ECC memory

Error correction code memory (ECC memory) is a type of computer data storage that uses an error correction code (ECC) to detect and correct n-bit data corruption which occurs in memory. ECC memory is used in most computers where data corruption cannot be tolerated, like industrial control applications, critical databases, and infrastructural memory caches.

Typically, ECC memory maintains a memory system immune to single-bit errors: the data that is read from each word is always the same as the data that had been written to it, even if one of the bits actually stored has been flipped to the wrong state. Most non-ECC memory cannot detect errors, although some non-ECC memory with parity support allows detection but not correction.

Description

Error correction codes protect against undetected data corruption and are used in computers where such corruption is unacceptable, examples being scientific and financial computing applications, or in database and file servers. ECC can also reduce the number of crashes in multi-user server applications and maximum-availability systems.

Electrical or magnetic interference inside a computer system can cause a single bit of dynamic random-access memory (DRAM) to spontaneously flip to the opposite state. It was initially thought that this was mainly due to alpha particles emitted by contaminants in chip packaging material, but research has shown that the majority of one-off soft errors in DRAM chips occur as a result of background radiation, chiefly neutrons from cosmic ray secondaries, which may change the contents of one or more memory cells or interfere with the circuitry used to read or write to them. Hence, the error rates increase rapidly with rising altitude; for example, compared to sea level, the rate of neutron flux is 3.5 times higher at 1.5 km and 300 times higher at 10-12 km (the cruising altitude of commercial airplanes).A Survey of Techniques for Modeling and Improving Reliability of Computing Systems
, IEEE TPDS, 2015 As a result, systems operating at high altitudes require special provisions for reliability.

As an example, the spacecraft ''Cassini–Huygens'', launched in 1997, contained two identical flight recorders, each with 2.5 gigabits of memory in the form of arrays of commercial DRAM chips. Due to built-in EDAC functionality, the spacecraft's engineering telemetry reported the number of (correctable) single-bit-per-word errors and (uncorrectable) double-bit-per-word errors. During the first 2.5 years of flight, the spacecraft reported a nearly constant single-bit error rate of about 280 errors per day. However, on November 6, 1997, during the first month in space, the number of errors increased by more than a factor of four on that single day. This was attributed to a solar particle event that had been detected by the satellite GOES 9.

There was some concern that as DRAM density increases further, and thus the components on chips get smaller, while operating voltages continue to fall, DRAM chips will be affected by such radiation more frequently, since lower-energy particles will be able to change a memory cell's state. On the other hand, smaller cells make smaller targets, and moves to technologies such as SOI may make individual cells less susceptible and so counteract, or even reverse, this trend. Recent studies show that single-event upsets due to cosmic radiation have been dropping dramatically with process geometry and previous concerns over increasing bit cell error rates are unfounded.

Research

Work published between 2007 and 2009 showed widely varying error rates with over 7 orders of magnitude difference, ranging from 10⁻¹⁰ error/bit·h (roughly one bit error per hour per gigabyte of memory) to 10⁻¹⁷ error/bit·h (roughly one bit error per millennium per gigabyte of memory). A large-scale study based on Google's very large number of servers was presented at the SIGMETRICS/Performance '09 conference. The actual error rate found was several orders of magnitude higher than the previous small-scale or laboratory studies, with between 25,000 (2.5 × 10⁻¹¹ error/bit·h) and 70,000 (7.0 × 10⁻¹¹ error/bit·h, or 1 bit error per gigabyte of RAM per 1.8 hours) errors per billion device hours per megabit. More than 8% of DIMM memory modules were affected by errors per year.

The consequence of a memory error is system-dependent. In systems without ECC, an error can lead either to a crash or to corruption of data; in large-scale production sites, memory errors are one of the most-common hardware causes of machine crashes. Memory errors can cause security vulnerabilities. A memory error can have no consequences if it changes a bit which neither causes observable malfunctioning nor affects data used in calculations or saved. A 2010 simulation study showed that, for a web browser, only a small fraction of memory errors caused data corruption, although, as many memory errors are intermittent and correlated, the effects of memory errors were greater than would be expected for independent soft errors.

Some tests conclude that the isolation of DRAM memory cells can be circumvented by unintended side effects of specially crafted accesses to adjacent cells. Thus, accessing data stored in DRAM causes memory cells to leak their charges and interact electrically, as a result of high cell density in modern memory, altering the content of nearby memory rows that actually were not addressed in the original memory access. This effect is known as row hammer, and it has also been used in some privilege escalation computer security exploits.

An example of a single-bit error that would be ignored by a system with no error-checking, would halt a machine with parity checking, or would be invisibly corrected by ECC: a single bit is stuck at 1 due to a faulty chip, or becomes changed to 1 due to background or cosmic radiation; a spreadsheet storing numbers in ASCII format is loaded, and the character "8" (decimal value 56 in the ASCII encoding) is stored in the byte that contains the stuck bit at its lowest bit position; then, a change is made to the spreadsheet and it is saved. As a result, the "8" (0011 1000 binary) has silently become a "9" (0011 1001).

Solutions

Several approaches have been developed to deal with unwanted bit-flips, including immunity-aware programming, RAM parity memory, and ECC memory.

This problem can be mitigated by using DRAM modules that include extra memory bits and memory controllers that exploit these bits. These extra bits are used to record parity or to use an error-correcting code (ECC). Parity allows the detection of all single-bit errors (actually, any odd number of wrong bits). The most-common error correcting code, a single-error correction and double-error detection (SECDED) Hamming code, allows a single-bit error to be corrected and (in the usual configuration, with an extra parity bit) double-bit errors to be detected. Chipkill ECC is a more effective version that also corrects for multiple bit errors, including the loss of an entire memory chip.

Implementations

Seymour Cray