Domain: planet-lab.org
Stories and comments across the archive that link to planet-lab.org.
Comments · 22
-
Re:I can see where this is going
If you're an ethical researcher wanting to run a distributed scan of the 'net, the proper way to do it is to use something like PlanetLab, which has been designed for uses like that and is freely available for research use. It's what everyone else uses, and it works great. Either that, or go and use your grant money to provision yourself appropriately for a job like this, which is what we did when I was in grad school. Commandeering routers and other devices for personal use is inexcusable.
Honestly, my first thought was, "What research ethics committee gave him the go-ahead?" My guess: the researcher didn't ask, because none of them would ever let him do it. Besides consuming bandwidth for tens or hundreds of thousands of Internet users without their consent (some of whom were likely capped), he's also loaded code onto their machines: code which they have no guarantee will work as expected in all circumstances. In fact, for all they know, they may have bricked tens of thousands of devices without realizing they did so, then taken their lack of response later as a simple incompatibility with his code.
When I was in grad school, we were doing web crawler and search engine research that was considered to be a bit on the edge of what was permissible (and our work resulted in serious threats of lawsuits aimed at our university), but we would never consider doing something like what they did. No credible conference or journal would publish this sort of work either, which is as it should be. Researchers have a responsibility to act responsible, and this anonymous one didn't.
Also, you've said it was useful research, but it really wasn't. These vulnerabilities are widely documented, and those researchers were not only able to publish earlier, they were also able to do so without engaging in gross ethical violations.
-
Link to the package
In case other people had as much trouble as I did finding the package: www.cs.princeton.edu/~sapanb/vsys [Download Vsys] http://git.planet-lab.org/?p=vsys.git;a=summary [git repository]
...also, interestingly, Vsys is not written in C, but Ocaml, which is a solid type-safe programming language. This is reassuring from the security standpoint given that it is a recent package. -
Re:Nail in the coffin?Doesn't this strike you as ironic? At first glance, sure. But when you realize that the bandwidth consumed is causing other research and educational use of the network to be severely degraded, it eases the irony.
If you want to test a new distribution scheme, you can use a lab setup. If you need a lot of nodes, you can use http://www.planet-lab.org/. If 99.9% of the use of the new technology has nothing to do with research, I'm not sure I see the value in letting that use hamper the rest of the research at your institution. -
Re:Queue "They Have no Right" posts
bastardizing a research network
Yeah right. Aside from using it to test some pretty fancy high speed protocols, the Internet2 in general is really nothing more than a fast pipe for college students to download music on, insulated from the original Internet by BGP. You never see an academic conference requiring "tests on the Internet2" because its geographic concentration is entirely in North America and its speed is totally beyond anything you see in the real Internet; that is why everyone wants PlanetLab instead.
- sm -
Re:YawnI believe that there is also some work being done on making Xen work on clusters.
http://www.cl.cam.ac.uk/Research/SRG/netos/xeno/
"This wide-ranging project has two main strands of work:- Development of the Xen virtual machine monitor, a high-performance hypervisor for hosting multiple commodity operating systems on a single x86-based server. This forms the core of each Xenoserver node, providing the resource management, accounting and auditing that we require. Xen finds numerous applications outside the Xenoserver project. These inclue server consolidation and secure computing platforms.
- Development of the Xenoserver Open Platform control software for managing networks of Xenoservers. Our research includes distributed storage, server discovery, resource management and authentication, authorization and accounting (AAA) functions. This work finds relevance to Grid computing and to globally distributed testbeds such as PlanetLab.
-
Some links with real information
A better news summary at:
http://www.internetnews.com/dev-news/article.php/3 406161
Or the organization itself:
http://www.planet-lab.org/ -
Re:waste of time
Want information? Read the abstract: http://www.planet-lab.org/PDN/PDN-02-001/pdn-02-0
0 1.pdf And of course, the official site: http://www.planet-lab.org/ -
Re:waste of time
Want information? Read the abstract: http://www.planet-lab.org/PDN/PDN-02-001/pdn-02-0
0 1.pdf And of course, the official site: http://www.planet-lab.org/ -
Re:You bet
Funny you should mention, "...you can forget about using linux on the desktop...". Planet Lab is built on Linux. Since this article is, in fact, about Planet Lab. Linux is a very large part of the picture.
-
Re:You bet
Funny you should mention, "...you can forget about using linux on the desktop...". Planet Lab is built on Linux. Since this article is, in fact, about Planet Lab. Linux is a very large part of the picture.
