At its re:Invent conference, Amazon Web Services has unveiled two new instance types for its EC2 service: X1, sporting 2TB of memory, and T2.Nano, aimed to the lower end of compute requirements.
According to AWS’s chief evangelist Jeff Barr, X1 instances will provide “a full order of magnitude larger” memory than current EC2 instances, up to 2TB of memory. X1 instances target enterprise customers and are designed for large in-memory databases, inclusing SAP HANA, Microsoft SQL Server, and Apache Spark; real-time analytics generation; and applications generating huge caches. X1 instances will be powered by up to four Intel Xeon E7 processors, which, Barr says, support high memory bandwidth and large L3 caches.
T2.Nano instances, on the other hand, are aimed to customers that need short bursts of processing power, such as to host dynamic websites, or to run microservices and monitoring systems. Instances from the T2 family provide the ability to accumulate unused CPU credits that can be later used when needed. T2.Nano offers the least performance within the T2 family, amounting to 1 vCPU and 512 MB of memory that can be used at full performance for one hour on full credit balance. According to Barr, T2 instances have been pretty successful and AWS customers are able to run them at full core performance with websites that usually require modest amount of constant traffic while also needing to serve sporadic traffic surges.
X1 instances should become available in the first half of 2016; T2.Nano instances at some point this year. Pricing details have not been disclosed yet.