This article was first published on null
Some 10 days ago, we published the first part of a Q&A that Piotr, our acting CEO (and all-time CTO) answered for one of our users. As promised, there was a part II (and probably, we will have a third instalment with follow-up questions).
This second part comes right on time as we prepare for the release of Graphene, the LibraryOS we have been building in a working group with Intel, ITL, Chia-Che Tsai and Don Porter. Last week, we published an interview with Mona Vij, R&D lead at Intel, that explained the corporation's relationship with the project. In this post, you can learn why we choose to go down the Intel SGX route, which resulted in Graphene.
Another fascinating topic Piotr covers in his answers is token (GNT) and velocity. As you may know, we want to provide avenues that open up to giving GNT more value (value in the context of the function, and not price) and the question on token velocity is an excellent primer to understand GNT.
Q: What are your thoughts on SGX? Would embracing this technology mean that Golem service providers would be limited to Intel chips only?
PJ: Firstly, I would like to underline that Golem is not limited to Intel CPUs - allow me to explain further below:
The discovery protocol mentioned in answer to question no 2 allows adding new types of resources. This applies to all classes of Trusted Execution Environments (TEEs), including Intel Software Guard Extensions (Intel SGX).
We chose to participate in the development of Graphene mostly because Intel SGX is, by far, the most mature one in terms of offered features and recognizability. The maturity of SGX is essential to us because as stated in question no 6, we are continuously looking for additional solutions to increase ...
To keep reading, please go to the original article at: