KR20180078112A - Apparatus and method for lifetime prediction of storage device - Google Patents

Abstract

The present invention relates to a method and a device for predicting a lifecycle of a storage device. More specifically, the present invention is as follows: measuring a used write traffic value (A) considering a write amplification factor (WAF), which represents a degree to which a write traffic value of a storage device is abnormally amplified, and predicting a lifecycle of the storage device; monitoring a read latency value of the storage device, and predicting a lifecycle of the storage device from the read latency value (B); and determining whether the predicted lifecycle is consistent with a predetermined condition, and if the predicted lifecycle is consistent with a predetermined condition, informing the predicted lifecycle to a user. Accordingly, the present invention considers an available capacity of the storage device and precisely predicts a lifecycle. Furthermore, the present invention protects important data through the lifecycle prediction in advance, and prevents problems that data is not saved in an unexpected situation.

Description

[0001] Apparatus and method for lifetime prediction of storage [

The present invention relates to an apparatus and method for predicting the lifetime of a storage device, and is a technology for predicting lifetime based on a write traffic value and a read latency value of a storage device without any hardware change. In particular, the present invention has a feature of predicting the lifetime of a storage device in consideration of a write amplification factor (WAF) indicating the degree of abnormally amplified write traffic.

SD card is one of the most suitable storage devices for things Internet devices. Small in size and easy to remove, it is easy to use in a variety of environments where an object Internet device is applied. Wide operating temperature and wear leveling techniques improve the reliability of SD cards used in Internet devices. The problem of applying such SD card to object internet devices is difficult to predict life. In the case of a black box, which is a type of Internet device, it stores streaming data in an SD card. The problem that the data is not stored in the black box due to the life characteristic of the SD card Lt; / RTI >

(SMART) and I / O traffic or I / O delay time (SMART), which is a representative conventional technique for predicting the life of a SD card, I / O latency). SMART (Self-Monitoring, Analysis and Reporting Technology), which is mainly used in SSD (Solid State Drive), is a system that monitors and reports potential failures through a reliability check. to be. In SD card, a function similar to S.M.A.R.T. (Self-Monitoring, Analysis and Reporting Technology) can be applied to the hardware to monitor the life span. With this monitoring tool, you can see the SD card's life span relatively accurately and visually. However, there is a problem that can not be used with a general SD card because it is expensive and requires hardware changes compared to an SD card.

처럼 플래시 메모리 특성상 가비지 컬렉션(Garbage collection)이 발생하게 되어, 와프(WAF : Write Amplification Factor) 때문에 실제 쓰여진 데이터 양보다 쓰여진 횟수가 증폭이 되는 것을 고려해야한다. 결국 입출력 트래픽(I/O traffic) 측면에서 수명예측을 할 때 가장 중요한 것은 와프(WAF : Write Amplification Factor)를 정확하게 예측하는 것이다. There are two ways to predict the life span of a SD card without changing its hardware. The first way is to consider the amount of data written to the SD card. The SD card, which consists of a flash memory, has a limited number of write cycles because of its endurance characteristics. The lifetime can be predicted by considering the limited total write count and the amount of data currently written. However,

, Garbage collection occurs due to the characteristics of the flash memory. Therefore, it is necessary to consider that the number of times of writing is amplified by the write amplification factor (WAF) rather than the amount of data actually written. As a result, it is important to accurately predict the write amplification factor (WAF) when predicting lifetime in terms of input / output traffic (I / O traffic).

FIG. 1 is a diagram showing a WAF (Write Amplification Factor) characteristic when data is sequentially written and when data is randomly written. Write Amplification Factor (WAF) has the property that memory utilization is inversely proportional to memory lifetime. In other words, the more memory it fills up (the more memory it uses), the less memory is available. In the case of Internet devices such as black boxes, data is written sequentially. It can be seen that the WAF (Write Amplification Factor) converges to almost 1 like the rightmost bar graph in FIG. 1. The fact that the WAF (Write Amplification Factor) converges to 1 means that there is almost no write amplification if the write request is in a sequential order with almost no random part. On the other hand, if the random portion increases in the write request (the random ratio increases toward the left stick in FIG. 1), the WAF (Write Amplification Factor) also increases. As a result, the write amplification factor (WAF) can be predicted based on the ratio of the random part in the write request.

The second method is to consider I / O latency. In the case of a normal memory, as shown in FIGS. 2 to 3, the read latency is usually less than 2 ms. However, if you use a lot of such normal memory, it becomes a state before wear due to heavy wear. This state is a phenomenon in which the RBT (Raw Bit Err / Rate) increases sharply and the read latency increases significantly due to an increase in the number of file retry operations (Error Retrieval) for ECC . That is, when the lifetime of the memory reaches its limit, there is a limitation that performance is surely deteriorated when operations such as erase and read are performed.

Korean Patent Laid-Open Publication No. 2016-0054933 Korean Patent Laid-Open Publication No. 2015-0017263 Korean Patent Registration No. 1190742 Japanese Laid-Open Patent Publication No. 2014-203467

It is an object of the present invention to provide a method and apparatus for predicting the lifetime of a storage device considering a write traffic volume, a read delay time and an available capacity of a storage device, .

It is another object of the present invention to provide an improved performance of an object Internet device using a lot of streaming data. In particular, in the case of a black box, which is a representative object Internet device, it is intended to solve the problem that the life prediction of the SD card is not sure and the important data is not stored.

The objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided a device for predicting the lifetime of a storage device, the device comprising: A first life predicting unit for predicting a lifetime of the storage device by measuring a write traffic value A of the storage device by monitoring a read latency value of the storage device, A second life predicting unit for predicting a life of the storage device from a read latency value B, a second life predicting unit for determining whether a life expectancy predicted from the first and second life predicting units corresponds to a predetermined condition And a notification unit for informing the user of the predicted life span when the predicted life span corresponds to a predetermined condition.

The first life predicting unit may include a maximum write traffic value calculating unit for calculating a maximum write traffic value by measuring a capacity value and a program / erase cycle of the storage device of the storage device, A waftage value calculation unit for calculating a write amplification factor (WAF) value by measuring a used write value and a random write value of the storage device; A write traffic value calculation unit for calculating a write traffic value A using the calculated write traffic value A and a maximum write traffic value And a remaining life ratio value calculating unit for calculating a remaining life ratio value of the apparatus.

In this case, the maximum write traffic value calculator is a value obtained by multiplying a capacity value of the storage device represented by Equation (1) by a P / E cycle (program / erase cycle) of the storage device.

&Quot; (1) "

The warp value calculating unit may calculate a value obtained by dividing a value calculated by dividing a write value (Write) using a random write value (Write) of the storage device represented by Equation (2) As a weight of write traffic (A).

&Quot; (2) "

The used write traffic value calculating unit is a value obtained by multiplying the used write value (A) expressed by Equation (3) by a value of a write amplification factor (WAF) derived from Equation (2).

&Quot; (3) "

The remaining lifetime ratio value calculator calculates a remaining lifetime ratio value, which is a value obtained by dividing a write traffic value (A) used in the storage device by the maximum write traffic value (Write traffic) represented by Equation (4) Is derived.

&Quot; (4) "

The predetermined condition of the determination unit is expressed by <condition 1> and <condition 2>. When the used write traffic A satisfies <condition 1> or the read latency value B, And executes the notification unit when it satisfies the < condition 2 >. On the other hand, if the used write traffic (A) does not satisfy <condition 1> and the read latency value (B) does not satisfy <condition 2> And the second life predicting unit is executed.

