[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Terminology harmonization: a proposal



Hi Nick,

I just talked to Maurizio. He already started defining some information models. We plan to define information models for describing specific packet selection methods (n in N, systematic, hash, etc.). In addition we want to define an additional generic information model that can be used to describe future or more exotic methods. 
Furthermore, (as you suggested) we plan to to include both filtering and sampling schemes  into the new version of the draft.
So far we think that having a categorization according to the input (filtering works on packet content and sampling not) is useful but maybe we get a differnt picture during working on the information model.

Regards
Tanja and Maurizio


Nick Duffield wrote:
3D8113ED.1399D14F@research.att.com">
Tanja,

comments below:

Tanja Zseby wrote:
Hi Maurizio and Nick and Rae,

In my opinion packet selectors can be distinguised according to
- the selection function: deterministic or random.
- input parameters (per packet): packet position, arrival time, packet
content
(as Rae pointed out input parameters also could be additional
information like interfaces, or some output of other functions on the
router (router reaction), etc..)

The question now is whether and how we want to categorize these packet
selectors into sampling and filtering primitives (maybe we need a
third categorie ?) . In your emails I saw various approaches for a
categorization. But I think we need a clear reasoning and a clear
definition how filtering and sampling is distinguished. So I tried to
find out the cases in question:

a) deterministic function on packet content (hashing also would fall
in this categorie) ==> filtering
b) deterministic function on packet positio n (every k-th packet) ==> ?
(currently denoted as systematic sampling)
c) deterministic function on arrival time (all packets in given time
interval ) ==> ? (currently denoted as systematic sampling)
d) random function on packet content. ==> probably not needed ?
e) random function on packet position ==> sampling
f) random function on arrival time ==> probably not needed ?

I see two alternatives for categorization:

1. Categorization according to selection function: filtering is
deterministic, sampling is random
If we categorize like this, systematic sampling (e.g. selection of
every k-th packet) would be a filtering primitive.

2. Categorization according to input parameters (as Maurizio
proposed): filtering works on packet content, sampling not
One reason to distinguish in this way could be that (most parts of
the) packet content would be the same at different measurement points
whereas packet posi tion and arrival time of the packet vary for
different observation points. That means we could specify a filter
which can be applied to various measurement points and ensures that
the same packets are selected at all points. Nevertheless this only
works if we limit packet content to the invariant fields in the packet
(so the TTL example from Rao would not work).

An alternative would be to define further categorizes or simply dont
categorize. We could just talk about packet selectors (defined by
function and parameters) and treat sampling and filtering as specific
forms of packet selection functions.
Any preferences or opinions ?

When it comes to writing a standard, my preference is for the last
alternative: concentrating on the functionality and parameters of the
configurable selectors.

At a lower level, we want to set of the universe of selection
operations which we will draw from. Given that we're not sure
whether some things are sampling or filtering, then perhaps it
doesn't make sense to have separate documents on each. Even
if we could agree as to which is what, the categories mentioned
in 1 and 2 would probably cut across the eventual functions,
and this could lead to confusion. Examples:

. hashing is deterministic, but a strong hash function with
suitable input can be used to implement apparently random sampling.

. in scheme 1, 1 in N selection is a filter (on some counter)
but functionally it might be thought of as (an implementation of)
1 in N sampling.

. we can have a function of "sampling 1/Nth of the packets on
average" which can have different implementations, some of which
are more deterministic than others e.g. (i) sample based on hash
of packet content, (ii) sample based on deterministic function of
packet position, (iii) use random number generator seeded with
packet content; (iv) use independent random number generator. We
should lay out all these possibilities in our universe.

(BTW, do people agree that these could just be different
implementations of the same function, or, could these actually
be different functions in the PSAMP device?).

