The new breed of enterprise applications, where massive amounts of corporate and customer data is constantly analyzed in real time, requires a lot of computing horsepower, and those workloads are best served by computing architectures that rely on parallel processing, where a lot of processes happen concurrently.
As it continues its rapid charge on the enterprise IT market, Amazon Web Services wants to make sure companies move as many workloads out of their corporate data centers into its cloud data centers as possible, and workloads that need the most computing muscle are among the ones that are hardest to move to the cloud.
This morning, on stage at the company’s re:Invent conference in Las Vegas, Amazon CTO Werner Vogels announced a new kind of cloud VM designed to make that move easier. The X1 instance, which the company expects to roll out next year, has 2TB of memory and 100 processor cores, enabling the kind of high concurrency Big Data analytics needs.
Powered by Intel’s latest Xeon E7 v3 processors, the instance is a full order of magnitude larger than Amazon’s current-generation high-memory instances, AWS chief evangelist Jeff Barr, wrote in a blog post. It is designed for applications like SAP Hana, Microsoft SQL Server, Apache Spark, and Presto.
AWS will be the first Infrastructure-as-a-Service provider industry wide to offer the might of Intel’s latest Xeon as a cloud service, Diane Bryant, head of Intel’s data center group, said during Thursday’s keynote at the show.
“Cloud computing is … the benchmark for delivering efficient and accessible technology,” Bryant said. But performance is critical for real-time, end-to-end data analytics solutions that can crunch through massive data sets, which means “cloud computing needs extreme performance.”
Public cloud has become big business for Intel, which often designs custom processors for the world’s leading cloud providers, such as Amazon and Microsoft. Last month, for example, Microsoft rolled out the latest DV2 cloud VMs on its Azure cloud, powered by customized Xeon E5 v3 processors.
In addition to announcing its most powerful cloud VMs yet, AWS also announced the tiniest cloud instances it has every announced. T2.nano, slated for availability later this year, has one virtual CPU core and 512 MB of memory.
Designed for applications like low-traffic websites, T2.nano is a “burstable” service, meaning a customer has the ability to burst to higher capacity than the tiny instance offers when necessary, but accumulate CPU credits during low-demand periods.