ICN

[Stephen Wolff <>]

Hi Alex,

Thank you for the rapid response to get us moving.

I believe the primary repository of genomic data on the ICN testbed is the 
Genomic Data Commons at University of Chicago.  UC Santa Cruz (also on the 
testbed) had CGhub and although I understand that project is complete perhaps 
the data are still accessible there?  It would be helpful if one of our UCSC 
colleagues could clarify.

Aspera and GridFTP are IP-based and are not compatible with the ICN 
protocols.  For your experiment, however, it is possible to set up a parallel 
topology - possibly even using the same AL2S links that support the ICN 
testbed - to run the IP-based transfers.  I think there are several 
possibilities, but maybe my colleague John Hicks could weigh in?

The second part of your experiment will require a greater understanding of 
naming and routing in ICN than I currently have.  I invite our Cisco and PARC 
colleagues to join this conversation to move the experimental design along.

Thanks again for breaking the ice!

-steve

> On Jan 23, 2017, at 5:45 PM, Alex Feltus 
> <>
>  wrote:
> 
> Hello ... everyone:
> 
> I am a genetics/bioinformatics associate professor at Clemson University 
> who sees tera/petascale biocomputing as the new normal in a few years.  
> Thus, I am also a cyberinfrastructure developer/user to help enable this 
> future for my research and facilitate that of others (e.g. 
> https://nsf.gov/awardsearch/showAward?AWD_ID=1443040).  
> 
> There are many things I would like to do, but based upon my limited 
> understanding of ICN, I would like to:
> 
> 1. After the hardware is installed, a first step would be to move a few TB 
> of Sequence Read Archive (SRA) data from NCBI to the Clemson Palmetto 
> cluster using the ICN network.  I would do this using Aspera and GridFTP 
> and process this data from a systems genetics project.  
> 2. Once I verify that the network functions at respectable flow, I would 
> like to have a discussion with I2/NCBI to discover and move data based upon 
> "named" experimental metadata directly as opposed to a manually curated URL 
> manifest.
> 3. I have many experiments that I could propose once the network is 
> flowing, I know more about ICN, and I know the right people to discuss 
> possibilities.
> 
> Please let me know if I am on/off track with this.
> 
> Alex
> 
> 
> ------
> Alex Feltus, Ph.D.
> CEO, Allele Systems LLC
> Associate Professor, Clemson University Dept. Genetics & Biochemistry
> Biosystems Research Complex Rm 302C
> 105 Collings Street Clemson, SC 29634, USA
> 864-656-3231 (office) 
> LinkedIn . Google Scholar
> 
> 
> 
> -----Original Message-----
> From: 
> 
>  
> [mailto:]
>  On Behalf Of Stephen Wolff
> Sent: Monday, January 23, 2017 3:47 PM
> To: 
> 
> Subject: [icn-rsch] Introduction
> 
> All,
> 
> As you know, the Cisco hardware is shipping, and after basic installation 
> some technical tasks remain - e.g., connection to the AL2S mesh, and API(s) 
> for moving data.
> 
> After all that is complete, and we have an up-and-running ICN testbed, the 
> question is "What do we do with it?"
> 
> I believe it is by no means too soon to start addressing that question, so 
> I would like us to start.  From my relatively untutored viewpoint, I can 
> see three general areas for work:
> 
> *  R&D of the technology and of testbed itself - e.g., 
> fragmentation/reassembly, forwarding, cacheing…  A subset of this area is 
> participation in the efforts to harmonize CCNx and NDN protocols,
> 
> *  Issues surrounding using ICN to move data in our chosen initial domain 
> of science application, genomics, and
> 
> *  Topics in the intersection requiring one foot in network R&D and one 
> foot in the science application (naming, for example).
> 
> To facilitate the discussion, I’ve created the list 
> .
>   If you’re reading this, you’re subscribed(!).  If you know others who 
> should be on the list, please let me know.
> 
> 1.  If your involvement with the ICN testbed is mostly (or purely) 
> custodial or administrative please feel free to unsubscribe or put us on 
> your spam list.
> 
> 2.  If on the other hand you are interested in working on one or more of 
> the “What do we do now?” bullet points above (or any other relevant issue), 
> PLEASE collect your thoughts and send a brief summary to us via this list 
> <>.
> 
> I have no wish to over-organize, to prescribe, to proscribe, or otherwise 
> interfere with testbed activities.  But I *do* want to start a discussion 
> to define and plan some projects “shovel-ready” (to use an old phrase) so 
> that as soon as the testbed is operational we can start using it to repay 
> our sponsors - PARC, Cisco, and Internet2 - for their generous investment.
> 
> Thank you for your participation in the ICN testbed.
> 
> Onward!
> 
> -steve
> 
> stephen wolff
> Principal Scientist
> Internet2
> +1 202 803 8971