Self-hosting
Self-hosting is deploying and operating software on infrastructure controlled or managed by the person or organization using it.
Self-Hosting LLMs at Scale
A deployment in which an organization runs and manages a model within infrastructure it controls.
- How it is used
- Frames self-hosting as an infrastructure decision driven by privacy, cost, control, vendor concentration, and regulation.
- How deep it goes
- Defines control boundaries and compares deployment choices, costs, privacy risks, and operating responsibilities; defers detailed sizing and cost calculations.
- Examples
- Uses a law firm’s confidentiality problem and a deployment whose monitoring and logs expose request data.
- What you can do after
- Apply it via vLLM to serve a model, measure it under load, and run a monitored Kubernetes service.
Kafka for Architects
An on-premises solution is software or infrastructure installed and operated within the customer’s own facilities.
- How it is used
- Frames the production deployment choice for a Kafka cluster against managed cloud and hybrid options.
- How deep it goes
- Compares responsibilities and trade-offs, naming operational tasks without providing a step-by-step installation procedure.
- Examples
- Uses a three-broker Customer 360 cluster and an AsyncAPI topic listing with replication settings.
- What you can do after
- Evaluate/understand when Kafka configuration control justifies in-house infrastructure and operational responsibility.
Think Like a CTO
The provision and operation of software on computing infrastructure owned or controlled at the organization's own premises.
- How it is used
- Frames self-hosting as an infrastructure choice within software purchasing and cloud-versus-on-premises decisions.
- How deep it goes
- Defines the ownership model and examines costs, hosting, capacity, and maintenance without giving hardware installation procedures.
- Examples
- Uses steady intensive computing, seasonal demand, and reallocating private-cloud servers for an online promotion to illustrate workload fit.
- What you can do after
- Evaluate/understand it by weighing workload, existing hardware, capacity limits, purchase costs, and ongoing maintenance.
LLM Customization and Fine-Tuning
A deployment arrangement in which systems run within an organization's own infrastructure rather than an external provider's environment.
- How it is used
- Selects a deployment arrangement when privacy, customization, or inference volume may justify infrastructure ownership.
- How deep it goes
- Explains hosting choices, cost comparisons, and operational responsibilities; identifies training tools without repeating the trainer setup.
- Examples
- Compares GPU and API costs for one million requests and lists cloud GPU VMs and on-premises servers as training infrastructure.
- What you can do after
- Evaluate self-hosting against managed platforms using privacy, customization, volume, and maintenance requirements.
Read how LLM Customization and Fine-Tuning explains self-hosting →
Fundamentals of Cybersecurity
On-premises infrastructure is computing infrastructure owned or operated by an organization at its own facilities rather than rented from a cloud provider.
- How it is used
- Keeps an organization's computing infrastructure under its own ownership and operation rather than renting cloud resources.
- How deep it goes
- Defines the model and explains control, staffing, security, compliance, and operational trade-offs without detailing implementation steps.
- Examples
- Uses owning a house versus staying in a hotel to compare on-premises and cloud infrastructure.
- What you can do after
- Evaluate whether direct control and customization justify the staffing, security, cost, and complexity of operating infrastructure.
Read how Fundamentals of Cybersecurity explains self-hosting →
Grokking AI Applications
Self-hosting is operating an application on infrastructure that an individual or organization owns or directly manages.
- How it is used
- Deploys a Langflow app on always-on infrastructure so users and other applications can reach it consistently.
- How deep it goes
- Defines the hardware setup and contrasts its costs and complexity with rented virtual machines and managed hosting.
- Examples
- Uses dedicated hardware for an organization with strict privacy needs as a practical case.
- What you can do after
- Evaluate deployment options by comparing infrastructure control, privacy needs, cost, and operational complexity.
Acing the CCNA Exam: Volume 2 Advanced Networking and Security
On-premises infrastructure is computing and network equipment located within facilities controlled by the organization that operates it.
- How it is used
- Frames infrastructure ownership and facility responsibility before comparing on-premises, colocation, and cloud deployment approaches.
- How deep it goes
- Defines location, ownership, control, and facility obligations, but stops short of configuration or implementation procedures.
