SMART ( Self-Monitoring, Analysis, and Technology Reporting , often written as SMART ) is a monitoring system that is included in the computer's hard disk drive (HDD ), solid-state drives (SSDs), and eMMC drives. Its main function is to detect and report on various drive reliability indicators with the intention of anticipating an imminent hardware failure.
When S.M.A.R.T. the data indicates a potential drive failure, the software running on the host system can notify the user so that precautions can be taken to prevent data loss, and the failed drives can be replaced and data integrity maintained.
Video S.M.A.R.T.
​​â € <â €
Hard disk failure (and Flash drive failure, but not exactly the same way) is included in one of two basic failure classes:
- Unpredictable failure, resulting from a slow process such as mechanical wear and gradual storage degradation. Monitoring can determine when the failure becomes more likely.
- Unpredictable failures, occurring without warning and from electronic components being damaged by sudden mechanical failure (which may be related to improper handling).
Mechanical failure accounts for about 60% of all drive failures. While ultimate failure may be disastrous, most mechanical failures result from gradual wear and tear and there is usually a certain indication that failure is imminent. This may include increased heat output, increased noise levels, problems with reading and writing data, or an increase in the number of damaged disk sectors.
PCTechGuide Page at S.M.A.R.T. (2003) commented that technology has gone through three phases:
In its original incarnation, S.M.A.R.T. provides a prediction of failure by monitoring certain hard drive activity online.
The next version of the standard increase predictive failure by adding automatic off-line read scanning to monitor additional operations. "S.M.A.R.T." latest technology not only monitors hard drive activity but adds failure prevention by trying to detect and correct sector errors. Also, while previous versions of the technology only monitor the hard drive activity for data retrieved by the operating system, this is S.M.A.R.T. latest tests all data and all drive sectors by using "off-line data collection" to confirm drive health during periods of inactivity.
Accuracy
A field study on Google covering over 100,000 consumer-grade hard disks from December 2005 to August 2006 found a correlation between S.M.A.R.T. certain. information and actual failure rate:
- Within 60 days after the first non-fixable drive error (SMART attribute 0xC6 or 198) detected as a result of offline scanning, the drive was, on average, 39 times more likely to fail than a similar drive that was not something like that goes wrong.
- The first error in reallocation, offline reallocation (S.M.A.R.T. attributes 0xC4 and 0x05 or 196 and 5) and the probational counting (attribute S.M.A.R.T. 0xC5 or 197) are also strongly correlated with higher probability of failure.
- In contrast, small correlations are found for temperature increase and there is no correlation for usage levels. However, research indicates that most (56%) of failed drive failures without recording counts in "four powerful S.M.A.R.T.sat" warnings are identified as scanning errors, recalculation, offline reordering, and probational counts.
- Furthermore, 36% of failed drives do so without recording S.M.A.R.T. error at all, except temperature, which means that S.M.A.R.T. data alone has limited utility in anticipating failure.
Maps S.M.A.R.T.
History and predecessors
Early hard disk monitoring technology was introduced by IBM in 1992 in the IBM 9337 Disk Array for AS/400 servers using IBM 0662 SCSI-2 disk drive. Lately it is named Predictive Failure Analysis (PFA) technology. It measures some health parameters of the main device and evaluates it in the drive firmware. Communication between physical units and monitoring software is limited to binary results: that is, "the device is fine" or "the drive is likely to fail soon".
Later, another variant, called IntelliSafe, was created by computer manufacturer Compaq and disk drive manufacturers Seagate, Quantum, and Conner. The disk drive will measure the "health parameters" of the disk, and its value will be transferred to the operating system and user-space monitoring software. Each disk drive vendor is free to decide which parameters to include for monitoring, and what their threshold should be. Unification is at the protocol level with the host.
Compaq filed IntelliSafe to the Small Form Factor (SFF) committee for standardization in early 1995. This was supported by IBM, by Compaq Seagate, Quantum, and Conner, and Western Digital development partners, who had no predictive failure timing systems. The committee chose the IntelliSafe approach, as it provides more flexibility. Compaq put IntelliSafe into the public domain on May 12, 1995. The resulting jointly developed standard was named S.M.A.R.T..
The SFF standard describes the communication protocols for ATA hosts to use and control monitoring and analysis on the hard disk drive, but does not specify specific metrics or analysis methods. Later, "S.M.A.R.T." finally understood (albeit without a formal specification) to refer to specific metrics and methods and to apply to protocols unrelated to ATA to communicate the same things.
Information provided
Technical documentation for S.M.A.R.T. in standard ATA Attachment (ATA). First introduced in 2004, it has undergone a regular revision, the latest in 2011.
The most basic information that S.M.A.R.T. providing is S.M.A.R.T. status. It gives only two values: "threshold not exceeded" and "threshold exceeded". Often these are represented as "OK drives" or "failed drives" respectively. The "threshold exceeded" value is intended to indicate that there is a relatively high probability that the drive will not be able to meet its specifications in the future: that is, the drive "will fail". Predictable failures can be catastrophic or perhaps something subtle such as the inability to write to specific sectors, or performance that may be slower than that stated by the manufacturer.
The S.M.A.R.T. status does not necessarily indicate the reliability of past or past drives. If the drive has failed synchronously, S.M.A.R.T. status may be inaccessible. Or, if the drive encountered a problem in the past, but the sensor no longer detects such problems, S.M.A.R.T. the status may, depending on the manufacturer's program, indicates that the drive now sounds.
The inability to read some sectors is not always an indication that the drive will fail. One way that unreadable sectors can be created, even when the drive works within the specification, is through a sudden power failure while the drive is writing. Also, even if the physical disk is damaged in one location, so that certain sectors can not be read, the disk may be able to use the backup space to replace the bad areas, so that the sector can be overwritten.
