Inspiration

Slower data retrieval from streaming and other cloud based application providers

How it works

The frequently accessed data and files are stored in In-memory
Whenever the user needs a frequently accessed file.We check if the file is present in the geode region.
If present we retrieve it from the region thereby reducing retrieval time and overhead,thereby improving overall performance

For Youtube

We get the most trending videos from youtube and favorites from the users playlist and store it in In-memory database so that the access time is reduced

For Spotify

We get the favorites form the users playlist and users favorite songs from spotify and store it in the In-memory database so that the access time is reduced

For Dropbox/other cloud based service providers

We get the frequently accessed documents by the user and store them in the In-memory database so that the access time is reduced

Challenges I ran into

Getting the geode server up and running.
Integrating Javascript with geode

Accomplishments that I'm proud of

Learned to use In-memory databases

What's next for InDocs

Real time Integration with major cloud based service providers like Spotify,Dropbox,etc

The alternate video link is :http://youtu.be/BLvPcRKJbBU

Share this project:
×

Updates