AWS EC2 Thoughts!
December 14, 2017

AWS EC2 Thoughts!

Anonymous | TrustRadius Reviewer
Score 10 out of 10
Vetted Review
Verified User

Overall Satisfaction with Amazon Elastic Compute Cloud (EC2)

EC2 is leveraged to meet the computing needs of various enterprise digital workloads. It is used by various projects within each LOB across the enterprise. It helps solve the business problems such as lower the TCO, position for greater innovation, faster delivery to market, minimize security risks of public info, greater agility, global launch etc.
  • Dynamic provisioning of resources.
  • Pay for what you exactly use by seconds.
  • Change the selected instance families, instance types under certain situations.
  • Through spot instances, execute your entire workloads for free.
  • Wide range of EC2 choices specifically tailored for needs such as storage, compute intensive, memory intensive, accelerated computing, data analytics etc.
  • EC2 defaults to a limit of 20 instances/region. This default # should increase to 50 and make autoscale mandatory.
  • Latencies amid instances.
  • Latencies amid storage.
  • Faster time to market. Very low infrastructure provisioning time period.
  • 50-70% reduction in TCO of servers.
  • Customer satisfaction is significantly improved.
We have evaluated other public cloud providers such as IBM Bluemix and Microsoft Azure. Our choice has clearly been on EC2 due to the expansive set of features and suitability to different workload types, flexibility around On-Demand instances, Reserved instances and spot instances. AWS EC2 instances and features offered are significantly superior to ones offered by other public cloud providers.
Amazon S3 (Simple Storage Service)
EC2 instances is suitable for all lightweight, stateless cloud ready applications. Applications whose workloads have significant volume peaks and valleys. Cross region communication is not available natively.
EC2 instances are definitely not suitable for heavyweight, stateful applications that require storage of immutable data in its local instance. Not suitable if instance types need to vary significantly.