Compute Instances v1.0.0

Learn about SLYD compute instances, how they work, and how to manage them effectively.

What are Compute Instances?

SLYD compute instances are virtual machines powered by LXD container technology. They provide isolated environments for running applications, with configurable resources (CPU, RAM, storage) and full operating system access. Instances can be created on demand, scaled up or down as needed, and managed through the SLYD platform or API.

Instance Types

SLYD offers a variety of instance types to match your specific workload requirements:

Standard Instances

Balanced compute resources for general-purpose workloads.

  • CPU: 2-16 cores
  • RAM: 4-64 GB
  • Storage: 100-500 GB SSD
Use cases:
Web Servers Development Testing Small Databases

Compute Optimized

High-performance CPU configurations for compute-intensive applications.

  • CPU: 8-96 cores
  • RAM: 16-192 GB
  • Storage: 250-1000 GB SSD
Use cases:
Batch Processing Scientific Computing Game Servers CI/CD Pipelines

Memory Optimized

High memory-to-CPU ratio for memory-intensive workloads.

  • CPU: 4-32 cores
  • RAM: 32-512 GB
  • Storage: 250-1000 GB SSD
Use cases:
In-Memory Databases Real-time Analytics Caching Layers Big Data Processing

GPU Accelerated

Popular

Instances equipped with powerful GPUs for accelerated computing.

  • CPU: 8-64 cores
  • GPU: NVIDIA A100, V100, T4
  • RAM: 32-512 GB
  • Storage: 500-2000 GB SSD
Use cases:
AI/ML Training Inference Rendering HPC

Instance Lifecycle

Understanding the lifecycle of a compute instance helps you manage resources effectively.

Create

Configure and provision a new instance

Start

Power on and run the instance

Stop

Shut down the instance temporarily

Terminate

Delete the instance permanently

Instance States

Instances can exist in several different states throughout their lifecycle:

Creating

Instance is being provisioned. You cannot connect to it yet.

Running

Instance is fully operational and accessible.

Starting

Instance is powering on. Wait a moment before connecting.

Stopping

Instance is shutting down. You cannot connect during this state.

Stopped

Instance is powered off but still exists. You are not charged for compute resources (only storage).

Terminating

Instance is being permanently deleted. This action cannot be undone.

Error

Instance encountered an error. Check the instance details for more information.

Creating an Instance

Follow these steps to create a new compute instance:

1

Navigate to Compute Marketplace

Go to the SLYD dashboard and select "Compute Marketplace" from the navigation menu.

2

Choose an Instance Type

Select the appropriate instance type based on your workload requirements.

3

Configure Instance

Specify instance details including name, region, operating system, and resource allocation.

Name: A descriptive name for your instance
Region: Geographic location for the instance
OS: Ubuntu, Debian, CentOS, etc.
Resources: CPU cores, RAM, and storage
4

Select Applications (Optional)

Choose pre-configured applications to install on your instance.

5

Set Rental Duration

Specify how long you want to rent the instance (hourly, daily, weekly, or monthly).

6

Launch Instance

Review your configuration and click "Launch Instance" to begin provisioning.

Pro Tip: Benson AI Recommendations

Not sure which instance type is right for your workload? Use Benson AI to get personalized recommendations. Just describe your project requirements, and Benson will suggest the optimal configuration.

Connecting to Your Instance

SLYD provides multiple ways to connect to your compute instance:

Web Terminal

Access your instance directly from your browser without any additional software.

1 Go to "Active Instances" in your dashboard
2 Click on your instance name
3 Select the "Terminal" tab

SSH

Connect using standard SSH tools for a native terminal experience.

SSH Command
ssh [email protected] -p 22

The SSH credentials are provided in the instance details page.

Web Applications

Access web applications running on your instance through secure endpoints.

1 Go to "Active Instances" in your dashboard
2 Click on your instance name
3 Navigate to the "Endpoints" tab
4 Click on the application URL

Managing Instances

You can perform various management tasks on your instances:

Start

Power on a stopped instance to resume operations.

Find your instance in "Active Instances"
Click the instance menu (three dots)
Select "Start"

Stop

Shut down an instance to pause operations and reduce costs.

Find your instance in "Active Instances"
Click the instance menu (three dots)
Select "Stop"

Restart

Reboot an instance while maintaining its configuration.

Find your instance in "Active Instances"
Click the instance menu (three dots)
Select "Restart"

Terminate

Permanently delete an instance and all its data.

Find your instance in "Active Instances"
Click the instance menu (three dots)
Select "Terminate"
Confirm termination
Warning: Termination is irreversible. All data will be lost.

Monitoring Instances

SLYD provides comprehensive monitoring capabilities for your instances:

Resource Utilization

Monitor CPU, memory, disk, and network usage in real-time.

CPU Utilization
65%
Memory Usage
80%
Disk I/O
32%
Network Traffic
48%

Alerts

Configure alerts for various metrics and events.

Resource Utilization

Alerts for high CPU, memory, or disk usage

Instance State Changes

Alerts when instances stop or encounter errors

Billing Thresholds

Alerts when approaching usage thresholds

Best Practices

Follow these recommendations to get the most out of your SLYD compute instances:

Security

  • Use strong passwords or SSH keys for authentication
  • Keep your operating system and software up to date
  • Follow the principle of least privilege when configuring access
  • Consider using additional security measures like firewalls

Cost Optimization

  • Right-size instances based on your workload requirements
  • Stop instances when not in use to avoid unnecessary charges
  • Consider long-term rental options for persistent workloads
  • Monitor usage regularly to identify optimization opportunities

Performance

  • Choose the appropriate instance type for your workload
  • Monitor resource utilization and scale as needed
  • Use benchmarking tools to measure performance
  • Optimize application configurations for the cloud environment

Data Management

  • Regularly back up important data
  • Use separate volumes for data that needs to persist
  • Implement proper data retention policies
  • Consider encrypted storage for sensitive information
An unhandled error has occurred. Reload 🗙