More details about drive health can be obtained by checking S.M.A.R.T. Attribute. SMART. Attributes are included in some ATA standard drafts, but are removed before the standard becomes final. The meaning and interpretation of attributes varies between manufacturers, and is sometimes considered a trade secret to one producer or another. Attributes are discussed further below.
Drive with S.M.A.R.T. can optionally retain a number of 'logs'. The log error records information about the latest error reported back to the host computer. Checking this log can help someone to determine whether the computer problem is related to disk or caused by anything else (timestamp error stamp may "wrap" after 2 32 ms = 49,71 days)
The impetus that implements S.M.A.R.T. can optionally apply a number of self-test or maintenance routines, and test results are stored in self-test log . The self-test routine can be used to detect unreadable sectors on the disk, so that it can be recovered from a backup source (for example, from another disk in RAID). This helps reduce the risk of permanent data loss.
Standard and implementation
Lack of general interpretation
Many motherboards display a warning message when the disk drive is nearing failure. Although industry standards exist among most hard drive manufacturers, the problem remains because attributes are deliberately undocumented to the public to distinguish between models between manufacturers.
From a legal perspective, the term "S.M.A.R.T." simply refers to the method of signaling between the internal disk drive electromechanical sensor and the host computer. Because this is the specification of S.M.A.R.T. fully vendor-specific and, while many of these attributes have been standardized between vendor drives, others remain vendor-specific. SMART. implementations are still different and in some cases may not have "general" or expected features like temperature sensors or just include some select attributes while still allowing the manufacturer to advertise the product as "S.M.A.R.T. compatible."
Visibility to the host system
Depending on the type of interface used, some motherboards are enabled by S.M.A.R.T.T. and related software may not communicate with the S.M.A.R.T hard disk. certain. For example, some external drives connected via USB and Firewire correctly send S.M.A.R.T. data through the interface. With so many ways to connect hard drives (SCSI, Fiber Channel, ATA, SATA, SAS, SSA, and so on), it's hard to predict whether S.M.A.R.T. reports will work correctly within the given system.
Even with hard drives and interfaces that implements specifications, the computer's operating system may not see S.M.A.R.T. information because drives and interfaces are summarized in lower layers. For example, they may be part of a RAID subsystem in which RAID controllers see S.M.A.R.T. capable drives, but the primary computer only sees the logical volume generated by the RAID controller.
On the Windows platform, many programs are designed to monitor and report S.M.A.R.T. information only works under administrator account.
Access
For a list of programs that allow reading of S.M.A.R.T. Data, see Comparison of S.M.A.R.T. tool.
ATA S.M.A.R.T. attribute
Each drive manufacturer defines a set of attributes, and sets a threshold value beyond the attributes that should not pass under normal operation. Each attribute has raw value , which means it is entirely dependent on the drive manufacturer (but often corresponds to the number or physical unit, such as degrees Celsius or second), normalization value , ranging from 1 to 253 (with 1 representing the worst case and 253 representing the best) and the worst value , representing the lowest normalized value recorded. The default initial attribute value is 100 but may vary between manufacturers.
Manufacturers who have applied at least one S.M.A.R.T. attributes in various products including Samsung, Seagate, IBM (Hitachi), Fujitsu, Maxtor, Toshiba, Intel, sTec, Inc., Western Digital and ExcelStor Technology.
Known ATA S.M.A.R.T. attribute
The following chart lists some S.M.A.R.T. attributes and the distinctive meaning of the raw value. Normalized values ​​are usually mapped so that higher values ​​are better (exceptions include drive temperatures, number of head load/disassemble cycles), but higher values ​​of the higher raw attributes may be better or worse depending on attributes and manufacturers. For example, the normalized "Countdown Sector Countdown" values ​​ decreased as the reallocated sector count increased . In this case, the raw attribute value will often indicate the actual number of re-allocated sectors, even if the vendor is not at all obliged to comply with this convention.
Since manufacturers do not always agree on the exact attribute and unit measurement definitions, the following list of attributes is just a general guide.
Drive does not support all attribute codes (sometimes abbreviated as "ID", for "identifier", in table). Some special codes for certain drive types (magnetic plates, flash, SSDs). Drive can use different code for the same parameters, for example, see code 193 and 225.
Threshold Limit Beyond Condition
Threshold Exceeds Condition (TEC) is an approximate date when the hard disk statistics attribute will reach its threshold value. When Drive Health software reports "Nearest T.E.C.", it should be considered a "Failure Date". Sometimes, no date is given and the drive can be expected to work without errors.
To predict the date, the drive tracks the rate of attribute change. Note that the TEC date is approximate only; the hard drive can and fails faster or slower than the TEC date.
Self-tests
SMART. drives may offer a number of self-tests:
- Short
- Check the electrical and mechanical performance and read performance of the disk. An electrical test may include a RAM buffer test, a read/write circuit test, or a read/write head element test. Mechanical tests include search and servo on data tracks. Scans a small part of the drive surface (vendor-specific area and there is a time limit on testing). Check out the list of pending sectors that may have read errors, and usually takes less than two minutes.
- Long/extended
- Longer and more thorough version of short self-test, scans the entire disk surface indefinitely. This test usually takes several hours, depending on the read/write speed of the drive and its size.
- Submission
- Intended as a quick test to identify the damage that occurred during the carriage of the device from the drive manufacturer to the computer manufacturer. Only available on the ATA drive, and usually takes a few minutes.
- Selective
- Some drives allow selective self-examination only from partial surfaces. Self test logs for SCSI and ATA drives are slightly different. It is possible for a long exam to pass even if a short test fails.
Source of the article : Wikipedia
0 Comments