[Mep-dev] TCP Performance Issues [oops]

Michelle w5nyv@yahoo.com
Sun, 30 Nov 2008 07:24:19 -0800 (PST)


--0-1237434424-1228058659=:567
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

I'm very much=A0looking forward to the rest of it. =0A=0ADo you (all)=A0thi=
nk that the the lack of pressure to make the service pay for itself (as opp=
osed to=A0possibly trying=A0to make the hardware come in at cost)=A0might m=
ake any difference to the design choices of the protocols and the link?=0A=
=0AMost of the=A0examples that I am familiar with spend a lot of time on ac=
counting. Most of=A0the satellite models=A0and most cellular models have to=
 keep track=A0of users and=A0how long and how much they use the link, in or=
der to charge for it and=A0make=A0money.=A0=0A=0AMy first=A0inclination is =
that the deletion of all the accounting overhead=A0would=A0do two things - =
it would require some redesign of the link, and it would=A0possibly increas=
e performance due to=A0a lessening of complexity.=0A=0AHowever, it was poin=
ted out to me=A0that relieving the financial accounting side of things does=
n't mean that those same functions are not useful and needed on MEP. So, my=
 second inclination is=A0perhaps the accounting=A0functions that are common=
ly=A0seen in satellite and cellular links=A0may=A0end up being required on =
MEP, but implemented for different=A0purposes.=A0Searching, rooms, filters,=
 friends lists, modeling,=A0quality of service, and other fun things that a=
re=A0enabled by a rich set of metadata certainly start to sound a lot like =
"accounting".=A0=0A=0AThe level of accountability (authentication and secur=
ity)=A0needs to be explored and discussed. How concerned should we be with =
harmful or unauthorized use? =0A=0ATraditionally, amateur radio is an open =
system that runs on the honor system and is self-policed. I would like to=
=A0remain true to the spirit of amateur radio whenever and wherever that sp=
irit is positive and worth preserving.=A0When considering a digital system =
compared to an analog system,=A0the issues of identification, authenticatio=
n, accounting, and security=A0naturally arise. A model to=A0consider is D-S=
tar, which has=A0repeater registration by call sign. This is a user-entry I=
D. So, to me, it is the equivalent of stating your call sign on the=A0air a=
nd remains=A0a voluntary, honor-system act. =0A=0AI=A0strongly agree with t=
he goal of making MEP a fully functional IP=A0network. That is the baseline=
 assumption, and should be=A0explored, challenged, and talked about until w=
e are=A0more than fully convinced it's the right=A0thing to do, where "it" =
is well-defined. This leads directly to identifying the requirements for fu=
lly supporting IP. =0A=0AWhat are the repercussions of amateur radio commun=
icating over a fully=A0functional IP network? What possibly unexpected requ=
irements do we incur upon ourselves by doing so? Are authentication and sec=
urity and identification controversial issues, or are they now so=A0embedde=
d in the communications ecosystem that to remove these functions would be n=
egative? Is the transport protocol truly unaffected by the type of traffic,=
 which is a hallmark axiom of layered protocols? In other words, is there a=
nything in particular about amateur radio that would cause a change in the=
=A0requirements for IP? =0A=0AMore soon,=A0-Michelle W5NYV =0A=0A=A0=0A=0A=
=0A=0A________________________________=0AFrom: Timothy J. Salo <salo@saloit=
s.com>=0ATo: mep <mep-dev@uppermeadow.com>=0ASent: Saturday, November 29, 2=
008 5:51:24 PM=0ASubject: Re: [Mep-dev] TCP Performance Issues [oops]=0A=0A=
> Now that I think about it some more, this topic is relevant=0A> to the ME=
P.=A0 First, I will try to outline the broader=0A> controversy and describe=
 three contrasting views.=A0 Then,=0A> I will show one area where the MEP p=
roject will probably=0A> need to, in some sense, pick sides.=0A=0AOops.=A0 =
The e-mail escaped before its time.=A0 I will resend it=0Awhen it is comple=
te.=0A=0A-tjs=0A=0A_______________________________________________=0AMep-de=
v mailing list=0AMep-dev@uppermeadow.com=0Ahttp://lists.uppermeadow.com/mai=
lman/listinfo/mep-dev=0A
--0-1237434424-1228058659=:567
Content-Type: text/html; charset=us-ascii

