ThousandEyes assesses the key performance differences between AWS, Azure and GCP
Plenty of factors have to be taken into account when choosing a cloud provider, from performance to price and everything in between. Indeed, many organisations have gone ahead of this process and are now weighing up more than one cloud provider, assessing which workloads fit best.
When it comes to the combination of price and performance, companies such as Cloud Spectator have shown the differences between the Amazons, Microsofts and Googles of this world and more specialised players. Yet for many organisations, the big three are a must.
ThousandEyes has put together what it claims to be the industry’s first report which measures the global performance of the public cloud behemoths. The report, which analysed more than 160 million data points, found subtle but crucial differences in how they worked.
For one thing, AWS takes a different approach to its brethren when it comes to connectivity. AWS’ traffic only enters its architectural backbone close to the target region; for instance, traffic from Singapore goes through multiple service providers before hitting Amazon’s systems in Dallas.
The reason for this is succinctly explained. “Why AWS chooses to route its traffic through the Internet while the other two big players use their internal backbone might have to do with how each of these service providers has evolved,” the report notes. “Google and Microsoft have the historical advantage. AWS, the current market leader in public cloud offerings, focused initially on rapid delivery of services to the market, rather than building out a massive backbone network.”
In terms of network performance, the research found generally consistent results, although there were exceptions in Asia and LATAM. The report took the cloud behemoth’s Eastern US data centre locations – Ashburn for AWS and Google and Richmond for Microsoft – and compared bi-directional latency between different geographies. Naturally, North America and Europe has the quickest times, with Asia and Oceania lagging, but when it came to fluctuations in latency, NA and Asia suffered the most.
Looking at specific countries, India proved a fascinating case. From each provider’s Mumbai region, Google struggled somewhat – particularly when it came to Europe, where almost three times the latency was experienced. Yet in terms of variance, AWS had particular difficulty when it came to Asia, although all providers recorded poor scores. “Such large swings can impact user experience and most likely corresponds to the relatively poor quality of the Internet in Asia,” the report noted.
The report also explored multi-cloud performance – with results here being consistent. Packet loss was at 0.01% across all relationships – AWS/Azure, Azure/GCP and GCP/AWS. “Multi-cloud performance reflects a symbiotic relationship,” the report explained. “Traffic between cloud providers almost never exits the three provider backbone networks, manifesting as negligible loss and jitter in end-to-end communication.”
“Multi-national organisations that are embracing digital transformation and venturing into the cloud need to be aware of the geographical performance differences between the major public clouds when making global multi-cloud decisions,” said Archana Kesevan, report author and senior product marketing manager at ThousandEyes.
You can read the full report here (email required).
Interested in hearing industry leaders discuss subjects like this and sharing their experiences and use-cases? Attend the Cyber Security & Cloud Expo World Series with upcoming events in Silicon Valley, London and Amsterdam to learn more.