Message boards :
Science :
What Is Our Mission At iThena?
Message board moderation
Author | Message |
---|---|
![]() Send message Joined: 26 Sep 19 Posts: 6 Credit: 208,458 RAC: 12 ![]() |
I can't find much in the way of information as to what the purpose of iThena is. I joined on the assumption that we would be performing good science here. I still am sure we are :) Just would like to be informed as to the course and intention of our shared scientific journey... :) The wonderful lady on my profile is Hayley Westenra- the finest classical soprano on the face of the earth. I highly recommend her music- Ceterum censeo Carthaginem delendam esse... |
![]() ![]() Send message Joined: 25 Aug 19 Posts: 409 Credit: 1,240,185 RAC: 0 ![]() ![]() |
This week I will develop a project description. I will post information in this forum thread. |
![]() ![]() Send message Joined: 25 Aug 19 Posts: 409 Credit: 1,240,185 RAC: 0 ![]() ![]() |
Well ... I didn't have enough time to complete the website 'About iThena' during the weekend. In the next week I will do it ;) |
![]() Send message Joined: 24 Sep 19 Posts: 17 Credit: 1,588,258 RAC: 12 ![]() ![]() ![]() |
Well ... Considering you're asking people to donate their computer time to you, for nothing in return, how about sparing a minute or two for a couple of sentences on the forum? What does your project do? * It analyzes protein structures * It factors large numbers * It solves chess problems * it's solving n-body orbital mechanics problems See?, It's not that difficult. It's also not that difficult to click on "No New Work" Want to find one of the largest known primes? Try PrimeGrid. Or help cure disease at WCG. ![]() |
![]() ![]() Send message Joined: 25 Aug 19 Posts: 409 Credit: 1,240,185 RAC: 0 ![]() ![]() |
Two sentences of the project description are available from the very beginning: The iThena distributed project concerns experimental mapping of network structures included in the Internet. The project is in closed beta phase. Currently, the only application available in the project (iThena CNode) performs a sequence of traceroute procedures from client computers. The resulting data is sent back to the server and submitted to the main database, where it can be further analyzed. |
![]() Send message Joined: 24 Sep 19 Posts: 17 Credit: 1,588,258 RAC: 12 ![]() ![]() ![]() |
Two sentences of the project description are available from the very beginning: Thank you. In 2008 there was a BOINC project called Depspid (wayback link: https://web.archive.org/web/20071222092002/http://www.depspid.net/). Depspid used BOINC hosts to crawl the Internet to essentially create some sort of map. Is iThena similar to that? It doesn't consume CPU, but instead is using network resources? If iThena is indeed similar to Depspid (i.e., the app is constantly sending outgoing traffic to the Internet), I'm curious how you avoided the pitfalls (reverse DDOS) that affected Depspid? Want to find one of the largest known primes? Try PrimeGrid. Or help cure disease at WCG. ![]() |
![]() ![]() Send message Joined: 25 Aug 19 Posts: 409 Credit: 1,240,185 RAC: 0 ![]() ![]() |
Many thanks for the link! I didn't know this project;) A valuable source of information. Depspid worked a little differently and had different purposes - if I understand correctly after reading the information quickly. Depspid mapped domains / subdomains / sites / directories etc. iThena is currently mapping the logical / physical structure of hops limiting itself to IPv4 addresses (in theory DNS should not be very involved but in my software there is probably a bug). The current application (only one) uses the network. This is not sooooo great use when we run one application. Over time, I may add more computational applications. However, this is a more distant future. |
![]() ![]() Send message Joined: 25 Aug 19 Posts: 409 Credit: 1,240,185 RAC: 0 ![]() ![]() |
This is a very quick answer - I will send more information soon. |
![]() Send message Joined: 24 Sep 19 Posts: 17 Credit: 1,588,258 RAC: 12 ![]() ![]() ![]() |
Many thanks for the link! Let me give you a little warning about DepSpid. It was an interesting and fun BOINC project -- but it was also VERY dangerous. It wasn't malicious, but running it had some serious and not-so-obvious risks. Running it on one computer worked just fine. Running it on two or three computers could swamp my home's Internet connection. Essentially, although it was working as intended, it had the same effect as a DDOS (Distributed Denial of Service) attack against your own network. My house has wired gigabit ethernet and at that time my Internet connection was probably about 30Mb down and 10MB up. You do the math. :) Some users have access to data centers full of servers and have legitimate authority to run BOINC on those servers. If they run an app similar to DepSpid on a lot of computers, they will bring their company's Internet connection to a grinding halt. That might affect their employment status, obviously, but there's also potential risk for you. Depending on the damage that is done, you (or your employer) may find yourself on the wrong side of a lawsuit in some court somewhere. In some jurisdictions you might even face criminal liability. In this scenario, the argument would probably be that software which you created and distributed caused financial harm to company X. It's not an unreasonable argument. What really sucks is that I don't think there's an easy fix for this. Making the network access slower doesn't really help because there's always going to be someone with a lot more computers and/or a much slower Internet connection. Please keep this in mind. People running this kind of app can inadvertently cause serious network disruptions. In my case, I had to deal with my kids complaining about the Internet being slow. That's much better than having a really awkward talk with my boss about how the network problems of the last day or two were traced back to that BOINC thing they let me run on computers "for science." :) Want to find one of the largest known primes? Try PrimeGrid. Or help cure disease at WCG. ![]() |