<html><head><style type="text/css"><!-- DIV {margin:0px;} --></style></head><body><div style="font-family:times new roman, new york, times, serif;font-size:12pt"><DIV></DIV>
<DIV>I'm very much&nbsp;looking forward to the rest of it. </DIV>
<DIV>&nbsp;</DIV>
<DIV>Do you (all)&nbsp;think that the the lack of pressure to make the service pay for itself (as opposed to&nbsp;possibly trying&nbsp;to make the hardware come in at cost)&nbsp;might make any difference to the design choices of the protocols and the link?</DIV>
<DIV>&nbsp;</DIV>
<DIV>Most of the&nbsp;examples that I am familiar with spend a lot of time on accounting. Most of&nbsp;the satellite models&nbsp;and most cellular models have to keep track&nbsp;of users and&nbsp;how long and how much they use the link, in order to charge for it and&nbsp;make&nbsp;money.&nbsp;</DIV>
<DIV>&nbsp;</DIV>
<DIV>My first&nbsp;inclination is that the deletion of all the accounting overhead&nbsp;would&nbsp;do two things - it would require some redesign of the link, and it would&nbsp;possibly increase performance due to&nbsp;a lessening of complexity.</DIV>
<DIV>&nbsp;</DIV>
<DIV>However, it was pointed out to me&nbsp;that relieving the financial accounting side of things doesn't mean that those same functions are not useful and needed on MEP. So, my second inclination is&nbsp;perhaps the accounting&nbsp;functions that are commonly&nbsp;seen in satellite and cellular links&nbsp;may&nbsp;end up being required on MEP, but implemented for different&nbsp;purposes.&nbsp;Searching, rooms, filters, friends lists, modeling,&nbsp;quality of service, and other fun things that are&nbsp;enabled by a rich set of metadata certainly start to sound a lot like "accounting".&nbsp;</DIV>
<DIV>&nbsp;</DIV>
<DIV>The level of accountability (authentication and security)&nbsp;needs to be explored and discussed. How concerned should we be with harmful or unauthorized use? </DIV>
<DIV>&nbsp;</DIV>
<DIV>Traditionally, amateur radio is an open system that runs on the honor system and is self-policed. I would like to&nbsp;remain true to the spirit of amateur radio whenever and wherever that spirit is positive and worth preserving.&nbsp;When considering a digital system compared to an analog system,&nbsp;the issues of identification, authentication, accounting, and security&nbsp;naturally arise. A model to&nbsp;consider is D-Star, which has&nbsp;repeater registration by call sign. This is a user-entry ID. So, to me, it is the equivalent of stating your call sign on the&nbsp;air and remains&nbsp;a voluntary, honor-system act. </DIV>
<DIV>&nbsp;</DIV>
<DIV>I&nbsp;strongly agree with the goal of making MEP a fully functional IP&nbsp;network. That is the baseline assumption, and should be&nbsp;explored, challenged, and talked about until we are&nbsp;more than fully convinced it's the right&nbsp;thing to do, where "it" is well-defined. This leads directly to identifying the requirements for fully supporting IP. </DIV>
<DIV>&nbsp;</DIV>
<DIV>What are the repercussions of amateur radio communicating over a fully&nbsp;functional IP network? What possibly unexpected requirements do we incur upon ourselves by doing so? Are authentication and security and identification controversial issues, or are they now so&nbsp;embedded in the communications ecosystem that to remove these functions would be negative? Is the transport protocol truly unaffected by the type of traffic, which is a hallmark axiom of layered protocols? In other words, is there anything in particular about amateur radio that would cause a change in the&nbsp;requirements for IP? </DIV>
<DIV><BR>More soon,&nbsp;</DIV>-Michelle W5NYV
<DIV></DIV>
<DIV>
<DIV><BR>&nbsp;</DIV>
<DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif"><BR>
<DIV style="FONT-SIZE: 13px; FONT-FAMILY: arial, helvetica, sans-serif"><FONT face=Tahoma size=2>
<HR SIZE=1>
<B><SPAN style="FONT-WEIGHT: bold">From:</SPAN></B> Timothy J. Salo &lt;salo@saloits.com&gt;<BR><B><SPAN style="FONT-WEIGHT: bold">To:</SPAN></B> mep &lt;mep-dev@uppermeadow.com&gt;<BR><B><SPAN style="FONT-WEIGHT: bold">Sent:</SPAN></B> Saturday, November 29, 2008 5:51:24 PM<BR><B><SPAN style="FONT-WEIGHT: bold">Subject:</SPAN></B> Re: [Mep-dev] TCP Performance Issues [oops]<BR></FONT><BR>&gt; Now that I think about it some more, this topic is relevant<BR>&gt; to the MEP.&nbsp; First, I will try to outline the broader<BR>&gt; controversy and describe three contrasting views.&nbsp; Then,<BR>&gt; I will show one area where the MEP project will probably<BR>&gt; need to, in some sense, pick sides.<BR><BR>Oops.&nbsp; The e-mail escaped before its time.&nbsp; I will resend it<BR>when it is complete.<BR><BR>-tjs<BR><BR>_______________________________________________<BR>Mep-dev mailing list<BR><A href="mailto:Mep-dev@uppermeadow.com"
 ymailto="mailto:Mep-dev@uppermeadow.com">Mep-dev@uppermeadow.com</A><BR><A href="http://lists.uppermeadow.com/mailman/listinfo/mep-dev" target=_blank>http://lists.uppermeadow.com/mailman/listinfo/mep-dev</A><BR></DIV></DIV></DIV></div></body></html>
--0-1237434424-1228058659=:567--