Although most end users never get a clear view of the infrastructure underlying the services they consume via Amazon’s Elastic Compute Cloud, Accenture Research Manager Huan Liu recently estimated that a whopping 454,400 individual blade servers are currently being used to power that product.
In a post to his personal blog, Liu said that he used a combination of publicly available data and DNS queries within EC2 to arrive at the total number of server racks in use by the cloud service, then multiplied that by the number of individual servers in each.
However, the researcher said, there are a couple of obvious caveats to his investigation. The total number of servers in each rack is an estimation on his part, and Amazon may well configure its systems differently than he imagines. What’s more, any rack without an active instance running on it would be impossible to count, throwing off Liu’s accuracy.
Nevertheless, there has been no shortage of media reaction to his post, thanks in part to the fact that it represents one of the best estimates yet of the Amazon cloud’s size. The company does not divulge much information on EC2, making such educated guesses necessary.
Related Articles
Oracle and Amazon Web Services announce strategic partnership designed to simplify migration of enterprise workloads
“My long-term aspiration and ambition for IFS is for us to dominate asset and service” – Mark Moffat
“We can bring simplicity and efficiency to our customers when it comes to adopting AI.” – Ranjith Kaippada, Cloud Box Technologies