CN103761057B - A kind of method improving hard disk reliability - Google Patents

Abstract

The invention discloses a kind of method improving hard disk reliability, including: obtain the parameter of different sampling instant hard disk array, calculate the membership values corresponding to parameter of different sampling instant hard disk array, including temperature, performance, operation time and power consumption, obtain hard disk temperature, performance, the meansigma methods A of the membership values that the operation time is corresponding, B, the minima D of C and power consumption correspondence membership values, obtain the polymerizing value F=aA+bB+cC+dD of this hard disk array, wherein a=0.1, b=0.2, c=0.3, d=0.4, change the size of disk array, and repeat the above steps, the polymerizing value corresponding to obtain different size disk array, and find the disk array making the maximum correspondence of polymerizing value as the most reliable disk array.Compared with the existing methods, the method for the present invention can reduce the temperature of ten to three 12 percent.Hard disk error rate can be efficiently reduced, improve the reliability of hard disk.

Description

A kind of method improving hard disk reliability

Technical field

The invention belongs to Computer Storage field, more particularly, to a kind of method improving hard disk reliability.

Background technology

Storage system is faced with the problems such as the huge electricity charge and high error rate, and these are all relevant with the excessive temperature of storage system, and experiment statistics finds that temperature affects the life-span of hard disk to a great extent.For the problem that energy consumption is bigger, the Energy Saving Strategy of current storage system is hard disk by load centralization to minority as far as possible, allows more hard disk in a dormant state, reduces the energy consumption that high speed rotating causes.Namely pay close attention to a large amount of loads on a small amount of hard disk, but this strategy can cause that high capacity hard disk is in hot operation state all the time, cause hard disk failure rate to steeply rise.And carrying out too much I/O operation on a small amount of hard disk and may result in the difference of temperature between hard disk, such system can not be freezed effectively.Compared with CPU, storage system carrying out temperature under existence conditions and controls increasingly difficult, first, the relation between temperature and electricity in storage system is non-linear and discrete, and this just controls to bring great amount of calculation to temperature；Secondly, the feature storing system response slow determines under precise control, and system easily enters labile state；Finally, the management of hard disk must also be considered that the access times of disk are limited.It addition, the raising of temperature can make storage system just face performance bottleneck.Therefore, control temperature, thus improving storage equipment reliability, there is very great meaning.

Summary of the invention

For disadvantages described above or the Improvement requirement of prior art, the invention provides a kind of method improving hard disk reliability, its object is to, by introducing temperature prediction and feedback mechanism, be effectively improved the reliability of storage device.

For achieving the above object, according to one aspect of the present invention, it is provided that a kind of method improving hard disk reliability, comprise the following steps:

(1) obtain the parameter of different sampling instant hard disk array, including temperature (t0, t1 ..., tn), performance (p0, p1 ..., pn), run the time (o0, o1 ..., on) and power consumption (c0, c1 ..., cn), wherein n represents the quantity of sampled point；

(2) membership values corresponding to parameter of different sampling instant hard disk array is calculated, including temperature (T_t0,T_t1,…,T_tn), performance (P_p0,P_p1,…,P_pn), run time (O_o0,O_o1,…,O_on) and power consumption (C_c0,C_c1,…,C_cn)；

(3) the meansigma methods A=(T of the membership values that obtaining step (2) obtains hard disk temperature, performance, operation time are corresponding_t0+T_t1+…+T_tn)/n, B=(P_t0+P_t1+…+P_tn)/n, C=(O_t0+O_t1+…+O_tnThe minima D of)/n and power consumption correspondence membership values；

(4) the polymerizing value F=aA+bB+cC+dD of this hard disk array is obtained, wherein a=0.1, b=0.2, c=0.3, d=0.4；

(5) change the size of disk array, repeat the above steps (1) to (4), the polymerizing value corresponding to obtain different size disk array, and find the disk array making the maximum correspondence of polymerizing value as the most reliable disk array.

Preferably, step (2) specifically adopts below equation:

$T_{\mathrm{tn}}=\left\{\begin{array}{cc}1,& \mathrm{tn}\&le;30\\ \frac{50-\mathrm{tn}}{50-30},& 30<\mathrm{tn}<50\\ 0,& \mathrm{tn}\&GreaterEqual;50\end{array}\right.$

$P_{\mathrm{pn}}=\left\{\begin{array}{cc}0,& \frac{\mathrm{pn}}{p0}\&le;0.8\\ \frac{\frac{\mathrm{pn}}{p0}-0.8}{1.2-0.8},& 0.8<\frac{\mathrm{pn}}{p0}<1.2\\ 1,& \frac{\mathrm{pn}}{p0}\&GreaterEqual;1.2\end{array}\right.$

$O_{\mathrm{on}}=\left\{\begin{array}{cc}0,& \mathrm{on}\&le;30\\ \frac{\mathrm{on}-30}{60-30},& 30<\mathrm{on}<60\\ 1,& \mathrm{on}\&GreaterEqual;60\end{array}\right.$

$C_{\mathrm{cn}}=\exp (-\frac{{\left(S_n{P}_n\right)}^2}{2{\mathrm{\&sigma;}}^2})$

Wherein, S_nFor the Disk State vector in n moment, P_nFor the power consumption vector of n moment each disk power consumption composition, σ is constant, and its value is the half of maximum power dissipation in sampling interval.

