managed objects/nodes, reconciliation frequency, scheduled jobs, API activity and worker concurrency Time the job and identify CPU, storage or external-service waits before replacing hardware.
Imports, processing and simultaneous users.Hardwareanforderungen für Backend.AI: RAM, CPU, Speicher, Control Plane und verwaltete Workloads
Planen Sie RAM, CPU, persistenten Speicher und Service-Abhängigkeiten für Backend.AI – mit einer an die Workload angepassten Dimensionierung von ZimaOS, Backups und praktischen Hardware-Empfehlungen.
Backend.AI hardware requirements at a glance
Backend.AI transforms GPU complexity into operational simplicity. Our container-level GPU virtualization maximizes performance while minimizing costs. The platform supports the entire AI lifecycle from data analysis to training and inference, streamlining AI development for businesses of all scales. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
- RAM
- The upstream README recommends 8 GB or more for development. This is not a complete production or GPU-worker minimum.
- CPU
- The development guidance recommends 4 or more CPU cores. Model jobs, agent workers and accelerators require independent sizing.
- Storage
- Include development dependencies and, separately, selected model/dataset artifacts, job workspaces and persistent service state.
- Required services
- Follow the supported upstream runtime and complete Backend.AI deployment topology for the selected release. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
- Main sizing variable
- managed objects/nodes, reconciliation frequency, scheduled jobs, API activity and worker concurrency
- Practical Zima starting point
- Conditional lab, data-storage or supporting-service hardware only; a complete production deployment requires a separately verified host meeting the upstream requirements. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
From official requirements to the right setup
Plan the complete Backend.AI workload before selecting a server.
-
Official requirements
Choose the exact edition, stable release and supported container/runtime. Follow the supported upstream runtime and complete Backend.AI deployment topology for the selected release. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
-
Confirm your needs
Test a representative busy operation: managed objects/nodes, reconciliation frequency, scheduled jobs, API activity and worker concurrency Record peak resident memory, CPU, disk latency and completion time, including other containers.
-
Leave room to grow
Place active state on reliable persistent storage. Persist declarative configuration, credentials, controller state, workflow history and recoverable management state. Budget working space for an upgrade or import and test recovery before relying on the deployment.
-
Run it on ZimaOS
Follow the supported upstream runtime, operating system and complete service topology for the selected release. On ZimaOS, consider Custom Install only when a maintained compatible amd64 image and every required host capability have been verified. Otherwise use a separately supported Linux environment; this guide does not certify an unverified container deployment. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
Check every playback client
- Exact edition and stable release
- Supported upstream runtime and architecture
- All required companion services
- Realistic peak activity
- Persistent data and file ownership
- Free host ports and reachable URLs
- Working space and backup retention
- Successful database/file restore
Official minimum requirements
Reviewed November 11, 2026 against the linked upstream deployment documentation. Numerical values retain their stated minimum, recommendation or example scope; select release-compatible instructions before installation.
AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker. Numerical examples above retain their stated deployment scope; co-hosted services and independent backups need additional capacity.
| Requirement | Official minimum | What this supports |
|---|---|---|
| RAM guidance | The upstream README recommends 8 GB or more for development. This is not a complete production or GPU-worker minimum. | Official guidance keeps its original test, recommendation or example scope. |
| CPU guidance | The development guidance recommends 4 or more CPU cores. Model jobs, agent workers and accelerators require independent sizing. | Core counts do not predict a fixed user, device or document capacity. |
| Storage planning | Include development dependencies and, separately, selected model/dataset artifacts, job workspaces and persistent service state. | Installation space is separate from retained content, working files and backups. |
| Service topology | Follow the supported upstream runtime and complete Backend.AI deployment topology for the selected release. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker. | Include every stateful companion service in the host budget. |
| Persistent state | Persist declarative configuration, credentials, controller state, workflow history and recoverable management state. | An image rebuild or container replacement is not a backup. |
| Deployment boundary | AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker. | Check the linked project instructions for the selected edition and release. |
When to upgrade your hardware
Use observed Backend.AI workload pressure to decide when an upgrade is warranted.
Busy operations exceed the available window
The complete stack runs short of memory
Persistent data approaches available capacity
Inspect the application and each required service for OOM events, sustained swap or growing queues. Reduce unnecessary concurrency before adding capacity.
Shared hosts and background work.Persist declarative configuration, credentials, controller state, workflow history and recoverable management state. Track primary state and backup retention separately.
Long-lived installations and growing collections.Plan hardware growth with confidence
Address storage, memory, compute and deployment limits separately.
Use suitable active storage
Put latency-sensitive application/database state on durable local SSD storage and keep recoverable copies.
SATA or PCIe SSD according to measured data growth.Add memory for the complete stack
Choose a higher-RAM host when peak application and companion-service usage leaves inadequate OS headroom.
ZimaBoard 2 832 or ZimaBoard 2 1664 according to the measured constraint.Separate capacity from compute
A multi-drive system helps when retained content and backup storage are the problem; it is not automatically a faster application.
ZimaCube 2 when a broader multi-drive workload justifies it.Fix the observed bottleneck
AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
Check service configuration, I/O and network behavior before selecting a stronger CPU.Can it run on ZimaOS?
Follow the supported upstream runtime, operating system and complete service topology for the selected release. On ZimaOS, consider Custom Install only when a maintained compatible amd64 image and every required host capability have been verified. Otherwise use a separately supported Linux environment; this guide does not certify an unverified container deployment.
