On Sun, Jun 23, 2019 at 9:17 PM David Anderson <d...@natulte.net> wrote:
> Looks very nice, I look forward to checking it out. > > I'm curious: why do you maintain separate trees for the single node vs. > cluster version? Naively I would have assumed that single node == cluster > codebase with a few features turned off at runtime, and there seems to be a > lot of overlap between the branches. So I'm wondering what the history is :) > Though cluster version and single-node version have a big share of common code, cluster version has non-trivial amount of non-trivial code for RPC aka communications between nodes over the network. Single-node version should fit the majority of users because it is quite fast and scalable <https://medium.com/@valyala/measuring-vertical-scalability-for-time-series-databases-in-google-cloud-92550d78d8ae>. So I decided moving non-trivial code related to cluster version to a separate branch, so users could understand and hack the code for single-node version without the need to investigate and understand the cluster code. -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/golang-nuts/CAPbKnmC0HhJiHj6E7iiR43HJM1o1QmLxWpYGyMKX42GtPddcpg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