In general, by the contemplated above technical scheme of the present invention compared with prior art, it is possible to obtain following beneficial effect:

(1) present invention can be effectively improved the reliability of storage device: owing to have employed step (1) and step (2), the method taking fuzzy control to storage system, temperature is obtained in sample point, performance, the occurrence of operation time and these four parameters of power consumption, calculates it in each self-corresponding membership values of each sample point.Relative to accurate control, adopt fuzzy control not only to greatly reduce the amount of calculation of storage system, it is thus also avoided that the undesirable state easily often occurred under precise control, make storage system be easier to enter steady statue.

(2) present invention adopts the mechanism of temperature prediction, in step (2), the temperature of the latter half, performance, operation time and power consumption are all obtained by prediction, wherein the prediction of temperature is according to the heat derivation of equation out, adopts forecasting mechanism can reduce the expense of storage system to a great extent.

(3) polymerizing value can be repeatedly polymerized by the present invention, in step (3), to temperature, the membership values of performance and the time of operation, averages；Membership values to power consumption, goes the minima of membership values in all sampled points, and this is first time polymerization.To four values obtained in step (3), step (4) carries out polymerization again and calculates, obtain an end value.By the polymerization of twice, it is possible to make the such a abstract concept of reliability of storage system be expressed intuitively by a numerical value.

(4) present invention can repeatedly adjust disk array, finds and makes storage system reliability the highest.In step (5), repeatedly adjust the size of disk array, find and make the disk array that in step (4), end value is maximum.The comparison of system reliability transfers the comparison of end value size to, so can be more easily found the disk array that reliability is the highest.

Accompanying drawing explanation

Fig. 1 is the flow chart that the present invention improves the method for hard disk reliability.

Detailed description of the invention

In order to make the purpose of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein is only in order to explain the present invention, is not intended to limit the present invention.As long as just can be mutually combined additionally, technical characteristic involved in each embodiment of invention described below does not constitute conflict each other.

A kind of method that the Integral Thought of the present invention is in that to provide fuzzy control, so-called fuzzy control, it is exactly to temperature, power consumption, runs the time, and performance calculates with respective membership function respectively and obtains result, the result of membership function is on the interval of 0 to 1, obtained result, closer to 1, means that the requirement more meeting control, and 0 represents undesirable.Wherein the membership function of power consumption is exponential form, and its value is 0 never, namely represents that power consumption always meets the requirements.Other then use linear membership function, and result can be 0.

Controller is to predict the performance of the storage system in a sampling section.Concrete grammar is to be divided into two parts the time period that will predict, front portion is the control stage, and the latter half is forecast period.Controlling the stage, controller can carry out different input, it was predicted that the stage, according to the input controlling the stage, can be predicted the storage system temperature in this stage, power consumption, run time, performance.Wherein the prediction of temperature is the temperature difference and two relating to parameters that go out between two sampled points according to the heat derivation of equation, then obtains the value of the two parameter according to recursive least squares.Thus can according to the temperature prediction of a upper sampled point to temperature now.

By the data that control stage and test phase obtain, the power consumption obtained by the whole each sampled point of sampling section, temperature, performance, the membership values of operating time is polymerized respectively, the polymerizing value obtained again is polymerized, and obtains a final value.

According to the input that the control stage is different, it is possible to the end value obtained is also different.So constantly adjust the input in control stage, select a maximum value.The control input that this value is corresponding, it is simply that by power consumption, performance, temperature, the input policing that the operating time combines best.

As it is shown in figure 1, the method that the present invention improves hard disk reliability comprises the following steps:

(1) parameter of different sampling instant hard disk array is obtained, including temperature (t0, t1, ..., tn), performance (p0, p1, ..., pn), run time (o0, o1, ..., on) and power consumption (c0, c1, ..., cn), wherein n represents the quantity of sampled point, its span is 50 to 100, and the interval of sampling is 75 minutes to 90 minutes；

(2) membership values corresponding to parameter of different sampling instant hard disk array is calculated, including temperature (T_t0,T_t1,…,T_tn), performance (P_p0,P_p1,…,P_pn), run time (O_o0,O_o1,…,O_on) and power consumption (C_c0,C_c1,…,C_cn)；T_tn, P_pn, O_on, C_cnConcrete formula as follows:

