You’re mixing use cases. StorJ Labs has specifically targeted the distributed storage space, not distributed processing power like SETI. These have hugely different requirements for a ‘node’ as well as expectations for node hardware, internet and speed of response.
StorJ - low CPU, low RAM, large drive, large bandwidth, respond on request
SETI - big CPU, big RAM, small drive, small bandwidth, respond whenever or never
The very nature of a ‘hot’ storage system like StorJ means that the nodes need to be robust, otherwise the data would be lost and customers would leave. Taking a week to respond to a SETI request just means someone else might have been given the job, there’s no penalty and no-one complains.
With 12000 active nodes on StorJ, and an average internet bandwidth of 50Mbps the total network bandwidth is circa 600Gb/s which should be more than enough for existing customers.
Again StorJ labs have designed a system to be compatible with the largest online storage API, so it can be a direct drop-in replacement for Amazon S3 buckets. As has been discussed before they are building the underlying infrastructure but it for others to build the platforms on top of the storage: