Page 1 of 6
Journal for Studies in Management and Planning
Available at http://edupediapublications.org/journals/index.php/JSMaP/
e-ISSN: 2395-0463
Volume 02 Issue 11
November 2016
Available online:http://edupediapublications.org/journals/index.php/JSMaP/ P a g e | 153
Coding Script to Avoid Data Corruption To Protect
Memories
1Kukkala Venkatesh, 2Mr.P.V.Vara Prasad Rao
1
PG Scholar, Department of ECE, SLC's Institute of Engineering and Technology, Piglipur Village, Hayathnagar
Mandal, Near Ramoji Film City, Ranga Reddy District, Hyderabad, Telangana
2Assosciate Professor, Department of ECE, SLC's Institute of Engineering and Technology, Piglipur Village,
Hayathnagar Mandal, Near Ramoji Film City, Ranga Reddy District, Hyderabad, Telangana
ABSTRACT:
The primary challenge is the fact that individual’s codes should minimize the delay and area
penalty. One of the codes which have been considered for memory protection is Reed-Solomon
(RS) codes. This limits using ECCs in high-speed recollections. It has brought to using simple
codes for example single error correction double error recognition (SEC-DED) codes. However,
as technology scales multiple cell upsets (MCUs) be common and limit using SEC-DED codes
unless of course they're coupled with interleaving. To prevent data corruption, error correction
codes (ECCs) are broadly accustomed to safeguard recollections. ECCs introduce a delay penalty
in being able to access the information as encoding or deciphering needs to be carried out. An
identical issue happens in some kinds of recollections like DRAM which are typically arranged
in modules made up of several products. In individual’s modules, the security against a tool
failure instead of isolated bit errors can also be desirable. In individual’s cases, one option is by
using more complex ECCs that may correct multiple bit errors. These codes derive from non- binary symbols and for that reason can correct multiple bit errors. Within this paper, single
symbol error correction codes according to Reed-Solomon codes that may be implemented with
low delay are suggested and evaluated. The outcomes show that they'll be implemented having a
substantially lower delay than traditional single error correction RS codes.
Keywords: Error correction codes, reed-solomon codes, DRAM memorymodule, soft errors.
I. INTRODUCTION
ECCs then adds additional parity check bits
to every memory word so that errors could
be detected and remedied. These extra bits
lessen the effective capacity from the
memory. Errors could be caused for instance
Page 2 of 6
Journal for Studies in Management and Planning
Available at http://edupediapublications.org/journals/index.php/JSMaP/
e-ISSN: 2395-0463
Volume 02 Issue 11
November 2016
Available online:http://edupediapublications.org/journals/index.php/JSMaP/ P a g e | 154
by radiation caused soft errors affecting a
number of memory cells and alter their
values. Data corruption brought on by errors
is really a serious problem in recollections
other kinds of failures cause permanent
damage so that the unit no more provides
correct data. To make sure that information
is not corrupted when failures occur, error
correction codes (ECCs) are broadly utilized
in recollections. Other overheads created by
the ECC would be the encoding and
deciphering circuitry. This circuitry comes
with an impact also around the delay
because the data needs to be encoded when
writing in to the memory and decoded when
studying from this. When multiple errors
affect cells which are physically close, out
of the box the situation of radiation caused
multiple cell upsets (MCUs), SEC-DED
codes could be coupled with interleaving to
make sure that the errors affect only one bit
per logical word. Generally, the deciphering
is much more complex compared to
encoding and limits the rate from the ECC.
Typically single error correction double
error recognition (SEC-DED) codes are
utilized to safeguard recollections. These
codes possess a minimum Hamming
distance of 4 so that single bit errors could
be remedied while double errors are detected
and never miscorrected. SEC-DED codes
could be implemented having a relatively
low area and delay overhead and a few
optimizations happen to be suggested
recently. Multiple bit errors are a problem
when SEC-DED codes are utilized. That's
even the situation when a mistake causes the
malfunction of the device inside a memory
module. For the reason that situation, the
term is split in sub-blocks as well as an ECC
can be used for all of them [1]. Then your
sub-blocks are interleaved within the
products so that inside a device there's only
one small given sub-block. However, using
interleaving comes with an effect on the
memory design and may increase area and
power. For any memory module, using
interleaving increases the amount of parity
check bits needed, to supplement bits are
needed per each one of the sub-blocks.
Finally, when multiple errors come from
independent error occasions, more effective
ECCs are necessary to make sure the
correction of errors. A large quantity of
multiple bit ECCs happen to be suggested to
safeguard recollections. Included in this are
Bose-Chaudhuri-Hocquenghem (BCH),
Euclidean Geometry, Difference Set,
Orthogonal Latin Squares and Reed- Solomon codes. Reed Solomon (RS) codes
Page 3 of 6
Journal for Studies in Management and Planning
Available at http://edupediapublications.org/journals/index.php/JSMaP/
e-ISSN: 2395-0463
Volume 02 Issue 11
November 2016
Available online:http://edupediapublications.org/journals/index.php/JSMaP/ P a g e | 155
possess a distinct feature when in
comparison using the other codes: they
aren't binary. They will use symbols from
the Galois Field so that each symbol is
symbolized by multiple bits. Therefore a
SEC RS code can correct multiple bit errors
as lengthy because they affect just one
symbol. This is extremely attractive for
memory modules as when the amount of bits
within the products match’s individuals
from the symbols within the RS code,
failures in a single device could be
remedied. Actually, because of this RS
codes are generally accustomed to safeguard
primary recollections in personal computers
for space programs, for example individuals
described [2]. Generally, the information
that forms an RS code wordis thought as
polynomial coefficients with values of the
Galois Field. The polynomial akin to a code
word is really a multiple of the specific
polynomial, known as generator polynomial
g(x). Interested visitors can make reference
to a textbook on error control codes for more
particulars. Within the situation of SEC RS
codes, the code word consists by appending
two check symbols to some data word of k
symbols. To define the SEC RS codes,
normally the matrix representation is
preferred, thus the symbols creating the code
word are vectors of Galois Field elements.
For binary codes, a code is determined
utilizing a matrix H, known as parity check
matrix. A code word is really a vector v. The
encoding procedure for an information
vector d is carried out beginning from the
generator matrix G, by computing Gd,
which helps to ensure that Hv. Because the
SEC RS code is generally separable, the G
matrix assumes the shape G, where Ik is
really an identity matrix and P is of
dimension k x 2. Just like other advanced
ECCs one problem for using RS codes in
recollections may be the delay created by
the deciphering. Within the situation of RS
codes, several arithmetic procedures within
the Galois Field are necessary to encode or
decode a block. This produces a much
bigger delay compared to traditional SEC- DED codes. To mitigate the outcome on
delay when utilizing advanced ECCs, one
choice is to do error recognition first and
just when errors are detected go to the
correction phase. As errors are rare, the
typical delay will bond with those of the
mistake free situation that is given when
needed to do error recognition only. This
really is reduced compared to time required
for correction. However, despite this
modification delay could be large because
