AWS Compute
Elastic Compute Cloud (EC2)
AWS Elastic Compute Cloud (EC2) provides stable, secure, and customizable compute resources based on specific needs. Using a simple interface, EC2 can configure server instances quickly, allowing users to efficiently scale capacity as requirements change and pay only for what they needed.
EC2 can be used to run hundreds or even thousands of server instances simultaneously, and provides a range of options for operating systems, server instance types, and software packages. Integrated with many popular AWS services like AWS S3, EC2 utilizes Amazon’s proven network infrastructure and data centers.
Users can also take advantage of products like StarCluster and AWS Batch that work with EC2 to improve efficiency, simplify configuration, and control costs. Learn more about AWS Elastic Compute Cloud (EC2)
AWS Batch
Batching computing is a method of executing jobs without user intervention, making it easier to work with large amounts of data or run hundreds or even thousands of jobs. Properly sequencing and scheduling these programs and allocating the correct compute and storage resources is especially important, since the goal is minimal interaction requirements. And since the resource needs may vary from job to job, it can be difficult to manage costs
AWS Batch is a free service that automatically plans, schedules, and executes batch computing jobs, eliminating the need to install and manage batch computing software to run your jobs. Using AWS Batch can also help keep costs down by dynamically engaging resources based on the requirements of the submitted batch jobs and by automatically bidding on Spot Instances.
AWS Batch is designed to work seamlessly with AWS data storage products and services like S3. Learn more about AWS Batch.
While there is no additional charge for AWS Batch, you will be responsible for costs associated with the AWS resources AWS Batch creates to store and execute your jobs.
StarCluster
StarCluster is an open-source toolkit for EC2 that streamlines and simplifies the configuration and management of virtual machine clusters while helping to control costs.
Users can control network access by configuring a new security group for each cluster, and then letting you apply different firewall rules to the entire group. A simplified naming convention makes managing the nodes more user-friendly. And once users log in to the master node, they can use SSH to connect between any nodes in the cluster without a password.
StarCluster consolidates time-consuming and tedious processes like creating and formatting EBS Volumes, building S3 and EBS AMIs, and copying data with simplified commands.
Users can also manage their costs by using StarCluster to scale their resource use up or down, depending on their immediate needs. StarCluster can automatically bid on spot instances to take advantage of reduced-cost opportunities, and its load balancing feature can help ensure that jobs use only the resources they need at any given time.
And while its default configurations are designed to help get users up and running quickly, StarCluster allows users to further customize their configuration with support for plugins that use StarCluster’s API to interact with the nodes. Learn more about StarCluster.