Amazon Web Services has further expanded its usage of AMD EPYC-based machines for its Elastic Compute Cloud (EC2) instances. Last week the company started to offer its new EPYC-powered T3a instances, which enable customers to balance their instance mix based on cost and the amount of throughput they require at a given moment.
AWS’s T3a instances offer burstable performance and are intended for workloads that have low sustained throughput needs, but experience temporary spikes in usage. Amazon says that users of T3a get an assured baseline amount of processing power and can scale it up “to full core performance” when they need more for as long as necessary.
T3a instances are offered in seven sizes in the US East (N. Virginia), US West (Oregon), Europe (Ireland), US East (Ohio), and Asia Pacific (Singapore) Regions in On-Demand, Spot, and Reserved Instance form. The specifications look as follows:
This is Amazon’s third announcement of AMD EPYC-powered instances. Previously AWS started to offer M5, R5, M5ad, and R5ad instances based on AMD’s latest server processors.
Related Reading:
- Amazon Offers More EPYC: M5ad & R5ad Instances
- AMD’s EPYC CPUs Now Available on Amazon Web Services
- AMD Previews EPYC ‘Rome’ Processor: Up to 64 Zen 2 Cores
- Naples, Rome, Milan, Zen 4: An Interview with AMD CTO, Mark Papermaster
- CES 2019 Question and Answer Session with AMD CEO, Dr. Lisa Su
Source: Amazon Web Services
from AnandTech http://bit.ly/2WmEVC4
via IFTTT
0 comments:
Post a Comment