The current blockchain in testnet is 30 gigs+. My opinion is that the design of Ethereum may have been over engineering toward a particular goal. The problem they are having is scalability.
The Ethereum team has a lot of smart people involved with it and I have no doubt they will eventually solve this problem I just hope they take a different approach. They are in a position to learn from the Bitshares team, and from the SAFE Network team, both who in my opinion solved the scalability problem in two different ways.
SAFE Network (in my opinion) will scale particularly well. It's an ant colony algorithm which in theory becomes more efficient as you scale it up, because that is how swarm intelligence works in general. They took a biomemetic approach which but their ability to scale depends on certain factors beyond their control which isn't due to their design (such as bandwidth caps).
Bitshares took the approach of studying the best solution, and going for a very specialized kind of scalability.
cally suited for an exchange. Bitcoin similarly is looking like it will go with the Lightening Network which specifically is well suited for transactions.
Ethereum being more general purposed might not benefit from either of these two approaches. At the same time it does handle transactions. All I can say is it's going to take perhaps a month for them to solve these problems which means the launch is probably going to be delayed until after Bitshares 2.0.
I hope the Ethereum team can launch a very scalable network because I want to make use of Ethereum. I think the smart contract functionality makes it special and unique at this time but the longer they take to launch the less of a competitive advantage they'll have.
https://www.reddit.com/r/ethereum/comments/3b1ikz/how_much_space_do_we_need_to_host_a_node/