Re: [Pyxida-users] Pyxida developer + help

[Jonathan L. <jon...@no...>]

Your inclination is correct.  See forwarded message.

-jonathan

On Apr 9, 2008, at 1:53 PM, ext Eric Chan-Tin wrote:

> Thank you for the clarification.
>
> I've been reading through your published papers and code and I must  
> say I learned a lot about how Pyxida works :)
>
> I still have a quick and easy (hopefully) question. Each Pyxida node  
> has a small amount of neighbors and learn about other nodes through  
> other nodes. I am running on about 200 machines, and it seems like  
> all the Pyxida nodes are adding each other to their upNeighbors set.  
> NCClient.java has the MAX_NEIGHBORS set to 512.
>
> I wanted to make sure that I am not misreading the code, but  
> shouldn't each Pyxida node keep a small number of neighbors in its  
> upNeighbors set. Else, it will be a lot of updates to send a  
> gossipMessage to all 200 neighbors, and it also means that a Pyxida  
> node has the whole network in its upNeighbors set - that seems  
> wrong. I am inclined to change the value of MAX_NEIGHBORS but wanted  
> to make sure that my understanding is correct first.
>
> Thank you,
>
> Eric
>
> On Mar 13 2008, Jonathan Ledlie wrote:
>
>>
>> Eric,
>>
>> IIRC, we set up two coordinate systems so we could experiment with   
>> different algorithms for updating them in parallel.  In other  
>> words,  the same sets of latency measurements would be experienced  
>> at the same  time -- how do different algorithms respond to them.   
>> You might well  find this functionality useful.  If you wanted to  
>> expand it to include  more than two, I'm sure this wouldn't be too  
>> tricky.
>>
>> So, both coordinates are for the same node, but different  
>> coordinate  systems.
>>
>> -jonathan
>>
>> On Mar 13, 2008, at 1:45 PM, Peter Pietzuch wrote:
>>
>>> Jonathan, can you shed some light into this?
>>>
>>> Cheers,
>>>  Peter
>>>
>>> Eric Chan-Tin wrote:
>>>> Thanks for the help.
>>>>
>>>> The problem was that I was using jdk 1.6 and even though I  
>>>> cleaned  and
>>>> recompiled util-prp and ASync and copied the jar files to Pyxida/  
>>>> lib, it
>>>> still failed with the same error message. But I downgraded to  
>>>> jdk  1.5.0_09
>>>> and seems to be working fine now. I assume that some of the jar   
>>>> files in
>>>> Pyxida/lib were compiled under jdk1.5.
>>>>
>>>> One more question though. Why are there "update1" and "update2"?  
>>>> More
>>>> specifically, why would CoordClient have two NCClient - a  
>>>> primaryNC  and a
>>>> secondaryNC?
>>>>
>>>> Let's say there are two Pyxida clients - A and B. In an update,   
>>>> update1
>>>> would refer to A and update2 would refer to B (B sends A a   
>>>> message). Is
>>>> that a correct interpretation?
>>>>
>>>> Thank you,
>>>>
>>>> Eric
>>>>
>>>> On Mar 11 2008, Peter Pietzuch wrote:
>>>>
>>>>> Dear Eric,
>>>>>
>>>>> Eric Chan-Tin wrote:
>>>>>> I am interested in helping with the Pyxida project. I read a   
>>>>>> couple of
>>>>>> papers about how network coordinate systems (for example GNP  
>>>>>> or  Vivaldi)
>>>>>> are insecure and wanted to implement some countermeasures  
>>>>>> against  those
>>>>>> attacks. Instead of creating my own Vivaldi implementation, I   
>>>>>> thought of
>>>>>> just using Pyxida instead.
>>>>> It's great to hear that you want to help out with the Pyxida   
>>>>> project. Let
>>>>> me know when you have code to contribute and I'll set you up with
>>>>> developer access.
>>>>>
>>>>>> [quote] WA 1205186886103 NetUtil : Could not deserialize java   
>>>>>> object:
>>>>>> java.io.StreamCorruptedException: invalid type code: 80 WA   
>>>>>> 1205186886103
>>>>>> ObjComm$ReadConnHa: No valid msgObject received from
>>>>>> remoteAddr=<address>:55504. Ignoring. [\quote]
>>>>> Usually, deserialisation problems like this one indicate that   
>>>>> class files
>>>>> were compiled using different Java versions. Have you tried   
>>>>> recompiling
>>>>> everything from scratch using, say, Java 1.5? Make sure that  
>>>>> you  also
>>>>> rebuild some of the helper libraries (i.e. prp-util).
>>>>>
>>>>> Let me know if you have any other questions.
>>>>>
>>>>> Cheers,
>>>>> Peter
>>>
>>>
>>> -------------------------------------------------------------------------
>>> This SF.net email is sponsored by: Microsoft
>>> Defy all challenges. Microsoft(R) Visual Studio 2008.
>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>>> _______________________________________________
>>> Pyxida-users mailing list
>>> Pyx...@li...
>>> https://lists.sourceforge.net/lists/listinfo/pyxida-users
>>