Over the past few months we've been conducting some intercloud network testing. To get started, we first setup testing nodes in 28 different infrastructure/server clouds (a.k.a IaaS). Twice daily at random times (hourly for latency), these nodes initiate throughput and latency network tests with each of the other 27 nodes in the test group. These tests consist of uploading and downloading a 5 MB test file and recording the throughput, and pinging to determine latency. The purpose is to determine which clouds are best connected to each other. In the results tables below, we've compiled averages from all of these tests ordered by downlink throughput and displayed the top 10 best connected clouds to each of the 28 we tested.
As reliable as cloud services generally are, they do occasionally fail. We believe a good cloud adoption strategy involves use of more than one cloud service to avoid risks of extended downtime (i.e. natural disaster, power failures, etc.). Such a strategy could be as simple as maintaining backups in a separate cloud for cold restores, to complex load balanced cross cloud clusters of application servers and data repositories. In both cases, good throughput and low latency between clouds will allow data to be transmitted quickly and efficiently. If latency is high or throughput low, you may run into data consistency issues and performance bottlenecks.
A few admin notes regarding our testing:
- With a few exceptions, most cloud services limit Internet uplinks to 100 Mb/s or less
- A 5MB test file is not sufficiently large to determine accurate throughput capacity between clouds. The downlink and uplink values provided here are for comparison purposes only and not meant to represent actual throughput capacity (which could be even higher for larger data transfers)
- We've found that routing is not always symmetrical between uplink and downlink tests. This is usually the reason for large discrepancies between uplink and downlink throughput
- The results are broken down by cloud. The table shows the 10 best connected clouds for each. The downlink, uplink and latency values are measured by the cloud shown in the left column of the table
These tests are ongoing. Over the next few months, we plan to make this data available in real-time via web services and on our website to include filtering criteria. If you have any suggestions on how we might improve the accuracy of these tests, please feel free to comment.
The Results:
EC2 - US East
EC2 - US West
EC2 - EU West
ElasticHosts
San Francisco, CA
New York, NY
Linode
LinodeLinode - Atlanta, GA
Linode - Dallas, TX
Linode - Fremont, CA
Linode - London, UK
Linode - Newark, NJ
NewServers
Linode - Dallas, TXLinode - Fremont, CALinode - London, UKLinode - Newark, NJNewServers
Well done! I didn't even know that many IaaS providers existed. I hope you plan on creating a basic price list for Iaas providers soon so that we can easily find potential vendors for windows/linux virtual servers.
ReplyDeleteI could be wrong but I think perhaps your use of the word "intra" is not correct, I think you mean "inter-cloud" - between clouds, from one to another as in "interstate highway", goes between states. Also consider "intranet" vs. "internet." Intra-cloud network testing would be between servers in a given cloud, which is not what you've done here, right?
ReplyDeleteGood point, thanks for bringing it up. I guess it depends on how you define cloud. If each service is a separate cloud, then yes, inter-cloud would be the better way of putting it. If all services reside in a single cloud, then intra-cloud is accurate. We tend to envision "the cloud" as a single large entity wherein all cloud services reside. Just a subtle difference in interpretation I suppose :)
ReplyDeleteThanks for your rapid and thoughtful response.
ReplyDeleteHow is a collection of clouds different than the internet at large? How is the connection between say GoGrid and EC2 different than the peering agreement between two pre-cloud managed hosting providers? (Hint: they probably do in fact have a peering agreement).
The network inside a cloud or even inside a data center (e.g. EC2-east) is what's really new and exciting and key to how these clouds will compete with each other. Intra-cloud networking is characterized by free (as in beer), low latency, and high bandwidth (and as secure as you care to make it).
Cheers,
Marc
Hi,
ReplyDeleteThe latency mentioned above is in milliseconds, isn't it?