headers
Jeff Barnette | 4 Oct 2002 17:06

Re: Why not use smaller packets? Was: RE: client versions - AIX

On Fri, 4 Oct 2002, waldo kitty wrote:

> 
> Jason Hartzell wrote:
> > 
> > >>we're still receiving work from clients working on DES and RC5-56.
> > 
> > That's too funny.
> 
> maybe the thought is to use DES and RC5-whatever as a replacement fuel
> source?  <<GD&R>>
> 
> 
I, for one, would like to know if any of my machines are still working on 
the old contests.  Is there any way I can get that info?

Jeff

--
To unsubscribe, send 'unsubscribe rc5' to majordomo@...
rc5-digest subscribers replace rc5 with rc5-digest
Permalink | Reply | 0 comments |
headers
Ryan Malayter | 4 Oct 2002 17:21

RE: Why not use smaller packets? Was: RE: client versions -AIX

Do you have your clients making logfiles?

Ryan Malayter
Sr. Network & Database Administrator
Bank Administration Institute
Chicago, Illinois, USA
PGP Key: http://www.malayter.com/pgp-public.txt
:::::::::::::::::::::::::::::::
We have just enough religion to make us hate, but not enough to make us
love one another. 
     -Jonathan Swift

-----Original Message-----
From: Jeff Barnette [mailto:barnette@...] 
Sent: Friday, October 04, 2002 10:06 AM
To: rc5@...
Subject: Re: Why not use smaller packets? Was: RE: [RC5] client versions
-AIX

On Fri, 4 Oct 2002, waldo kitty wrote:

> 
> Jason Hartzell wrote:
> > 
> > >>we're still receiving work from clients working on DES and RC5-56.
> > 
> > That's too funny.
> 
> maybe the thought is to use DES and RC5-whatever as a replacement fuel
> source?  <<GD&R>>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Bruce Wilson | 4 Oct 2002 18:16

RE: OGR motivations/achievements in general

I have posted the status of OGR to this list in the past, but perhaps
it's time for a restatement.

Status of OGR
*  The OGR client has a bug which weakens the accuracy of our results.
A truncation error causes rulers with many segments to be aborted
incorrectly.  We were making some good progress towards fixing this bug
and releasing new clients when we solved RC5-64.  At that point, all OGR
development was put on hold in favor of RC5-72.

Because of the verification plan we are using, we do not expect to have
to discard work done with current buggy clients unless the bug affected
the results for a given stub.  Our original plan was to require two
identical results for the same stub from different participants.  Now we
will be revising that to say that one of the results must also come from
a "fixed" client.  If the buggy client gives the same results as the
fixed client, the bug was not a factor.  If the buggy client gives a
different result, then we wait for a confirmation from a different
participant using the fixed client.

Status of OGR stats
*  Progress is very hard to compute for OGR, relative to the RC5
projects.  For RC5, we only need to track whether the work was completed
or not (a single bit of information).  For OGR, it is necessary to track
what participant completed each stub (and now, we must also track
whether the client was pre-fix or post-fix).  We also keep track of how
many nodes were found in each stub, to verify that the participants got
the same results.  In all, the storage requirement per workunit is
orders of magnitude greater than RC5-64 - we go from one bit (1/8 of a
byte) per workunit to about 50 bytes.  This is just the beginning.
(Continue reading)

Permalink | Reply | 31 comments |
headers
Bruce Wilson | 4 Oct 2002 18:16

RE: DNet questions (repost)

The first, best and sometimes only way to get a new project on
distributed.net is for someone who understands the project to write a
core for our network.  We are eagerly looking for other projects we can
add to our network, but without coders who know how to distribute and
solve the problem, we have nothing but a nice wish.

__
Bruce Wilson <bwilson@...>
PGP KeyID: 5430B995, http://www.toomuchblue.com/ 

Build a man a fire and he'll be warm for a day.
Set a man on fire, he'll be warm for the rest of his life.

| -----Original Message-----
| From: owner-rc5@... 
| [mailto:owner-rc5@...] On Behalf Of Earl 
| '<--Havoc--{{{' Stenlund
| Sent: Friday, October 04, 2002 09:48
| To: rc5@...
| Subject: RE: [RC5] DNet questions (repost)
| 
| 
| I too would like to see Dnet move in the direction of medical 
| and other
| research besides encryption. Currently I'm looking for a 
| project of this
| nature that I can setup the way I have dnet setup. Basically 
| I have the
| client on the our servers here at work and when a user logs 
| in their log
(Continue reading)

Permalink | Reply | 7 comments |
headers
James Spinks | 4 Oct 2002 18:52
Picon

RE: OGR motivations/achievements in general

Quoting Bruce Wilson <bwilson@...>:

> I have posted the status of OGR to this list in the past, but perhaps
> it's time for a restatement.

Thanks for this Bruce, your continued presence here is much appreciated,  I
think while I knew most of this already the stuff about the database loss was
new to me (I don't know if that bit didn't make the lists before or if I just
missed it).  It's good for us all though that you are here and able to make
these occasional regurgitations to keep the masses quiet.

I think I'm gonna have to start keeping some of your emails so I can start
quoting replies from them in future if these questions keep coming up.

--

-- 
James Spinks
--
To unsubscribe, send 'unsubscribe rc5' to majordomo@...
rc5-digest subscribers replace rc5 with rc5-digest
Permalink | Reply | 0 comments |
headers
Timothy Marsh | 4 Oct 2002 20:19
Favicon

Re: DNet questions (repost)

Well put.

    I think it is important for evryone to remember that to develop a new
project for dnet requires not only an intimate knowledge of distributed
computing, but also an intimate knowledge of the application.  While it may
not be uncommon to find coders who understand the interworkings of
encryption and/or OGR's (being a mathematical concept), it is considerably
harder to find coders who understand biochemical applications.  Keep in mind
that most projects working on these type of biochemical applications are
performed by many different groups from different areas of study working
together.  Typically a medical research group designs an application for
analysis of biochemical data.  Then, if possible, the software may be
tailored (with the help of people knowledgeable in distributed computing) to
a distributed computing type of platform.  Then the project cunching begins.

    In short, most people behind medical research forms of distributed are
from much larger and multi-specialized groups than dnet.  While it would be
possible for dnet to run such an application, I don't believe it currently
has anyone experienced in these types of medical software.  I also don't
believe that one person experienced in this area would be able to create the
software by themselves, given the complexity.  Dnet would have to work
closely with another company to accomplish this sort of goal.  Sound
familiar.  The relationship with United Devices is accompliahing just such a
goal.  But there are many differences between these two organizations
including their complexity and purpose.  For that reason I am glad that both
are here and I have given many cycles to both worthy causes.

Timothy

----- Original Message -----
(Continue reading)

Permalink | Reply | 1 comment |
headers
Jeff Barnette | 4 Oct 2002 20:14

Clients still working on old contests Was Re: FW: Why not use smaller packets? Was: RE: [RC5] client versions -AIX

> -----Original Message-----
> From: owner-rc5@...
> [mailto:owner-rc5@...]On Behalf Of Ryan Malayter
> Sent: Friday, October 04, 2002 9:21 AM
> To: rc5@...
> Subject: RE: Why not use smaller packets? Was: RE: [RC5]
> client versions
> -AIX
>
>
> Do you have your clients making logfiles?
>
> Ryan Malayter
>
>> I, for one, would like to know if any of my machines are still working
>> on
>> the old contests.  Is there any way I can get that info?
>>
>> Jeff

All of the clients I know about are currently reporting to my perproxy;  
however, when I first installed the dnet clients on a few machines, they
were connecting directly to the dnet proxies.  What I'm wondering about is
whether any of these original machines (which were working on RC5-56 and
DES at the time) may have "gotten away from me" over the years.  I don't 
have visibility because they don't connect to my perproxy.  Only the dnet 
folks with access to the raw stats can tell me.

