Hi,
What is the difference between CRC,checksum and Frame check sequence.

I m aware that checksum is used as part of tcp/udp/ip headers.
At which layer CRC and FCS used.

I think purpose all three is Data integrity.Why three different
things.Any specific reason?

Thankx,
Saranjothy

Each layer can scramble data in different ways.

Broadly, the idea is a CRC is a more comprehensive check, intended to catch
a wide range of errors likely to occur over an analog transmission (e.g.
range of bit errors) and a checksum is designed to catch typical errors
made in software implementations.

That's the idea. Reality is somewhat different and I recommend reading
Jonathan Stone's SIGCOMM paper and Stanford doctoral dissertation (warning:
I was Jonathan's advisor and co-author on the SIGCOMM paper).

Craig

SaranJothy <saranjothy@gmail.com> writes:

>Hi,
>What is the difference between CRC,checksum and Frame check sequence.

>I m aware that checksum is used as part of tcp/udp/ip headers.
>At which layer CRC and FCS used.

>I think purpose all three is Data integrity.Why three different
>things.Any specific reason?

>Thankx,
>Saranjothy

On 30 Apr 2007, SaranJothy <saranjothy@gmail.com> wrote:

> What is the difference between CRC,checksum and Frame check sequence.

Checksum is the general term. CRC is one kind of checksum. An FCS is
a checksum (typically a CRC) appended to a frame during serial data
transmission, so an FCS is one use of a praticular type of checksum.

> I m aware that checksum is used as part of tcp/udp/ip headers.
> At which layer CRC and FCS used.
>
> I think purpose all three is Data integrity.Why three different
> things.Any specific reason?

Some are better than others in detecting specific kinds of errors.
Some are more easily computed in software or hardware. Since there are
a variety of different ways checksums are used, there is a variety of
different checksums.

Dave

--
D.a.v.i.d T.i.k.t.i.n
t.i.k.t.i.n [at] a.d.v.a.n.c.e.d.r.e.l.a.y [dot] c.o.m

While we're on the subject, what do you think about the idea of
Layer-3 check-summing. If you could redo UDP/TCP checksums, how would
you do it?

-Le Chaud Lapin-

On Apr 30, 12:24 pm, Craig Partridge <cra...@std.com> wrote:
> Each layer can scramble data in different ways.
>
> Broadly, the idea is a CRC is a more comprehensive check, intended to catch
> a wide range of errors likely to occur over an analog transmission (e.g.
> range of bit errors) and a checksum is designed to catch typical errors
> made in software implementations.
>
> That's the idea. Reality is somewhat different and I recommend reading
> Jonathan Stone's SIGCOMM paper and Stanford doctoral dissertation (warning:
> I was Jonathan's advisor and co-author on the SIGCOMM paper).
>
> Craig
>
> SaranJothy <saranjo...@gmail.com> writes:
> >Hi,
> >What is the difference between CRC,checksum and Frame check sequence.
> >I m aware that checksum is used as part of tcp/udp/ip headers.
> >At which layer CRC and FCS used.
> >I think purpose all three is Data integrity.Why three different
> >things.Any specific reason?
> >Thankx,
> >Saranjothy

In article <1177983225.756330.280380@h2g2000hsg.googlegroups. com>,
Le Chaud Lapin <jaibuduvin@gmail.com> wrote:

> While we're on the subject, what do you think about the idea of
> Layer-3 check-summing. If you could redo UDP/TCP checksums, how would
> you do it?

Modern computers are probably fast enough that a more powerful checksum
could probably be used without significant performance impact.

--
Barry Margolin, barmar@alum.mit.edu
Arlington, MA
*** PLEASE post questions in newsgroups, not directly to me ***
*** PLEASE don't copy me on replies, I'll read them in the group ***

On Apr 30, 8:37 pm, Barry Margolin <bar...@alum.mit.edu> wrote:
> In article <1177983225.756330.280...@h2g2000hsg.googlegroups. com>,
> Le Chaud Lapin <jaibudu...@gmail.com> wrote:
>
> > While we're on the subject, what do you think about the idea of
> > Layer-3 check-summing. If you could redo UDP/TCP checksums, how would
> > you do it?
>
> Modern computers are probably fast enough that a more powerful checksum
> could probably be used without significant performance impact.

You mean like a CRC instead of the current techniques?

-Le Chaud Lapin-