Compute Unit
Scalable, on-demand computing resources powering your DiPhyx projects
A Compute Unit in DiPhyx refers to a reserved computing resource that can be allocated on your preferred cloud platform, AWS, GCP, etc., eliminating the need for self-hosted resources. However, you can still use any other computing resources, including your own computer, to create projects and run software on DiPhyx.
This unit is the backbone for executing various computational tasks within the DiPhyx environment, providing scalable and efficient computational power on demand.
Every DiPhyx Project needs computing resources to run, so having a compute unit available—either by creating a new compute unit or accessing a shared one—is the necessary first step.
Equivalent Terms for Compute Units
Different providers use specific terms for the computing resources:
- Amazon Web Services (AWS): AWS uses the term "instance" to refer to the resources it provides for users. An instance in AWS is a virtual server that can run applications and workloads in the cloud, offering scalability and a range of configurations.
- Google Cloud Platform (GCP): GCP uses the term "virtual machine (VM)" for its computing resources. VMs on GCP are versatile and can be tailored to meet specific performance and capacity requirements, enabling users to deploy and manage applications efficiently. The term "VM instance" is also commonly used.
- Microsoft Azure: Azure refers to its computing resources as "virtual machines (VMs)." Azure VMs provide on-demand, scalable computing resources and can be used for various tasks, from development and testing to running critical applications and extending datacenters.
- Self Host: Self-hosting utilizes the computing power of your own computer or other external resources. In other words, you will connect the computing resources of your choice with the DiPhyx platform to run your projects on. This approach does not involve sharing your resources with DiPhyx or other users and offers complete control over the computing environment.
On DiPhyx, when you are creating a compute unit, you are selecting one of these resources depending on the resource provider. The term "compute unit" includes all of these options, giving you flexibility and choice in how you run and manage your projects.
Main Features of Compute Units
Compute units in DiPhyx are designed to optimize efficiency and enhance the learning and work experience for you, your team, or your organization by offering the following features and capabilities:
- On-Demand Resources: You can access computing power exactly when you need it, eliminating the need for maintaining expensive, underutilized infrastructure. This flexibility makes it cost-efficient and ensures that you only pay for what you use.
- Multi-Platform Support: Whether you prefer AWS instances, GCP VMs, Azure VMs, or your own self-hosted resources, compute units on DiPhyx support a wide range of platforms. This flexibility allows you to choose the best environment for your specific needs without being locked into a single provider.
- Scalability: Compute units provide scalable computational power, allowing you to adjust resources based on the workload demand. This ensures that projects of any size, from small simulations to large-scale computations, can be handled efficiently without performance bottlenecks.
- High Performance and Efficiency: Compute units offer high performance, leveraging advanced hardware and optimized configurations to handle intensive computational tasks. This ensures that projects run faster and more efficiently, reducing time to insight and accelerating R&D cycles.
Different Statuses of Compute Units
Once a compute unit is created, it goes through different statuses, each indicating a specific situation the compute unit is in. These statuses include:
- PENDING: DiPhyx sends a request to the provider, and the unit is not created yet. The resources are being provisioned by the provider. Your account is not charged during this status.
- PREPARING: The provider has given access to the requested resources. Software requirements are being prepared, so the unit is not ready to use yet. This is when the compute unit starts to charge your account.
- READY: The compute unit is ready to use. The computing resources are up and running.
- STOPPING: This is the transitioning status when you "stop" a compute unit.
- STOPPED: When the compute unit is STOPPED, it is not charging, but the compute disk is still charged. It is resumable but not available for use in this status.
- TERMINATING: This is the transitioning status when you "terminate" a compute unit.
- TERMINATED: The compute unit has been permanently terminated. The related entry will soon disappear from the compute units table.
- FAILED: This status indicates something went wrong during the process you intended to perform. You should troubleshoot the issue for it to resolve. Regarding billing, if the unit was previously in PENDING status and now FAILED, no credit is subtracted. Otherwise, the billing will follow the previous status.
Pricing
The pricing of compute units depend on two related factors:
- Compute type
- Compute disk