$T_{\mathrm{tn}}=\left\{\begin{array}{cc}1,& \mathrm{tn}\&le;30\\ \frac{50-\mathrm{tn}}{50-30},& 30<\mathrm{tn}<50\\ 0,& \mathrm{tn}\&GreaterEqual;50\end{array}\right.$

$P_{\mathrm{pn}}=\left\{\begin{array}{cc}0,& \frac{\mathrm{pn}}{p0}\&le;0.8\\ \frac{\frac{\mathrm{pn}}{p0}-0.8}{1.2-0.8},& 0.8<\frac{\mathrm{pn}}{p0}<1.2\\ 1,& \frac{\mathrm{pn}}{p0}\&GreaterEqual;1.2\end{array}\right.$

$O_{\mathrm{on}}=\left\{\begin{array}{cc}0,& \mathrm{on}\&le;30\\ \frac{\mathrm{on}-30}{60-30},& 30<\mathrm{on}<60\\ 1,& \mathrm{on}\&GreaterEqual;60\end{array}\right.$

$C_{\mathrm{cn}}=\exp (-\frac{{\left(S_n{P}_n\right)}^2}{2{\mathrm{\&sigma;}}^2})$

Wherein S_nFor the Disk State vector in n moment, i.e. the operation halted state of disk, P_nFor the power consumption vector of n moment each disk power consumption composition, σ is constant, and its value is the half of maximum power dissipation in sampling interval.

(3) the meansigma methods A=(T of the membership values that obtaining step (2) obtains hard disk temperature, performance, operation time are corresponding_t0+T_t1+…+T_tn)/n, B=(P_t0+P_t1+…+P_tn)/n, C=(O_t0+O_t1+…+O_tnThe minima D of)/n and power consumption correspondence membership values；

(4) the polymerizing value F=aA+bB+cC+dD of this hard disk array is obtained, wherein a=0.1, b=0.2, c=0.3, d=0.4；

(5) change the size of disk array, repeat the above steps (1) to (4), the polymerizing value corresponding to obtain different size disk array, and find the disk array making the maximum correspondence of polymerizing value as the most reliable disk array.

Those skilled in the art will readily understand; the foregoing is only presently preferred embodiments of the present invention; not in order to limit the present invention, all any amendment, equivalent replacement and improvement etc. made within the spirit and principles in the present invention, should be included within protection scope of the present invention.

Claims (2)

1. the method improving hard disk reliability, it is characterised in that comprise the following steps:

(1) parameter of different sampling instant hard disk array is obtained, including temperature (t0, t1, ..., tn), performance (p0, p1, ..., pn), run time (o0, o1, ..., on) and power consumption (c0, c1 ..., cn), wherein n represents the quantity of sampled point；

(2) membership values corresponding to parameter of different sampling instant hard disk array is calculated, including temperature (T_t0,T_t1,…,T_tn), performance (P_p0,P_p1,…,P_pn), run time (O_o0,O_o1,…,O_on) and power consumption (C_c0,C_c1,…,C_cn)；

(3) the meansigma methods A=(T of the membership values that obtaining step (2) obtains hard disk temperature, performance, operation time are corresponding_t0+T_t1+…+T_tn)/n, B=(P_t0+P_t1+…+P_tn)/n, C=(O_t0+O_t1+…+O_tnThe minima D of)/n and power consumption correspondence membership values；

(4) the polymerizing value F=aA+bB+cC+dD of this hard disk array is obtained, wherein a=0.1, b=0.2, c=0.3, d=0.4；

(5) change the size of disk array, repeat the above steps (1) to (4), the polymerizing value corresponding to obtain different size disk array, and find the disk array making the maximum correspondence of polymerizing value as the most reliable disk array.

2. the method for raising hard disk reliability according to claim 1, it is characterised in that step (2) specifically adopts below equation:

$T_{tn}=\left\{\begin{array}{cc}1,& tn\&le;30\\ \frac{50-tn}{50-30},& 30<tn<50\\ 0,& tn\&GreaterEqual;50\end{array}\right.$

$P_{pn}=\left\{\begin{array}{cc}0,& \frac{pn}{p0}\&le;0.8\\ \frac{\frac{pn}{p0}-0.8}{1.2-0.8},& 0.8<\frac{pn}{p0}<1.2\\ 1,& \frac{pn}{p0}\&GreaterEqual;1.2\end{array}\right.$

$O_{on}=\left\{\begin{array}{cc}0,& on\&le;30\\ \frac{on-30}{60-30},& 30<on<60\\ 1,& on\&GreaterEqual;60\end{array}\right.$

$C_{cn}=\exp (-\frac{{\left(S_n{P}_n\right)}^2}{2{\mathrm{\&sigma;}}^2})$

Wherein, S_nFor the Disk State vector in n moment, P_nFor the power consumption vector of n moment each disk power consumption composition, σ is constant, and its value is the half of maximum power dissipation in sampling interval.