Jeff
(Continue reading)

Permalink | Reply | 4 comments |
headers
Ryan Malayter | 4 Oct 2002 22:46

RE: Clients still working on old contests

I found a few old clients on my netowrk by changing a few logging rules
on my firewall and checking the logs a week later.

Ryan Malayter
Sr. Network & Database Administrator
Bank Administration Institute
Chicago, Illinois, USA
PGP Key: http://www.malayter.com/pgp-public.txt
:::::::::::::::::::::::::::::::
There is only one basic human right, the right to do as you damn well
please. And with it comes the only basic human duty, the duty to take
the consequences.
     - PJ O'Rourke 

-----Original Message-----
From: Jeff Barnette [mailto:barnette@...] 
Sent: Friday, October 04, 2002 1:14 PM
To: rc5@...
Subject: [RC5] Clients still working on old contests Was Re: FW: Why not
usesmaller packets? Was: RE: [RC5] client versions -AIX

> -----Original Message-----
> From: owner-rc5@...
> [mailto:owner-rc5@...]On Behalf Of Ryan Malayter
> Sent: Friday, October 04, 2002 9:21 AM
> To: rc5@...
> Subject: RE: Why not use smaller packets? Was: RE: [RC5]
> client versions
> -AIX
>
(Continue reading)

Permalink | Reply | 1 comment |
headers
Greg Wooledge | 5 Oct 2002 00:12

Re: OGR motivations/achievements in general

Jack Beglinger (jackb@...) wrote:

> 9 yr old, 3 yr old, 18 months....  The 9yr shares computer/office space with 
> dad - leaves the main gates open to run back and forth, the 3yr releases the 
> 18 months from play area by lifting side - 3 wants to play with 18. and 
> together the come giggling into the computer area... fun in sues.

Just wait until you start pulling french fries, used band-aids, etc. from
your floppy disk drive.  (Or have you already been through that with the
oldest one?

--

-- 
Greg Wooledge                  |   "Truth belongs to everybody."
greg@...              |    - The Red Hot Chili Peppers
http://wooledge.org/~greg/     |
Permalink | Reply | 0 comments |
headers
Bruce Wilson | 5 Oct 2002 01:02

RE: Clients still working on old contests Was Re: FW: Why not use smaller packets? Was: RE: [RC5] client versions -AIX

No, we have no way of tracing back the machines submitting RC5-56 and
DES work.  This work is submitted on a completely different TCP port
(2056) than our current network (2064) and the port we'll be using with
the RC5-72-capable clients (2072 - see the pattern?).  To reduce the
load on our servers, we simply block the old port.  It may not be the
most graceful way to handle the problem, but it is simple and effective.

__
Bruce Wilson <bwilson@...>
PGP KeyID: 5430B995, http://www.toomuchblue.com/ 

Build a man a fire and he'll be warm for a day.
Set a man on fire, he'll be warm for the rest of his life.

| -----Original Message-----
| From: owner-rc5@... 
| [mailto:owner-rc5@...] On Behalf Of Jeff Barnette
| Sent: Friday, October 04, 2002 13:14
| To: rc5@...
| Subject: [RC5] Clients still working on old contests Was Re: 
| FW: Why not use smaller packets? Was: RE: [RC5] client versions -AIX
| 
| 
| > -----Original Message-----
| > From: owner-rc5@...
| > [mailto:owner-rc5@...]On Behalf Of Ryan Malayter
| > Sent: Friday, October 04, 2002 9:21 AM
| > To: rc5@...
| > Subject: RE: Why not use smaller packets? Was: RE: [RC5]
| > client versions
(Continue reading)

Permalink | Reply | 3 comments |

Gmane