<Condition 1>

(한계 값 P)(Used write traffic value A)

(Limit value P)

<Condition 2>

N(The value of the number of times the read delay time value B satisfying {(read delay time value B) (2 ms XQ) among the number of times M monitoring the latest read delay time value B)

N

(P, Q, N, and M are predetermined values)

According to an aspect of the present invention, there is provided a method of predicting the lifetime of a storage device, the method comprising: predicting a life expectancy of a storage device based on a WAF A first life predicting step of predicting a life time of the storage device by measuring a used write traffic value A considering an amplification factor of the storage device, Estimating a lifetime of the storage device from the read latency value (B) by monitoring a value of the read latency value (B) from the first lifetime prediction step A determination step of determining whether or not a predicted life span corresponds to a preset condition; and a determination step of determining whether the predicted life span corresponds to a predetermined condition, And a notification step of notifying the user.

At this time, in the first life predicting step, the maximum write traffic value calculating unit measures a capacity value and a P / E cycle (program / erase cycle) of the storage device of the storage device to calculate a maximum write traffic value Calculating a maximum write traffic value and a wafue value calculation unit calculating a waf (WAF) value by measuring a write value and a random write value used in the storage, Calculating a used write traffic value (A) of the storage device from the calculated write amplification factor (WAF), calculating a used write traffic value (A) Calculating a remaining lifetime ratio value of the storage device using the write traffic value A and the calculated maximum write traffic value ; And

The maximum write traffic value calculating step is a value obtained by multiplying a capacity value of the storage device represented by Equation (1) by a program / erase cycle of the storage device.

&Quot; (1) &quot;

The warp value calculation step is a value obtained by dividing a value calculated by dividing by a write value (Write) using a random write value (Write) of the storage device represented by Equation (2) Is applied as the weight of the write traffic (A).

&Quot; (2) &quot;

The used write traffic value calculating step is a value obtained by multiplying the used write value (A) expressed by Equation (3) by a write amplification factor (WAF) derived from Equation (2).

&Quot; (3) &quot;

The remaining lifetime ratio value calculation step calculates the remaining lifetime ratio value by multiplying the write traffic (A) used in the storage device by the maximum write traffic value (write traffic) represented by Equation (4) And deriving a ratio value.

&Quot; (4) &quot;

The predetermined condition of the determination step is expressed as <condition 1> and <condition 2>, and the write traffic (A) used by the notification part satisfies <condition 1> or the read latency value B) satisfies < Condition 2 >, the notifying step is executed. On the other hand, if the write traffic (A) used by the first and second life predicting units does not satisfy <condition 1> and the read latency value B satisfies <condition 2> The first life predicting step and the second life predicting step are executed.

<Condition 1>

(한계 값 P)(Used write traffic value A)

(Limit value P)

<Condition 2>

N(The value of the number of times the read delay time value B satisfying {(read delay time value B) (2 ms XQ) among the number of times M monitoring the latest read delay time value B)

N

(P, Q, N, and M are predetermined values)

According to the apparatus and method for predicting the life of the storage device as described above,

First, the life expectancy factor of the storage device can be more accurately predicted considering the write traffic volume, the read delay time, and the available capacity of the storage device.

Second, there is an effect of improving the performance of the object Internet device using a lot of streaming data.

Third, in the case of black box, which is a typical object Internet device, it is possible to predict the life of the SD card, and thus it is possible to store important data.

Fourth, it has the effect of predicting the life without changing the hardware.

Fifth, since the predicted life time point of the storage device to be received by the user can be separately set (P, Q, N, and M are predetermined values), it is possible to prevent the problem that data is not stored in an unexpected situation .

FIG. 1 is a diagram showing a WAF (Write Amplification Factor) characteristic when data is sequentially written and when data is randomly written. FIG.
FIG. 2 is a diagram showing a result of measuring a read latency according to 100 commands; FIG.
FIG. 3 is a diagram showing a result of measuring a read latency according to 1000 commands; FIG.
4 is a block diagram illustrating a device for predicting the life of a storage device according to an embodiment of the present invention.
5 is a flowchart illustrating a method of predicting the service life of a storage device according to an embodiment of the present invention.
FIG. 6 is a flowchart for explaining the first life prediction step in FIG. 5; FIG.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An apparatus and method for estimating the service life of a storage device according to embodiments of the present invention will be described in detail with reference to the accompanying drawings. The present invention is capable of various modifications and various forms, and specific embodiments are illustrated in the drawings and described in detail in the text. It is to be understood, however, that the invention is not intended to be limited to the particular forms disclosed, but on the contrary, is intended to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. Like reference numerals are used for like elements in describing each drawing. In the accompanying drawings, the dimensions of the structures are enlarged to illustrate the present invention, and are actually shown in a smaller scale than the actual dimensions in order to understand the schematic structure. Also, the terms first and second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. On the other hand, unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Do not.

