Oh, so riffing on my last blog post about specialized iteration syntax. for par x.await? in y {} What if we had something similar but for vectorized operations? Speeding up loops depends on heuristics-driven loop vectorizers (e.g. LLVM). What if we had syntax for this?
-
-
I don't really know enough about SIMD operations to make any definitive statements here. But I think this would be intriguing to explore. Rust's borrow checker removes the need for various compiler heuristics by providing complete information. Wonder if loops could be similar.
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Must make sure to pay attention to the universal scalability law, and not distribute things when the cost of distribution + collection (which is surprisingly high) cancels the benefit of parallelization. Basically it only makes sense for rarer throughput-oriented workloads
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.