Rather than trying to solve the sampling/filtering conundrum,
maybe the thing to do is describe the universe in a more neutral
way:
. inputs
- some contents from packets
- layer 2?
- associated routing state?
- subsidiary quantities, perhaps calculated from the packet
contents, (possibly from multiple packets) e.g.
o hashes
o packet seeded random-like numbers
o counters and timers, and rules for their update
- anything else?
. selection
just filters on the inputs?

Any of our functional primitives must be describable by combining these,
although if there is more than one combination that meets a functional
requirement, then the standard would be neutral as to which one should
be
used: its an implementation choice.

Nick

Regards
Tanja

Maurizio Molina wrote:

Nick Duffield wrote:

Maurizio,
thanks for opening up the discussion on terminology.
Comments below:
Maurizio Molina wrote:

Hi,
After browsing through Nick's draft, the PSAMP mailing list,
and
Tanja's draft proposal, I see there's still the need to harmonize
some
terminology (and concepts) regarding packet selection, sampling
and
filtering.
1) From Nick's draft:
3.2 Packet selection
..........Packet selection is performed through combination a
number
of measurement primitives described below.......
- Hashing
...........
- Filtering
...........
- Sampling
...........
2) From a Nick's mail:
......With the words currently at my disposal, my usage is:
1. sampling = 1 in N (periodic or statistical) or hash-based
2. filtering = filtering
3. (primitive) selectors = either 1 or 2, and further methods TBD
4. (composite) selectors = composites of methods from 3
3) Tanja's draft: it focuses on samping, but it mentions some
methods
(e.g. stratified sampling, or sampling dependent on the packet
content)

that im my view are already a combination of filtering +
sampling. This
point is clarified later.
My proposal for harmonizing terminology/concepts would be the
following:
a) "primitives" for packet selection are only sampling and
filtering.
Composite packet selection methodologies can then be built by a
combination of the two.
b) sampling is always "blind" to packet content. A packet is
sampled out
of a stream only dependening on the packet position (which can be
spatial or temporal) and/or on the result of a sampling algorithm
(which
can be deterministic or probabilistic).

I think it will be too limiting to say that type (a) sampling must
be blind to packet content. Implementations may want to use
content
from the packet stream as a cheap way to insert or seed randomness
into
sampling decisions. (In a different arena, this approach has been
proposed for importance sampling of flow statistics)

Nick,
Clearly distinguishing between filtering (based ONLY on packet
content) and
sampling (NOT based on packet content) is aimed at finding a way to
formally
describe any packet selection mechanism, i.e. an information model
that could
be used to configure packet selectors in a standard way.
Most of the packet selection examples described in Tanja's draft
could be in
fact easily described as a combination of the two. E.g., sampling
dependent
on the packet length -> first describe a filter that create
different
substreams on the basis of packet length, then describe different
samplers one
for each substream with different sampling frequencies. This clear
distinction
applies to the formal definition only. The way it is implemented can
then be a
real "hybrid", as you mention.
However, at least a packet selection method is actually quite
difficult to
express as a combination of the two: namely, what is
described in Tanja's doc
(4.2.3 - packet content based trigger) that is (I guess) also what
you refer
to when you say "....Implementations may want to use content from
the packet
stream as a cheap way to insert or seed randomness into sampling
decisions.....".
Triggering a sampling procedure on the basis of the packet content,
and then
sampling according to some other method, can be viewed as something
really
"hybrid". In the formalization attempt I'm currently working on,
however, I'm
trying to still represented it by a composition of the two basic
selectors
(sampling & filtering).

c) filtering is on the contrary "blind" to the packet position in
the
stream, but it is based on the packet properties. A packet
property may
be simply its content, or the content of a set of subfields, or
the
result of a function taking as an input (part of) the packet
content.

I agree that at a high level one can regard hash-based selection
as a
type
of filtering, since it relies on a complex but deterministic
function of
the packet contents. Perhaps it is worth noting that any good
hash-based
selection function would be infeasible to express as a composition
of
match/mask filters; I've been asked this.
But I'm concerned that the (a) /(b) division may close the door to
future
selectors that people may invent, or implementations (e.g. of
sampling)
that are really a hybrid of the two approaches.

I agree. We should avoid to say that EVERY current and future
selection
methodology can be expressed by these two building blocks.

And why stop there? At an abstract level one could regard all the
selectors as (i) calculating some quantity depending on the packet
content and/or other variables; then (ii) selecting packet if the
quantity falls in a given range. A danger with trying to couch the
framework this way is of losing focus of the basic functionality
that we want (e.g. sample 1 in N packets, somehow) and dwelling on
implementations (e.g. different ways of getting 1 in N sampling in
this
framework, e.g. decrement a counter, or calculate a hash, or use a
well-known random number generator, or seed a counter with the
packet
stream).

I'm not sure I fully understood this  point. If the only sampling
type to be
supported by a standard is 1 in N sampling (how it is done being
implementation specific) then there's not much to do in PSAMP. But I
guess the
scope of PSAMP is exactly creating a set of commonly agreed standard
procedures so that users of sampled data are more confident of what
they're
receiving. Also, as you explained in your papers, applications like
trajectory
sampling need a common hash function on all the crossed nodes....
Of course, not all the bits of an implementation need to be
standardized, but
I guess that defining the border between what needs to be
standardized and
what not is exactly PSAMP's job, isn't it?
Maurizio

Some notes:
n1) filtering is always deterministic.
n2) hashing is a sub-case of filtering.
n3) how "complex" a composite selector can be still needs to be
discussed, but by sure the methodologies Tanja mentioned
(stratified
sampling, or sampling dependent on the packet content) can be
implemented by a cascaded filter->sampler.
n4) Another example of a filtering function could be taking the
source
and/or destination address, lookup the source/destination AS and
filter
on the basis of the result. While such a complex filtering
function
doesn't make sense at the line rate, it may make sense if a
sampler is
placed in front of a filter to reduce the rate of packets to be
processed. In this respect, the text appearing in Nick's draft at
the
bottom of 3.2 reported below (unavailability of router state to
measurement primitives) should be reconsidered.
"In order to be able to fun ction at line rates,
each measurement
primitive take as its input only a packet itself, or
quantities
that have been calculated from the packet previously by other
measurement primitives. Router state is not assumed to be
available
to the measurement primitives."

Yes, I have been wondering whether this should be reworked after
Peram
brought up this point a little while ago.
The reason for excluding router state from the primitive
operations
was arhictectural: we didn't want to assume that the routing state
would
be available to a filter that could be required to operate at line
rate.
(Any comments on this from implementors?)
But we do assume that routing state, if present in the measuring
network
element, will be available to form the packet reports, so it
should be
feasible to do filtering based on routing state when reports
are formed i.e. after all the other selection primitives have
operated.
Nick

Maurizio

--
to unsubscribe send a message to psamp-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/psamp/>

--
Dipl.-Ing. Tanja Zseby
FhI FOKUS/Global Networking Email: zseby@fokus.fhg.de
Kaiserin-Augusta-Allee 31 Phone: +49-30-3463-7153
D-10589 Berlin, Germany Fax: +49-30-3463-8153
--------------------------------------------------------------------------------------
"Living on earth is expensive but it includes a free trip around the sun." (Anonymous)
--------------------------------------------------------------------------------------

--
to unsubscribe send a message to psamp-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/psamp/>

-- 
Dipl.-Ing. Tanja Zseby			    	      	
FhI FOKUS/Global Networking			Email: zseby@fokus.fhg.de	
Kaiserin-Augusta-Allee 31				Phone: +49-30-3463-7153
D-10589 Berlin, Germany				Fax:   +49-30-3463-8153
-------------------------------------------------------------------------------------- 
"Living on earth is expensive but it includes a free trip around the sun." (Anonymous)
--------------------------------------------------------------------------------------