FIG. 4 is a block diagram illustrating an apparatus for predicting the life of a storage device according to an embodiment of the present invention.

As shown in FIG. 4, the device for predicting the service life of a storage device includes a write traffic (write traffic) considering a write amplification factor (WAF) indicating a degree of abnormally amplifying a write traffic value of a storage device, A first latency predicting unit 100 for predicting a lifetime of the storage device by measuring a value A of the storage device and a read latency value of the storage device, A second life predicting unit 200 for predicting the life of the storage device from the value B, a predicted life expectancy from the first and second life predicting units 100 and 200, And a notification unit 400 for informing the user of the predicted life span when the estimated life span corresponds to a predetermined condition. The first service life predicting unit 100 includes a maximum write traffic value calculator 110 for calculating a maximum write traffic value by measuring a capacity value and a P / E cycle (program / erase cycle) A warp value calculation unit 120 for calculating a write amplification factor (WAF) by measuring a write value and a random write value used in the storage device, and a wobble value calculation unit 120 for calculating a write amplification factor A used write traffic value calculating unit 130 for calculating a write traffic value A used in the storage device and a write traffic value calculating unit 130 for calculating a write traffic value A calculated using the calculated maximum write traffic value A, And a remaining life ratio value calculation unit 140 for calculating a remaining life ratio value of the storage device using a write traffic value.

The maximum write traffic value calculation unit 110 calculates a maximum write traffic value of the storage device by multiplying a capacity value of the storage device expressed by Equation (1) by a P / E cycle (program / erase cycle) .

&Quot; (1) &quot;

The warp value calculator 120 calculates a warp value by adding a value calculated by dividing a write value (Write) using a random write value (Write) of the storage device represented by Equation (2) Is calculated.

&Quot; (2) &quot;

The used write traffic value calculator 130 multiplies the used write value A expressed by Equation 3 by a write amplification factor WAF derived from Equation 2, Calculate the used write traffic value of the device.

&Quot; (3) &quot;

Accordingly, the remaining lifetime ratio calculation unit 140 calculates a value obtained by dividing the write traffic (A) used in the storage device expressed by Equation (4) by the maximum write traffic value (Write traffic) The remaining life ratio value of the storage device is calculated. In this case, the maximum write traffic value is a value calculated from the maximum write traffic value calculating unit 110, and the write traffic value (A) used by the storage device is the write traffic value (130). &Lt; / RTI &gt;

&Quot; (4) &quot;

Accordingly, the determination unit 300 determines whether the lifetime of each of the storage devices predicted by the first and second life predicting units 100 and 200 corresponds to predetermined conditions. The predetermined condition of the determination unit 300 is expressed by < Condition 1 > and < Condition 2 >. The determination unit 300 determines whether the used write traffic A corresponds to the <condition 1>, and the read latency value B corresponds to <condition 2> .

<Condition 1>

(한계 값 P)(Used write traffic value A)

(Limit value P)

<Condition 2>

N(The value of the number of times the read delay time value B satisfying {(read delay time value B) (2 ms XQ) among the number of times M monitoring the latest read delay time value B)

N

(P, Q, N, and M are predetermined values)

When the used write traffic A satisfies <condition 1> or the read latency value B satisfies <condition 2> (<condition 1>), Or < Condition 2 > is satisfied), the notification is executed.