- Examples
- Uses company-owned servers and key network devices located within company facilities as the concrete infrastructure example.
- What you can do after
- Evaluate deployment choices by comparing equipment ownership, facility responsibilities, control, and shared-resource access.
Read how Acing the CCNA Exam: Volume 2 Advanced Networking and Security explains self-hosting →
Microsoft Azure in Action
An on-premises system is computing infrastructure operated within an organization's own facilities rather than hosted by a cloud provider.
- How it is used
- Compares self-hosting with cloud infrastructure when choosing how to meet business, security, and data-location needs.
- How deep it goes
- Defines benefits and responsibilities, notes scaling costs, and stops short of covering on-premises operations in detail.
- Examples
- Uses data sovereignty, company-specific security, existing hardware, and a data-center-to-Azure VPN as concrete cases.
- What you can do after
- Evaluate whether data control, security requirements, existing infrastructure, and operational responsibilities support self-hosting.
Rearchitecting LLMs
Self-hosting is operating a model on infrastructure controlled by the organization deploying it rather than relying on an external service.
- How it is used
- Evaluates replacing an agent’s API call with a specialized SLM in production.
- How deep it goes
- Calculates break-even and per-request costs, then calls for latency and task-quality measurements.
- Examples
- Uses an illustrative 2,950-requests-per-hour deployment and hypothetical latency comparisons between an SLM and a frontier API.
- What you can do after
- Evaluate whether request volume and escalation costs make self-hosting cheaper, then compare latency and task quality.
Also covered in
- Platform Engineering on Kubernetes
- Architecting an Apache Iceberg Lakehouse
- Machine Learning Platform Engineering
- The Cloud at Your Service: The when, how, and why of enterprise cloud computing
- Terraform in Depth: Infrastructure as Code with Terraform and OpenTofu
- Amazon Web Services in Action, Third Edition: An in-depth guide to AWS
- The AI Pocket Book
- Build AI-Enhanced Web Apps
- Designing AI SystemsIn early access (MEAP)
- Kubernetes in Action, Second Edition
- Data-Driven Cybersecurity
- Machines that ThinkIn early access (MEAP)
What to know first
- language modelA language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- contractA contract is an agreed specification of the conditions, expectations, rights, or obligations that govern behavior or interactions between parties or software components.
- fine-tuningFine-tuning is the continued training of a pretrained model on targeted data or objectives to adapt its parameters or behavior to a task, domain, or use.
- infrastructureInfrastructure is the foundational physical, digital, and operational resources and systems that support the operation of organizations, computing systems, or services.
Related concepts
- infrastructureuses · Infrastructure is the foundational physical, digital, and operational resources and systems that support the operation of organizations, computing systems, or services.
- cloud computingcontrasts with · Cloud computing is a model for on-demand network access to shared computing resources and services that can be provisioned and scaled as needed.
- hardwareuses · Hardware is the physical equipment and components of a computing system, including processors, memory, storage, and peripherals.
- privacyaddresses · Privacy is the protection of information about people or other sensitive subjects from inappropriate collection, access, use, or disclosure, including control over how it is handled.
- managed APIcontrasts with · A managed API is an interface to a service whose provider operates the underlying software and infrastructure, typically accessed over a network.
Questions about self-hosting
What is self-hosting?
Self-hosting is deploying and operating software on infrastructure controlled or managed by the person or organization using it.
Which Manning books explain self-hosting?
Self-Hosting LLMs at Scale, Kafka for Architects, Think Like a CTO, LLM Customization and Fine-Tuning, Fundamentals of Cybersecurity, Grokking AI Applications, Acing the CCNA Exam: Volume 2 Advanced Networking and Security, Microsoft Azure in Action and 1 more.
What should I know before learning self-hosting?
Manning books that teach self-hosting usually introduce language model, contract, fine-tuning and infrastructure first.
What is the difference between self-hosting and cloud computing?
Self-hosting is deploying and operating software on infrastructure controlled or managed by the person or organization using it. Cloud computing is a model for on-demand network access to shared computing resources and services that can be provisioned and scaled as needed.
What is self-hosting also called?
Self-hosting is also called on-premises, self-host, self-hosted deployment or on-prem.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.








