AWS offers a wide selection of EC2 instance types for different compute workloads. Instance types come in different combinations of CPU, memory, storage, and networking capacity that allow you to build a cloud architecture that fit your application needs. You can pick the instance sizes that match your workloads to help you scale your resources.
An earlier article presents a summary of all EC2 instance types. In this post, we will review AWS compute optimized instances to help you choose the right resources to increase cloud performance while lowering the cost.
AWS compute optimized instances are well-suited for what applications?
These instances drive compute intensive applications that run on high performance processors. Examples are batch processing workloads, media transcoding, high performance web servers, scientific modeling, dedicated gaming servers, ad servers and machine learning inference. The later generations are EBS optimized by default.
C5 Instances
C5 instances are the latest generation instances optimized for compute-intensive workloads. They deliver high performance at a lower price.
Use Cases
- High performance web servers
- Scientific modelling
- Batch processing
- Distributed analytics
- High-performance computing (HPC)
- Machine/deep learning inference
- Ad serving
- Highly scalable multiplayer gaming
- Video encoding
C5 and C5n instances are built on Intel processors, while C5a instances run on AMD processors. Bare metal instances, such as C5.metal, provide applications direct access to server processors and memory.
SSD storage performance
The C5d variant is optimized for storage. These instances have local NVMe-based SSD block level storage that physically connects to the host server. They are ideal for media processing applications, such as video encoding and image manipulation, that need access to high-speed, low latency local storage. They are also a good fit for providing temporary storage of data for batch and log processing and applications that need caches and scratch files.Networking performance
C5n instances are ideal for network intensive applications, such as High Performance Computing (HPC) workloads, data lake and analytics software, and network appliances. Network bandwidth of up to 25 Gbps enable C5 and C5d instances to run compute intensive workloads that are not network bound. They provide 33% higher memory footprint compared to C5 instances.What is the difference between C5 and C4 instances?
C5 instances offer up to 50% better price/performance over C4 instances. They pack more memory per vCPU and offer up to 25 Gbps of network bandwidth, more than double that of C4 instances. New C5 and C5d instances in 12xlarge, 24xlarge, and metal sizes run on custom 2nd generation Intel Xeon Scalable Processors (Cascade Lake) with a sustained all core Turbo frequency of 3.6GHz and single core turbo frequency of up to 3.9GHz, significantly improving performance over Haswell-based C4 instances.C6g Instances
C6g instances were recently introduced by AWS to deliver up to 40% better price performance over C5 instances for compute-intensive applications. They are powered by AWS Graviton2 processors based on 64-bit Arm architecture, and EBS-optimized by default.
Use Cases
- High performance computing (HPC)
- Batch processing
- Ad serving
- Video encoding
- Gaming
- Scientific modelling
- Distributed analytics
- CPU-based machine learning inference
Compute optimized instances are a compact set of resources that pack a lot of power to run your applications on AWS. Parquantix monitors EC2 usage in real time to ensure maximum instance utilization.
Contact us for a 30-minute consultation to find out how we can help you.
Optimizing AWS Spend for EC2 Compute Optimized Instances
Cost management is integral to optimizing your AWS architecture. As your compute requirements increase your cloud costs can quickly add up. To control costs you can take advantage special AWS pricing models that provide significant discounts for a volume commitment of instances.
Reserved Instances
As you steadily scale your production environments in the cloud you will benefit from substantial savings through Reserved Instances. You can reserve a set number of compute optimized instances for 1 or 3 years, to receive savings of up to 60% over on-demand rates. While this sounds like a no-brainer, Reserved Instances require that you dedicate time and effort to continuously monitor your workloads so that you can apply reservations to match actual usage for optimum utilization and protect your cost advantage. Otherwise, instances can expire before being used and quickly wipe out your savings.
The cost optimization tool by Parquantix tracks usage to resize your reserved instances with its AI-driven algorithm in real time. It adjusts your RI mix and allocation based on changes in your workloads and business needs. The tool will sell unused instances in the AWS Marketplace before their term expires, to recover costs. You can let the tool to work its magic for you while you focus on building and running your applications on AWS.
Savings Plans
If you commit a specific dollar amount of instance usage over a 1 or 3 year period, Savings Plans offer significant discounts. They can be purchased not just for EC2 instances, but also for other compute resources such as AWS Fargate and Lambda.
While Reserved Instances provide higher discounts than Savings Plans, your application requirements will ultimately decide which pricing model best fits your needs, Reserved Instances or Savings Plans. You can view our article that compares the two pricing plans to better understand the differences.
The table below illustrates how much you can expect to save on compute optimized instances through these volume-based pricing plans. An c5.xlarge running on Linux in US East (N. Virginia) can be purchased for considerable savings over on-demand rates.
Optimize Your AWS Spend with an Automated Tool
Building and running your cloud architecture with the appropriate mix of compute optimized instance types requires a careful consideration of all pricing options. Parquantix can help by formulating a cloud cost management strategy based on your application and business needs. We will put the strategy to work with our automated tool to maximize your workload utilization and performance. Our AI-driven cost optimization tool calculates the best instance architecture to deliver the highest value for your applications in the cloud.
Are you ready to start optimizing your applications with compute optimized instances? Contact us to schedule a 30-minute consultation.