![]() ![]() Send message Joined: 25 Aug 19 Posts: 409 Credit: 1,240,185 RAC: 0 ![]() ![]() |
You are right. The current procedure slows down the client's link. This can be problematic. I am aware of the problem. Packages, however, go for different targets (conceptually in a random pool). There is no question of targeting a particular host (impossible event). (Currently there is a problem with DNS load - I remember that). Being in a company you can start many machines with different priority on a company computer and also harm the company. There is no rule for this. You don't need iThena app at all. The possibilities of this application are very limited. However, I think that if someone consciously knows the realities, they will not launch 100,000 applications simultaneously. Equally, wget may be harmful. Depends on the user. A valuable thread. I have to add a sentence with proper attention. |
![]() Send message Joined: 2 Oct 19 Posts: 20 Credit: 4,947,466 RAC: 222 ![]() ![]() ![]() |
Any idea how this was different than netdimes.org? https://web.archive.org/web/20170919072818/http://www.netdimes.org/new/ Last I consulted the admin of that project, they had stopped with the research and moved it to a private company. IIRC Netdimes was an NCI application that had its own client and was not a BOINC project. My other thought was surveil@home but it had a different agenda - https://web.archive.org/web/20140211221835/http://surveill.dei.uc.pt/surveill/ ![]() |
![]() ![]() Send message Joined: 25 Aug 19 Posts: 409 Credit: 1,240,185 RAC: 0 ![]() ![]() |
Yes. Netdimes is significantly ideologically closer at this moment. |
![]() Send message Joined: 2 Oct 19 Posts: 20 Credit: 4,947,466 RAC: 222 ![]() ![]() ![]() |
I used to email back and forth with Ido Blutman who was one of the programmers but he stopped replying around 2014 and I think my last attempt to contact him was 2015. However, I would have to look through maybe another email account at work to see my last response from someone involved that explained them moving on. If you are interested in picking any of their brains of course. I think their client required Java. Not sure what yours requires. However, I don't recall their app wreaking havoc on the network and I ran over a hundred clients for them. https://stats.free-dc.org/user/dimes/Coleslaw Edit: Looking through my installation backups, I actually still have a copy of their older 0.5.5.3 agent (client) for Windows. IIRC it worked much better than their last relase...lol. For some screenshots, we had a few how-to's for setting up the client. Windows - https://hardforum.com/threads/dimes-installation-walk-through.1768712/ Ubuntu - (no longer has pics) https://hardforum.com/threads/dimes-ubuntu-installation-guide.1804110/ ![]() |
Send message Joined: 6 Nov 19 Posts: 2 Credit: 0 RAC: 0 ![]() |
After 12 years of operations, we could not fund DIMES and moved on. Ido left the university, and I myself still doing hundreds of millions of traceroute a week from dedicated servers and use them to identify IP hijack attacks. DIMES contributed to the science of networking by sharing freely its data, which was used by many academians in papers. The papers I published with this data can be found in http://www.eng.tau.ac.il/~shavitt/Mapping.html Among them, The first paper that describe the project: http://www.eng.tau.ac.il/~shavitt/pub/CCR05.pdf A nice analysis: http://www.eng.tau.ac.il/~shavitt/pub/PNAS07.pdf and a recent paper that looks at IP hijack attacks: https://scholarcommons.usf.edu/mca/vol3/iss1/7/ I will be happy to help on this project, and would strongly encourage you to make the data available to the community. Yuval Shavitt Tel Aviv University |
![]() ![]() Send message Joined: 25 Aug 19 Posts: 409 Credit: 1,240,185 RAC: 0 ![]() ![]() |
Hi Yuval! Many thanks for the message! I think we can cooperate. I probably have a lot of questions. I will send you a message today on priv ;) |
![]() ![]() Send message Joined: 25 Aug 19 Posts: 409 Credit: 1,240,185 RAC: 0 ![]() ![]() |
A very interesting problem. New article (2018). China still has 8 POPs in the US, and the US in China exactly zero? |
Send message Joined: 6 Nov 19 Posts: 2 Credit: 0 RAC: 0 ![]() |
There was a reason DIMES was not using BOINC. We cared about what agent get which task and this was not easy to implement with BOINC. Over the years we run many random experiments but also many experiments that require specific tasks to specific locations. |
Send message Joined: 13 Jun 20 Posts: 3 Credit: 86,306 RAC: 0 ![]() |
Where can I find the data visualization now? That link doesn't work anymore. |
![]() ![]() Send message Joined: 25 Aug 19 Posts: 409 Credit: 1,240,185 RAC: 0 ![]() ![]() |
New visualization page address: https://vi.ithena.net/ Additionally, you can read a new article by iThena on Everipedia: https://everipedia.org/wiki/lang_en/ithena |
![]() ![]() Send message Joined: 23 Sep 19 Posts: 9 Credit: 3,241,384 RAC: 169 ![]() ![]() ![]() ![]() |
New visualization page address: https://vi.ithena.net/ Any way this could be made into a screensaver? ![]() ![]() |
© 2019-2024 iThena. All rights reserved. | Private Policy