You plan sounds legit. Basically, Ignite splits your files into chunks of data and puts them into an ordinary distributed cache.
You can find general recommendation on configuring the Ignite cluster in the documentation: https://apacheignite.readme.io/docs/ Regarding IGFS: you are going to need to configure data region size for IGFS cache. You can either configure size of the default data region, or specify a cache name in *FileSystemConfiguration#dataCacheName* property and configure data region for this cache separately. More on data region configuration: https://apacheignite.readme.io/docs/memory-configuration You can also try to tune IGFS-specific parameters, that are available in the *FileSystemConfiguration <https://ignite.apache.org/releases/latest/javadoc/org/apache/ignite/configuration/FileSystemConfiguration.html> * class. Denis вт, 26 июн. 2018 г. в 17:41, matt <[email protected]>: > Hi, > > I've recently started using the Ignite FileSystem (igfs) in our API, to > fully buffer an incoming stream of byte[] values (chunked InputStream). I'm > doing this because that stream then needs to be sent along to another > remote > service, and I'd like the ability to retry without telling the sender to > send again. The thinking is that if this all gets "buffered" into Ignite, > then pulling the "file" out again and sending/retrying should be possible > and present no burden on the original sender. After the file has been > successfully sent, it is then deleted from Ignite -- this all seems to > work, > however, is there a better way? > > If this approach is a good one, I have questions on how to configure. I had > to look around quite a bit to get a working configuration (version 2.3) and > even now, I'm not clear as to what is needed in order to get a good > configuration setup, based on environment/memory/hardware etc.. Is it OK to > just use the default settings? > > > > -- > Sent from: http://apache-ignite-users.70518.x6.nabble.com/ >