In contrast, the first and second lifetime predicting units 100 and 200 may determine that the write traffic A used does not satisfy the <condition 1> and the read latency value B) does not satisfy <condition 2> (when both <condition 1> and <condition 2> are not satisfied), the first life prediction and the second life prediction are executed.

5 is a flowchart illustrating a method of predicting the service life of a storage device according to an embodiment of the present invention.

As shown in FIG. 5, the method of predicting the lifetime of a storage device may be a method of predicting the lifetime of a storage device, such as a WAF (write amplification factor) indicating a degree of abnormally amplifying a write traffic value of a storage device, And estimates the life of the storage device by measuring the used write traffic value A in step S100. The second life predicting unit 200 monitors the read latency value of the storage device and predicts the lifetime of the storage device from the read latency value B ). The determination unit 300 determines whether the life expectancy predicted from the steps S100 and S200 corresponds to a predetermined condition (S300). If the predicted life span corresponds to the predetermined condition, the notification unit 400 informs the user of the predicted life span (S400).

FIG. 6 is a flowchart for explaining the step S100 in FIG.

6, the maximum write traffic value calculator 110 measures a capacity value and a P / E cycle (program / erase cycle) of the storage device of the storage device to determine a maximum write traffic (S100). The warp value calculating unit 120 measures a write value used in the storage device and a random write value to calculate a write amplification factor (WAF) (S120). The used write traffic value calculating unit 130 calculates the write traffic value A used by the storage device from the calculated write amplification factor WAF at step S130. The remaining lifetime ratio value calculating unit 140 calculates the remaining life ratio value of the storage device using the calculated write traffic value A and the calculated maximum write traffic value (S140).

The process for predicting the service life of the storage device in step S100 is as follows.

In step S110, the maximum write traffic value of the storage device is calculated by multiplying the capacity value of the storage device by the P / E cycle (program / erase cycle) of the storage device using Equation (1) do.

In step S120, the value calculated by dividing the write value (Write) using a random write value (Write) of the storage device by 1 is added to the value stored in the storage device using Equation (2) Is calculated.

In step S130, as described above, the value of the write (A) used in the above equation (3) is multiplied by the value of a write amplification factor (WAF) derived from the equation (2) And calculates the used write traffic value.

As a result, in step S140, the value calculated by dividing the write traffic (A) used in the storage device by the maximum write traffic value (Write traffic) is subtracted from 1 by using Equation (4) The remaining life ratio value of the storage device is calculated.

For example, if the number of P / E cycles (program / erase cycles) that can be written to and erased from a storage device of 8 GB is 10,000, the maximum write traffic value of 80,000, which is 8 X 10,000, .

In this case, if the user actually uses the storage device with a write value of 20,000 GB, and if the random write value of the storage device is 4,000 GB, the write amplification factor (WAF) The value is 1.2, which is 1+ (4,000 / 20,000).

Then, the calculated Write Amplification Factor (WAF) value 1.2 is applied as the weight of the write traffic value (A) actually used by the storage device. Therefore, the write traffic (A) actually used in the storage device becomes 24,000, which is 20,000 X 1.2 when the above Equation (3) is used.

If the remaining life ratio value of the storage device is obtained by using Equation (4) described above, the value becomes 0.7 (70%) which is 1- (24,000 / 80,000).

In step S300, the determination unit 300 determines whether the lifetime of each storage device predicted from steps S100 and S200 corresponds to a predetermined condition. The predetermined condition of step S300 is expressed by < Condition 1 > and < Condition 2 > as described above. In step S300, it is determined whether or not the used write traffic value A corresponds to the <condition 1>. If the read latency value B corresponds to the <condition 2> do.

If the used write traffic (A) satisfies <condition 1> or the read latency value (B) satisfies <condition 2> (<condition 1> or <condition 2> If any one of them is satisfied, the alarm unit 400 executes step S400. In operation S400, the lifetime recommendation of the storage device may be applied as one embodiment, and a lifetime recommendation is provided to the user through TCP / IP (Transport Control Protocol / Internet Protocol) or Bluetooth communication. Further, the lifetime recommendation and notification as an embodiment of the present invention can be implemented in various forms that the user can visually recognize through the LED or inform the user through the speaker through voice.

