Is the common depiction of a surface code to be taken literally as a real-space image of the actual hardware?

Question

I'm currently reading the paper "Surface codes: Towards practical large scale quantum computing" and have a couple of very basic questions that if answered will help me contextualize and organize the information in this paper much better. I understand the requisite info for basic QC topics but have the sneaking feeling I'm missing/misunderstanding some implicit abstraction core to the topic of surface codes. And so, would like to double check my intuitions against someone's actual knowledge.

1. Is a surface code an architecture onto which logical qubits, logical operations, and their connections are physically implemented? That is, is the common depiction of a surface code (the pattern of the measure qubits and data qubits) to be taken literally as a real-space image of the actual hardware or does it correspond to some abstraction in software that I'm not quite grasping?

2. Are quantum computers currently designed to compile programs into a surface code implementation or are surface codes still a theoretical framework due to the massive amount of qubits that are needed?

Thank you.

Answer 1

is the common depiction of a surface code (the pattern of the measure qubits and data qubits) to be taken literally as a real-space image of the actual hardware?

Correct. The surface code is physically implemented by a planar grid of qubits.

Left image source

are surface codes still a theoretical framework due to the massive amount of qubits that are needed?

Correct. To do error corrected quantum computation you need on the order of a thousand physical qubits per logical qubit. And your physical qubits must have fewer than 1 error per thousand operations (especially the 2-qubit operations). No one has a hundred physical qubits yet, and the best demonstrated error rates are ~5 in a thousand.