-
Re:Well...
Intel's press release about the same speech has a little bit more information, although nothing technical.
http://www.intel.com/pressroom/archive/releases/20 040909corp.htm
Also interesting: a link to the open platform website:
http://www.planet-lab.org/
Interesting quotes:
"Applications run on PlanetLab are decentralized, with pieces running on many machines spread across the global Internet. They can also self-organize to form their own networks, and include some form of application processing inside the network (instead of at the edges), adding new intelligence and capabilities to the Internet."
"It would provide a platform on which Web services can run and a way to connect grid computing sites and utility data centers. It sits above the new physical infrastructure supplied by Internet 2 and above the networking layer where IPv6 functions, adding a new stratum of higher-level functionality to the Internet."
Why it has to replace the current TCP/IP-infrastructure is still unclear (apart from selling more hardware). -
Publicity
They're doing what they can to get some publicity to the PlanetLab project.
-
This is what Coral is using
NYU's Coral Project, which seems to be the rage of the so-called "karma whores" on Slashdot these days seems to be a consumer of PlanetLab. Seems to be a strange coincidence.
-
Re:What's in it for Intel?
You're on the right thread.
The internet as we know it is a collaboration. This entire discussion about Intel owning this planetlab by posters above is kinda funny, shows that they don't read fully, when the links are provided, or even do research before posting.
PlanetLab is also a collaboration, not just Intel, but also includes acadamics and other corporations such as google and hp. Look at the consortium link to see who is involved at http://www.planet-lab.org/
Also, be thankful that corporations and academics have banded together to keep things advancing. None of you would not be connected currently if it wasn't for Intel! Remeber DIX? Probably not.... thank DIX for ethernet people. Who is DIX? Digital, Intel and Xerox. They created the first ethernet standard which became the basis for 802.3 in 1980.
No, I'm not an Intel employee... not even employed in tech. Just sick of people not thinking before they post. -
Re:Can I joinActually www.planet-lab.org/consortium lays it out quite well. The trick is they don't like people as such only as members of organizations. It sounds non-profits don't pay any fee.
Let us know when your organization is part, then we can join your organization instead of going to the consortium directly.
Or maybe slashdot.org can sign up.
Stephan
-
Re:Can I joinThe Faq at www.planet-lab.org says this
Individuals are not allowed to directly join PlanetLab. Your home institution must be a member of the PlanetLab Consortium. To see if your institution is already on PlanetLab, click here. This page also identifies your site's Principal Investigator (PI), who is the person that approves PlanetLab accounts at your institution.
So is your institution on the list? Or is it a second planet-lab? Stephan -
Re:Can I joinThe Faq at www.planet-lab.org says this
Individuals are not allowed to directly join PlanetLab. Your home institution must be a member of the PlanetLab Consortium. To see if your institution is already on PlanetLab, click here. This page also identifies your site's Principal Investigator (PI), who is the person that approves PlanetLab accounts at your institution.
So is your institution on the list? Or is it a second planet-lab? Stephan -
Give back platform access, not (only) code ...
Google should consider "giving back" by providing a platform for services to run on -- a PlanetLab for the SourceForge world. Google code isn't about a single server, it's about clusters of servers, and few people have the budget and experience to set up distributed clusters on their own.
-
Adding Lowers Value: Right, but how bout...I agree with Dave and Doc that in most cases mucking around with the physical and code layers of the internet is a Miserble idea.
Not only technically will it likely muck things up, but in the real world, some big gorilla of a firm will find a way to take advantage for them selves, at the expense of others.
But, as I once told my ex-crush, Never Say Never baby.
As I picked up from MIT's Tech Review Planet Lab. Seems to me like a good idea, but not sure. Particularly after all the time's I've read Lessig pound the end-to-end point home. Here's a snippet from the Intel press release on Planet Lab. what do you think?SANTA CLARA, Calif, June, 24, 2003 -- Researchers from the University of California, Berkeley, HP, Intel Corporation, Princeton University, the University of Washington and more than 60 universities from around the world have joined together to form PlanetLab, a global test bed for inventing and testing prototype Internet applications and services. The researchers aim to spark a new era of innovation by using "overlay" networks to upgrade and expand the Internet's features and capabilities.
PlanetLab may lead to new ways of protecting the Internet from viruses and worms. It could also enable new capabilities, such as persistent storage, the idea of giving the Internet a "memory." For example, 100 years from now a piece of data could still be found, even though the original computer on which it was posted no longer exists. In addition, this research could influence the future design of servers and network processors.