On the other hand, when the used write traffic (A) does not satisfy <condition 1> and the read latency value (B) does not satisfy <condition 2> (<condition 1> and <condition 2> are not satisfied), the first life predicting unit 100 to the second life predicting unit 200 execute steps S100 to S200.

For example, in the case of <condition 1>, as the threshold value P is set higher, the write traffic value (A) used by the storage device receiving the notification is higher. Conversely, the lower the threshold value P is set, the lower the write traffic value (A) used by the storage device which receives the notification is.

For example, if <Condition 2> is set to M = 20, Q = 5 and N = 15, the read delay time of 10 ms (the read delay time of normal storage device (2 ms) Read latency) is 15 or more, it is determined that the condition <2> is satisfied, and the user is informed of the expected lifetime of the storage device.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. It is clear that the present invention can be suitably modified and applied in the same manner. Therefore, the above description does not limit the scope of the present invention, which is defined by the limitations of the following claims.

Claims (16)

The used write traffic value (A), which considers a write amplification factor (WAF) indicating the degree to which the write traffic value of the storage device is abnormally amplified, is measured to predict the life of the storage device A first life predicting unit;
A determination unit for determining whether the life expectancy predicted from the first life predicting unit corresponds to a predetermined condition; And
And a notification unit for informing the user of the predicted life span when the predicted life span corresponds to a preset condition. The used write traffic value (A), which considers a write amplification factor (WAF) indicating the degree to which the write traffic value of the storage device is abnormally amplified, is measured to predict the life of the storage device A first life predicting unit;
A second life predicting unit monitoring the read latency value of the storage device to predict the life of the storage device from the read latency value B;
A determination unit for determining whether a life expectancy predicted from the first and second life predicting units corresponds to a predetermined condition; And
And a notification unit for informing the user of the predicted life span when the predicted life span corresponds to a preset condition. 3. The method according to claim 1 or 2,
The first life predicting unit predicts,
A maximum write traffic value calculation unit for calculating a maximum write traffic value by measuring a capacity value and a P / E cycle (program / erase cycle) of the storage device;
A waf value calculation unit for calculating a write amplification factor (WAF) by measuring a used write value and a random write value of the storage device;
A used write traffic value calculating unit for calculating a write traffic value A used in the storage device from the calculated write amplification factor (WAF); And
And a remaining life ratio value calculating unit for calculating a remaining life ratio value of the storage device by using the calculated write traffic value A and the calculated maximum write traffic value, Wherein the lifetime prediction unit estimates the lifetime of the storage device. The method of claim 3,
The maximum write traffic value calculation unit may calculate,
(1) is a value obtained by multiplying a capacity value of the storage device expressed by Equation (1) by a P / E cycle (program / erase cycle) of the storage device.
&Quot; (1) &quot;

The method of claim 3,
The warp value calculating unit calculates,
Is a value obtained by dividing a value calculated by dividing a write value (Write) using a random write value (Write) of the storage device represented by Equation (2) by 1 and adding the value to the write traffic value A) is a weight of the storage device.
&Quot; (2) &quot;

The method of claim 3,
The used write traffic value calculator calculates,
Is a value obtained by multiplying the used write value (A) expressed by Equation (3) by a WAF (Write Amplification Factor) value derived from Equation (2).
&Quot; (3) &quot;

The method of claim 3,
The remaining lifetime ratio value calculation unit calculates,
The remaining lifetime ratio value that is a value obtained by dividing a value obtained by dividing the used write traffic (A) used in the storage device expressed by Equation (4) by the maximum write traffic value (Write traffic) is deduced. Wherein the storage device is a storage device.
&Quot; (4) &quot;

