__NOTOC__
Chipkill is
IBM
International Business Machines Corporation (using the trademark IBM), nicknamed Big Blue, is an American Multinational corporation, multinational technology company headquartered in Armonk, New York, and present in over 175 countries. It is ...
's trademark for a form of advanced
error checking and correcting (ECC)
computer memory
Computer memory stores information, such as data and programs, for immediate use in the computer. The term ''memory'' is often synonymous with the terms ''RAM,'' ''main memory,'' or ''primary storage.'' Archaic synonyms for main memory include ...
technology that protects memory systems from single memory chip failures and multi-bit errors from any portion of a single memory chip. One simple scheme to perform this function scatters the bits of a
Hamming code
In computer science and telecommunications, Hamming codes are a family of linear error-correcting codes. Hamming codes can detect one-bit and two-bit errors, or correct one-bit errors without detection of uncorrected errors. By contrast, the ...
ECC word across multiple memory chips, such that the failure of any single memory chip will affect only one ECC bit per word. This allows memory contents to be reconstructed despite the complete failure of one chip. Typical implementations use more advanced codes, such as a
BCH code
In coding theory, the Bose–Chaudhuri–Hocquenghem codes (BCH codes) form a class of cyclic error-correcting codes that are constructed using polynomials over a finite field (also called a '' Galois field''). BCH codes were invented in ...
, that can correct multiple bits with less overhead.
Chipkill is frequently combined with
dynamic bit-steering, so that if a chip fails (or has exceeded a threshold of bit errors), another, spare, memory chip is used to replace the failed chip. The concept is similar to that of
RAID
RAID (; redundant array of inexpensive disks or redundant array of independent disks) is a data storage virtualization technology that combines multiple physical Computer data storage, data storage components into one or more logical units for th ...
, which protects against disk failure, except that now the concept is applied to individual memory chips. The technology was developed by the
IBM Corporation
International Business Machines Corporation (using the trademark IBM), nicknamed Big Blue, is an American multinational technology company headquartered in Armonk, New York, and present in over 175 countries. It is a publicly traded company ...
in the early and mid-1990s. An important
RAS feature, Chipkill technology is deployed primarily on
SSDs
The Ship Self-Defense System (SSDS) is a combat system specifically designed for anti-air defense of aircraft carriers, and most other non-Aegis United States Navy combat ships. It coordinates several existing shipboard systems. Multi-sensor i ...
,
mainframes
A mainframe computer, informally called a mainframe or big iron, is a computer used primarily by large organizations for critical applications like bulk data processing for tasks such as censuses, industry and consumer statistics, enterprise ...
, and midrange servers.
An equivalent system from
Sun Microsystems
Sun Microsystems, Inc., often known as Sun for short, was an American technology company that existed from 1982 to 2010 which developed and sold computers, computer components, software, and information technology services. Sun contributed sig ...
is called ''Extended ECC'', while equivalent systems from
HP are called ''Advanced ECC'' and ''Chipspare''. A similar system from Intel, called ''
Lockstep memory
Lockstep systems are fault-tolerant computer systems that run the same set of operations at the same time in parallel. The redundancy (duplication) allows error detection and error correction: the output from lockstep operations can be compared ...
'', provides
double-device data correction
Lockstep systems are fault-tolerant computer systems that run the same set of operations at the same time in parallel. The redundancy (duplication) allows error detection and error correction: the output from lockstep operations can be compared ...
(DDDC) functionality. Similar systems from
Micron
The micrometre (English in the Commonwealth of Nations, Commonwealth English as used by the International Bureau of Weights and Measures; SI symbol: μm) or micrometer (American English), also commonly known by the non-SI term micron, is a uni ...
, called ''redundant array of independent NAND'' (RAIN), and from
SandForce
SandForce was an American fabless semiconductor company based in Milpitas, California, that designed flash memory controllers for solid-state drives (SSDs). On January 4, 2012, SandForce was acquired by LSI Corporation and became the Flash Compo ...
, called ''RAISE level 2'', protect data stored on SSDs from any single NAND flash chip failure.
A 2009 paper using data from Google's data centers provided evidence demonstrating that in observed Google systems, DRAM errors were recurrent at the same location, and that 8% of DIMMs were affected each year. Specifically, "In more than 85% of the cases a correctable error is followed by at least one more correctable error in the same month." DIMMs with Chipkill error correction showed a lower fraction of DIMMs reporting uncorrectable errors compared to DIMMs with error-correcting codes that can only correct single-bit errors. A 2010 paper from the
University of Rochester
The University of Rochester is a private university, private research university in Rochester, New York, United States. It was founded in 1850 and moved into its current campus, next to the Genesee River in 1930. With approximately 30,000 full ...
also showed that Chipkill memory resulted in substantially fewer memory errors, using both real-world memory traces and simulations.
See also
*
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.
Typically, ECC memory maintains a memory system immun ...
*
Lockstep memory
Lockstep systems are fault-tolerant computer systems that run the same set of operations at the same time in parallel. The redundancy (duplication) allows error detection and error correction: the output from lockstep operations can be compared ...
*
Memory ProteXion
For computer memory, Memory ProteXion, found in IBM xSeries servers, is a form of " redundant bit steering". This technology uses redundant bits in a data packet to recover from a DIMM failure.
Memory ProteXion is different from normal ECC error c ...
*
Redundant array of independent memory
A redundant array of independent memory (RAIM) is a design feature found in certain computers' main random access memory. RAIM utilizes additional memory modules and striping algorithms to protect against the failure of any particular module and ...
*
Single-error correction and double-error detection
In computer science and telecommunications, Hamming codes are a family of linear error-correcting codes. Hamming codes can detect one-bit and two-bit errors, or correct one-bit errors without detection of uncorrected errors. By contrast, the ...
(SECDED)
References
{{Reflist, 30em
External links
Intel E7500 Chipset MCH Intelx4 Single Device Data Correction (x4 SDDC) Implementation and Validation Intel Application note AP-726, August 2002.
DRAM study turns assumptions about errors upside down Ars Technica
''Ars Technica'' is a website covering news and opinions in technology, science, politics, and society, created by Ken Fisher and Jon Stokes in 1998. It publishes news, reviews, and guides on issues such as computer hardware and software, sci ...
, October 7, 2009
Enabling Memory Reliability, Availability, and Serviceability Features on Dell PowerEdge Servers 2005
Chipkill correct memory architecture August 2000, by David Locklear
The Mathematics of Chipkill ECC October 2015, by Bob Day
Computer memory
Error detection and correction
IBM computer hardware