopted for determining the capacity of the storage device is: A formula of bit calculation, wherein the storage capacity=the number of storage units x storage word length; The formula of calculation in bytes is storage capacity=number of storage units x storage word length/8. Optionally, the performance of the storage device is represented by the performance of the node, and the node performance is represented by the number of requests waiting in the cache and in the I/O server and the number of failures of the I/O server. Optionally, the process for establishing the storage system I/O performance analysis model includes: The method comprises the steps of setting a queuing system of a computer, wherein the queuing system is in the form of time difference distribution representing sequential arrival of requests, distribution representing service time of a server, number of servers representing the computer, accommodation amount of the computer, number of requests and rules representing service of the computer. Optionally, substituting a preset negative index distribution and corresponding parameters into the queuing system; the negative exponential distribution with parameters is the distribution of request arrival time interval, 0< lambda </infinity, a k position at the front end, 0<k </infinity, if s is provided with an I/O server, 0<s is less than or equal to k </infinity, k-s cache spaces are provided, 0<k-s </infinity, the request source is ++A, the service time obeys the negative exponential distribution with parameters of mu i (i=1, 2), and 0< mu 1＜μ2 </infinity; And when t