3. The method according to claim 1 or 2,
The predetermined condition of the determination unit is represented by < Condition 1 > and < Condition 2 &
Executes the notification unit when the used write traffic value A satisfies <condition 1> or the read latency value B satisfies <condition 2>
When the used write traffic value A does not satisfy the condition 1 and the read latency value B does not satisfy the condition 2, And a second life predicting unit is executed.
<Condition 1>
(Used write traffic value A)

(Limit value P)
<Condition 2>
(The value of the number of times the read delay time value B satisfying {(read delay time value B) (2 ms XQ) among the number of times M monitoring the latest read delay time value B)

N
(P, Q, N, and M are predetermined values) The first life predicting unit measures a used write traffic value A considering a write amplification factor (WAF) indicating an extent of abnormally amplifying a write traffic value of the storage device, A first life prediction step of predicting the life of the device;
A judgment step of judging whether or not the life expectancy predicted from the first life prediction step corresponds to a predetermined condition; And
And a notification step of informing the user of the predicted life span when the notifying unit corresponds to the predicted life span of the predetermined condition. The first life predicting unit measures a used write traffic value A considering a write amplification factor (WAF) indicating an extent of abnormally amplifying a write traffic value of the storage device, A first life prediction step of predicting the life of the device;
A second life predicting step of predicting a lifetime of the storage device from the read latency value (B) by monitoring a read latency value of the storage device;
A judgment step of judging whether or not the life expectancy predicted from the first life expectancy step and the second life expectancy step corresponds to a predetermined condition; And
And a notification step of informing the user of the predicted life span when the notifying unit corresponds to the predicted life span of the predetermined condition. 11. The method according to claim 9 or 10,
Wherein the first life prediction step comprises:
A maximum write traffic value calculating step of calculating a maximum write traffic value by measuring a capacity value and a program / erase cycle of the storage device;
Calculating a warp value (WAF) by measuring a write value used in the storage device and a random write value;
A used write traffic value calculating unit calculates a write traffic value A used by the storage device from the calculated write amplification factor (WAF); And
The remaining lifetime ratio value calculating unit calculates a remaining lifetime ratio value (A) for calculating the remaining lifetime ratio value of the storage device using the calculated write traffic value (A) and the calculated maximum write traffic value And calculating the lifetime of the storage device. 12. The method of claim 11,
Wherein the maximum write traffic value calculation step comprises:
(1) is a value obtained by multiplying a capacity value of the storage device expressed by Equation (1) by a P / E cycle (Program / erase cycle) of the storage device.
&Quot; (1) &quot;

12. The method of claim 11,
The warp value calculating step may include:
Is a value obtained by dividing a value calculated by dividing a write value (Write) using a random write value (Write) of the storage device represented by Equation (2) by 1 and adding the value to the write traffic value A) is a weight of the storage device.
&Quot; (2) &quot;

12. The method of claim 11,
The used write traffic value calculating step may include:
Is a value obtained by multiplying the used write value (A) expressed by Equation (3) by a value of Write Amplification Factor (WAF) derived from Equation (2).
&Quot; (3) &quot;

12. The method of claim 11,
The remaining lifetime ratio value calculating step may include:
The remaining lifetime ratio value that is a value obtained by dividing a value obtained by dividing the used write traffic (A) used in the storage device expressed by Equation (4) by the maximum write traffic value (Write traffic) is deduced. Of the storage device.
&Quot; (4) &quot;

11. The method according to claim 9 or 10,
The predetermined condition of the judgment step is represented by < Condition 1 > and < Condition 2 &
The notification step is executed when the write traffic value A used by the notification unit satisfies <condition 1> or the read latency value B satisfies <condition 2>
If the write traffic A used by the first and second life predicting units does not satisfy <condition 1> and the read latency value B does not satisfy <condition 2> The first life prediction step to the second life estimation step is executed.
<Condition 1>
(Used write traffic value A)

(Limit value P)
<Condition 2>
(The value of the number of times the read delay time value B satisfying {(read delay time value B) (2 ms XQ) among the number of times M monitoring the latest read delay time value B)

N
(P, Q, N, and M are predetermined values)

Images