[Nfd-dev] Large NFD content store

[Junxiao Shi]

Hi Mohammad

On-disk Content Store is not an existing feature, but can be easily
experimented due to NFD's modular design.
Content Store lookup API has been changed to asynchronous, exactly for
supporting an on-disk Content Store.

To design an on-disk Content Store, some considerations are:

* The disk should be faster than network retrieval. SSD is needed.
* Index should be kept in memory. Full Data packets can be stored in disk.
This avoids forwardi delays in case of a cache miss.
* Keep some full Data packets in memory. You'll need an algorithm to decide
when to move full Data packets onto disk or back into memory.

Yours, Junxiao

On Fri, Feb 24, 2017 at 09:39 Mohammad Alhowaidi <malhowaidi at gmail.com>
wrote:

> Hello,
>
> I am working with large amount of data and I was wondering if there is a
> direct way in the current NFD implementation to save the data to the disk
> instead save it to memory due to the data size. OR should I change the NFD
> implementation to do that?
>
> Thanks,
> Mohammad
> _______________________________________________
> Nfd-dev mailing list
> Nfd-dev at lists.cs.ucla.edu
> http://www.lists.cs.ucla.edu/mailman/listinfo/nfd-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/nfd-dev/attachments/20170224/58c4a38d/attachment.html>