Cross-regional
1. Location Hints are just hints, not guarantees.
2. If you’re worried about the latency differences between Eastern and Western Europe, it will be negligible enough to ignore. The way the CF network splits Europe is not as you’d expect, so don’t focus on this too much.
3. This is more “big picture” stuff, but R2 is different from S3 in that we don’t really want you to think in “regions”. The endgame here is to have your data available wherever your users access it from, without you having to guess. Smart placement in Workers is a step in that direction, but this is a much harder problem for blob storage, so it’ll take time. I understand that this doesn’t mean much to you right now though.
4. In the mean time, you can use the CF cache to get around cross-regional data access. Caches are datacenter-local, so in most cases, you’ll be able to either automatically or manually cache your files so they get served quickly. If you absolutely must have your data replicated in multiple places, the best way to do that right now is to have multiple buckets and have copies of your data. I know this isn’t ideal, but it is very rare for someone to want to do this, so you should really consider what you might be getting into. I would naively expect this to only be an enterprise-level concern. If so, you should definitely be talking to a specialist at CF before you decide to do this, at least in my opinion.
5. Why are you trying to test latencies of different regions? If this is about Europe, see 2. Otherwise, geographical distance is a pretty good estimate right now because there are so few regions available.
6. What exactly are you building? I’m not trying to discourage you from thinking about performance, but just making sure that you’re not overthinking this. Pick a region where you expect the most traffic from.
2. If you’re worried about the latency differences between Eastern and Western Europe, it will be negligible enough to ignore. The way the CF network splits Europe is not as you’d expect, so don’t focus on this too much.
3. This is more “big picture” stuff, but R2 is different from S3 in that we don’t really want you to think in “regions”. The endgame here is to have your data available wherever your users access it from, without you having to guess. Smart placement in Workers is a step in that direction, but this is a much harder problem for blob storage, so it’ll take time. I understand that this doesn’t mean much to you right now though.
4. In the mean time, you can use the CF cache to get around cross-regional data access. Caches are datacenter-local, so in most cases, you’ll be able to either automatically or manually cache your files so they get served quickly. If you absolutely must have your data replicated in multiple places, the best way to do that right now is to have multiple buckets and have copies of your data. I know this isn’t ideal, but it is very rare for someone to want to do this, so you should really consider what you might be getting into. I would naively expect this to only be an enterprise-level concern. If so, you should definitely be talking to a specialist at CF before you decide to do this, at least in my opinion.
5. Why are you trying to test latencies of different regions? If this is about Europe, see 2. Otherwise, geographical distance is a pretty good estimate right now because there are so few regions available.
6. What exactly are you building? I’m not trying to discourage you from thinking about performance, but just making sure that you’re not overthinking this. Pick a region where you expect the most traffic from.
Welcome to the official Cloudflare Developers server. Here you can ask for help and stay updated with the latest news
86,529Members
Resources
Similar Threads
Was this page helpful?