Verify the supported upstream deployment
AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker. Follow the supported upstream runtime, operating system and complete service topology for the selected release. On ZimaOS, consider Custom Install only when a maintained compatible amd64 image and every required host capability have been verified. Otherwise use a separately supported Linux environment; this guide does not certify an unverified container deployment.
Review upstream deployment guidance ↗Use the official Backend.AI deployment instructions
Keep the selected stable release, persistent paths and companion services aligned with upstream documentation.
Review Backend.AI official installation ↗Check workload-specific deployment details
AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
Review Backend.AI deployment details ↗Which Zima hardware fits Backend.AI?
Conditional lab, data-storage or supporting-service hardware only; a complete production deployment requires a separately verified host meeting the upstream requirements.
Which workload best matches your Backend.AI deployment?
Conditional lab or supporting storage use only
Conditional lab, data-storage or supporting-service hardware only; a complete production deployment requires a separately verified host meeting the upstream requirements. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
- Supporting hardware onlyZimaBoard 2 832
These are practical workload recommendations, not official app requirements, compatibility certification, measured capacity limits or performance guarantees. Include all co-hosted services and independent backups.
| Zima hardware | Best for | Example workload | Core configuration | Recommended boundary | Next step |
|---|---|---|---|---|---|
| ZimaBoard 2 832 | Conditional lab or supporting-service role | managed objects/nodes, reconciliation frequency, scheduled jobs, API activity and worker concurrency |
|
Conditional lab, data-storage or supporting-service hardware only; a complete production deployment requires a separately verified host meeting the upstream requirements. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker. | Get Now |
| ZimaBoard 2 1664 | Conditional lab or supporting-service role | Follow the supported upstream runtime and complete Backend.AI deployment topology for the selected release. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker. |
|
Conditional lab, data-storage or supporting-service hardware only; a complete production deployment requires a separately verified host meeting the upstream requirements. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker. | Get Now |
What the Press Says
Highlights from trusted reviewers worldwide.
“ZimaCube 2: Not just another NAS, tested with 25TB storage, local AI agents, 4K transcoding, and real homelab workflows.”Read full review
“The ZimaBoard 2 is a compact x86 server board that can be turned into a mini NAS, home server, media box, or self-hosting hub.”Read full review
“ZimaCube 2: A modern, high-performance NAS with plenty of room to grow—built for users who want more than basic storage.”Read full review
“Coverage focused on ZimaCube 2's open hardware foundation, no monthly fee, and self-hosting flexibility.”Read full review
Loved by the Community
Stories and reviews from people who build with Zima every day.
★★★★★
Zima Blade Little yet Powerful
Maybe I am not digital natives but I live with PCs since 12 years old in 1984 when IBM PC clone come to my home. Many years have passed and many operating system I've tried. For me Zima blade and CasaOS was a quantum leap for home PC enthusiast and server lab machine to make me stay curious and relevant for this era.
★★★★★
Very good!!
I use ZimaCube Pro as 5th Proxmox cluster node. It runs several VMs and containers, including a VM with GPU passthrough to run a self-hosted LLM. A specific LXC container runs a Samba server for NAS capabilities using four of six RAID 6 SATA HDDs with ZFS.
★★★★★
Great innovation for mini server!
It is very useful and makes a powerful mini server for many purposes, including university and college students in engineering and electronics. Thank you so much for making this server.
★★★★★
Avaliação ZimaBoard 2
Construí um servidor de uso pessoal. O desempenho está muito bom e funciona perfeitamente onde quer que eu esteja. A surpresa é não dependermos de grandes estruturas para termos nosso próprio servidor de dados. Como iniciante, estou gostando bastante do ZimaOS, pois ele é simples e eficiente.
Frequently asked questions
Answers to Backend.AI RAM, CPU, storage, Docker, database and ZimaOS sizing questions.
How much RAM should I plan for Backend.AI?
The upstream README recommends 8 GB or more for development. This is not a complete production or GPU-worker minimum. Include ZimaOS, the application and its companion services rather than reserving the entire machine for one container.
What drives the CPU requirement?
The development guidance recommends 4 or more CPU cores. Model jobs, agent workers and accelerators require independent sizing. The workload to test is: managed objects/nodes, reconciliation frequency, scheduled jobs, API activity and worker concurrency
Which data must survive a container update?
Persist declarative configuration, credentials, controller state, workflow history and recoverable management state. Preserve a consistent database/file set and verify that your recovery procedure restores a working application.
What other services belong in the budget?
Follow the supported upstream runtime and complete Backend.AI deployment topology for the selected release. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
Does this workload need a dedicated GPU?
Accelerator needs depend on the selected inference, synthesis, browser or media backend. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
Can I deploy it on ZimaOS?
Follow the supported upstream runtime, operating system and complete service topology for the selected release. On ZimaOS, consider Custom Install only when a maintained compatible amd64 image and every required host capability have been verified. Otherwise use a separately supported Linux environment; this guide does not certify an unverified container deployment. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
What should I check before upgrading hardware?
AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker. Measure the same representative operation before and after a configuration or hardware change.
Which Zima device is a practical starting point?
Conditional lab, data-storage or supporting-service hardware only; a complete production deployment requires a separately verified host meeting the upstream requirements. AI/GPU orchestration and scheduling resources are separate from training/inference workers. Validate supported accelerators, drivers, cluster/runtime and model memory; none of the Zima hardware choices certifies an accelerator worker.
What sources informed this Backend.AI hardware guide?
This guide was reviewed against the linked project deployment documentation on November 11, 2026. Zima device choices are editorial workload guidance, separate from upstream requirements and measured performance.
