I just ran speedtest.net on a $5/mo 512MB DigitalOcean droplet in the San Francisco (SFO1) datacenter.
~ speedtest-cli
Retrieving speedtest.net configuration...
Testing from DigitalOcean (192.241.229.48)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Monkey Brains (San Francisco, CA) [5.93 km]: 2.132 ms
Testing download
Download: 921.09 Mbit/s
Testing upload
Upload: 705.31 Mbit/s
Can anybody run speedtest-cli[1] on a 512MB LightSail instance to compare network throughput?
The problem with speedtest-cli is that the servers are set up to test domestic accounts, not other servers. You can be lucky and get ~1GBs bandwidth, but just because you don't doesn't mean that it's the fault of the provider. The server specs and speedtest only require a 1Gbit/s port, so you will probably not get higher results than yours anyway.
Network speed should generally not be the issue with AWS, it's disk iops where the non-local SSDs will make a major impact.
Agree speedtest-cli is not perfect, but you can see from my test I got near the 1Gbps that DigitalOcean advertises. I am curious if AWS LightSail even breaks 100Mbps.
In terms of network speed not being important, that's not true. Lots of workloads are network bound not i/o bound (load balancers, web servers, etc).
That's what I meant. The results show that the DO box is fast enough, but a slow result doesn't indicate that you can't saturate traffic. It's really hard to test without real-world traffic, haven't found a reliable way to do so yet.
[1] https://github.com/sivel/speedtest-cli