Upgrading the Internet The Internet has been based on a small set of software protocols that direct routers inside the network to forward data from source to destination, while applications run on computers connected to the edges of the network. The simplicity of the software model enabled the Internet to rapidly scale into a critical global service; however, this success now makes it difficult to create and test new ways of protecting it from abuses, or from implementing innovative applications and services.
The PlanetLab concept was born when Intel researchers gathered a group of leading network and distributed systems researchers to discuss the implications of a new, emerging class of global services and applications on the Internet. This new class of services is designed to operate as "overlay" networks, which have emerged as a way of adding new capabilities to the Internet. The concept of an overlay or "on top of" approach might be familiar from text books where additional details are added to an image by laying a transparent sheet containing new graphics on top of an existing page. An example of this is overlaying an image of human muscles on top of an illustration of bones to show how the body works.
These overlay networks incorporate the Internet for packet forwarding, but integrate their own intelligent routers and servers on top of the Internet to enable new capabilities without affecting its performance today. These applications are decentralized, with pieces running on many machines spread across the global Internet, they can self-organize to form their own networks, and include some form of application processing inside the network (instead of at the edges), adding new intelligence and capabilities to the Internet.
One example of an overlay network enabling a new kind of Internet application is robust video multicasting. Today, a standard Web site that receives too many requests for the same video clip can bog down or crash; however, if this site were supported by an overlay network of smart routers and globally distributed content storage sites, it could redirect requests on-the-fly, sending them across the Internet to the nearest available content site to ensure the best viewing experience while keeping the site up and running.Sometimes you just have to say screw it, w
-
Go unspecialized CS, and explore
As a student who's about to finish his undergraduate degree and hoping to specialize in networking in graduate school, I'd like to offer some insight...
The reason you can't find a college with a major like "network engineering" is simple -- specializations such as these are for graduate students. Undergraduates in Computer Engineering, by contrast, go through a diversified program including math, physics, chemistry, humanities, etc., in addition to the real meat of their degree: programming, architecture, networking (your interest), and many other facets of computer science.
Now you might think that because of this well-rounded approach, you'll only take one class on computer networking. This is not true. Some universities offer more than one undergraduate class in computer networking. At the university I attend, there is a "general" networking class, but there is also a class and lab devoted entirely to the physical layer, and within a year a class on peer-to-peer technologies might be incorporated. Also, the CS department will typically allow you as an undergraduate to petition to take graduate-level computer CS classes, including any networking ones you might find. And if you can't even do that, don't sweat it -- a single professor specializing in networking at a university might have multiple projects available for undergraduates to work on. Earnestly, this is the best way to get your feet wet on the Internet2 and other networking-related projects that interest you. Through meeting with a professor, I'm now working on a peer-to-peer (Pastry DHT based) client that will, perhaps, see the light of day on PlanetLab, followed by a general release.
Another reason to not shy away from the "general" CS degree is that you might find something along this track that is more interesting than networking. Besides spending my time on that peer-to-peer project, I'm also volunteering on a project in the EE department developing software for Lego robots. Sure, I'm little more than a code monkey over there, but it's interesting to hang around the lab and learn of new things and new technologies. I've also done work in the physics lab, programmed in the Geology department (writing Matlab programs to simulate earthquakes and analyze seismographic data), and done technical support for campus computer users in my almost-four years here. If you go looking, you'll find a myriad of cool things at your university -- perhaps cooler than networking.
Best of luck in your searches,
shadowmatter -
Re:Is it just me...
Close, but not quite. Planetlab is not a closed, high performance network. Rather, it's more of an overlay testbed: The machines reside on the Internet (companies that host nodes) and on the Internet2 (research universities). That's part of what's so cool about it - the machines reside all over the world (see the map on the planetlab website - it's an accurate reflection of the location of the nodes). They have a lot of visibility into nooks and crannies on the Internet, and they're beginning to be deployed enough that there's often a planetlab node nearby, whereever in the network you are.
-
A meta-testbedA Planet Lab pagesays:
PlanetLab also serves as a meta testbed on which multiple, more narrowly-defined virtual testbeds can be deployed. That is, if we generalize the notion of a service to include what might traditionally be thought of as a testbed, then multiple virtual testbeds can be deployed on PlanetLab.
Any time a discussion starts to use the word meta you know you have achieved buzzword satori and can stop reading.
Anything you can do I can do meta. I can do anything meta than you.