Public Cloud Providers' Network Performance Wildly Varies (zdnet.com) 14
ThousandEyes, a cloud analysis company, in its second annual Cloud Performance Benchmark, has succeeded in measuring a major performance factor objectively: Public cloud providers' global network performance. ZDNet reports: In this study, ThousandEyes looked at the five major public cloud providers: Alibaba Cloud, Amazon Web Services (AWS), Google Cloud Platform (GCP), IBM Cloud, and Microsoft Azure. It did by analyzing over 320 million data points from 98 global metro locations over 30 days. This included measuring network performance from within the U.S. using multiple ISPs and global network measurements and by checking out speeds between availability zones (AZ)s and connectivity patterns between the cloud providers. Besides measuring raw speed, the company also looked at latency, jitter, and data loss.
First, ThousandEyes found some cloud providers rely heavily on the public internet to transport traffic instead of their backbones. This, needless to say, impacts performance predictability. During the evening Netflix internet traffic jam, if your cloud provider relies on the internet, you will see slowdowns in the evening. So, while Google Cloud and Azure rely heavily on their private backbone networks to transport their customer traffic, AWS and Alibaba Cloud rely heavily on the public internet for the majority of transport, IBM takes a hybrid approach that varies regionally.
What about AWS Global Accelerator? If you pay for this service, which puts your traffic on the AWS private backbone network, will you always see a better performance? Surprisingly, the answer's no. AWS doesn't always out-perform the internet. ThousandEyes found several cases, where the internet performs faster and more reliably than Global Accelerator -- or the results were negligible. For example, ThousandEyes discovered that from your headquarters in Seoul, you'd see a major latency improvement when accessing AWS US-East-1. That's great. But your office in San Francisco wouldn't see any improvement, while your group in Bangalore India would see a performance decrease. Generally speaking, Latin America and Asia have the highest performance variations across all clouds, whereas, in North America, cloud performance is generally comparable. You need to look at ThousandEye's detailed findings to pick out the best cloud provider on a per-region basis to ensure optimal performance. Regional performance differences can make a huge impact. Additionally, the ISP you use and whether or not you're moving traffic in or out of China also affects cloud performance.
For more on the report, see ThousandEyes' website.
First, ThousandEyes found some cloud providers rely heavily on the public internet to transport traffic instead of their backbones. This, needless to say, impacts performance predictability. During the evening Netflix internet traffic jam, if your cloud provider relies on the internet, you will see slowdowns in the evening. So, while Google Cloud and Azure rely heavily on their private backbone networks to transport their customer traffic, AWS and Alibaba Cloud rely heavily on the public internet for the majority of transport, IBM takes a hybrid approach that varies regionally.
What about AWS Global Accelerator? If you pay for this service, which puts your traffic on the AWS private backbone network, will you always see a better performance? Surprisingly, the answer's no. AWS doesn't always out-perform the internet. ThousandEyes found several cases, where the internet performs faster and more reliably than Global Accelerator -- or the results were negligible. For example, ThousandEyes discovered that from your headquarters in Seoul, you'd see a major latency improvement when accessing AWS US-East-1. That's great. But your office in San Francisco wouldn't see any improvement, while your group in Bangalore India would see a performance decrease. Generally speaking, Latin America and Asia have the highest performance variations across all clouds, whereas, in North America, cloud performance is generally comparable. You need to look at ThousandEye's detailed findings to pick out the best cloud provider on a per-region basis to ensure optimal performance. Regional performance differences can make a huge impact. Additionally, the ISP you use and whether or not you're moving traffic in or out of China also affects cloud performance.
For more on the report, see ThousandEyes' website.
Re: (Score:3)
"You can hate him, but you have to admit his hotels are damn fine."
Hilton?
Re: (Score:2)
You can hate him, but you have to admit his hotels are damn fine
No kidding (Score:2)
Re: (Score:1)
And then still take it with a grain of salt.
Internet routes change so often. I love my local ISP and I think they do an excellent job. They virtually never have congestion and seem to have premium routes such that intra and international routes are also virtually congestion free. But the routes can change week to week. Always high quality routes with low latency and loss, but one day I might have a flat 30ms RTT to New York City, and then next will be a flat 35ms.
And because they actively attempt to maintain reliable routes, when something does go
Re: No kidding (Score:1)
Oh gees... (Score:4, Interesting)
Is this news? I tell you I'm shocked!
For anyone vaguely in this area of tech, we've known for a long time that (say) Akamai isn't always as fast as (say) CD Networks, and Fastly isn't really all that fast, Limelight's expensive but good for streaming, and AWS Cloudfront is a commoditised CDN, so not always the best either. If that's news, then just wait until you hear that one provider may be the fastest in your region, but the slowest in another. Oh my! Amazing!
Probably about 8 years ago I met the founder of a company who had figured out that Cloud Provider + CDN provider speed could vary around the world. They constantly 'ping' your app, and using a DNS cname can route traffic through whichever CDN and cloud provider will give the fastest service. They used to even publish the rolled up data so you could see the "leaderboard". It looks like they've been sold to Citrix now though, so I'd imagine are now worse than useless.
Slows down in the evening? (Score:2)
Re: (Score:1)
They discussed per datacenter; so I assume it was per each datacenters local time zone.
Re: (Score:2)
Who gives a shit (Score:1)
The unattended backup with company X needs 7 minutes longer.
I'll commit Seppuku.
Re: (Score:3)