Glossary
Know the Language.
Definitions for the terms, concepts, and technologies that underpin sovereign infrastructure, AI compute, and Harch Corp's platform.
AI Inference
The process of running trained machine learning models to generate predictions, classifications, or outputs from new input data. Inference workloads are typically latency-sensitive and require optimized GPU or TPU hardware.
API
Application Programming Interface. A set of protocols and tools that allow software applications to communicate with each other. REST and gRPC are common API paradigms used in cloud infrastructure.
Air-Gapped
A security measure where a computer or network is physically isolated from unsecured networks, including the internet. Used in military, financial, and critical infrastructure environments where data must never leave the perimeter.
Cloud Computing
The delivery of computing services — servers, storage, databases, networking, software — over the internet ("the cloud") rather than on local hardware. Models include IaaS, PaaS, and SaaS.
Compute
The processing power provided by CPUs, GPUs, or TPUs that runs applications, models, and workloads. Compute is measured in FLOPS, vCPUs, or GPU-hours and is the fundamental resource of any cloud platform.
Containerization
Packaging an application with its dependencies into a standardized unit (container) that runs consistently across any environment. Docker and Kubernetes are the dominant containerization and orchestration technologies.
Data Center
A dedicated facility housing computer systems, storage, networking equipment, and power/cooling infrastructure. Modern hyperscale data centers consume 20–100MW and serve as the physical backbone of cloud computing.
Data Sovereignty
The concept that data is subject to the laws and regulations of the country in which it is collected or processed. Sovereign cloud ensures data never leaves national borders — critical for government, healthcare, and financial sectors.
DPA
Data Processing Agreement. A legally binding contract between a data controller and a data processor that governs how personal data is handled, stored, and protected. Required under GDPR for any third-party data processing.
Edge Computing
Processing data closer to the source of generation (e.g., IoT devices, sensors) rather than in a centralized data center. Reduces latency, conserves bandwidth, and enables real-time decision-making for remote or mobile applications.
ESG
Environmental, Social, and Governance. A framework for evaluating a company's sustainability and societal impact. In infrastructure, ESG metrics include carbon intensity, water usage, community investment, and board diversity.
Egress
Data flowing out of a cloud network to the internet or another network. Cloud providers typically charge for egress bandwidth, making it a significant cost factor in distributed architectures.
Fog Computing
An architecture that extends cloud computing to the edge of the network, using intermediate nodes (fog nodes) between end devices and the cloud. Enables processing at multiple tiers for latency-sensitive and bandwidth-constrained applications.
Free Tier
A cloud service pricing tier that offers limited resources at no cost, typically for experimentation and prototyping. Free tiers include usage caps on compute hours, storage, and API calls.
GPU
Graphics Processing Unit. Originally designed for rendering graphics, GPUs are now the primary hardware for AI training and inference due to their massively parallel architecture. NVIDIA H100 and A100 GPUs dominate AI workloads.
Green Computing
The practice of designing, manufacturing, using, and disposing of computing devices in an environmentally sustainable manner. Includes renewable-powered data centers, efficient cooling, and carbon-aware workload scheduling.
gRPC
A high-performance, open-source RPC (Remote Procedure Call) framework developed by Google. Uses HTTP/2 for transport and Protocol Buffers for serialization, offering lower latency and smaller payloads than REST APIs.
HarchOS
Harch Corp's proprietary distributed AI operating system. Orchestrates 1,798 GPUs across multiple data centers with the SENSE/THINK/ACT pipeline — providing sovereign compute infrastructure for African and global markets.
Hyperscale
A computing environment designed to scale rapidly to massive capacity — typically exceeding 100MW and hundreds of thousands of servers. Hyperscale data centers are operated by the largest cloud providers and require specialized power and cooling.
IaaS
Infrastructure as a Service. A cloud computing model where providers offer virtualized computing resources — servers, storage, networking — on demand. Customers manage OS, middleware, and applications while the provider manages the physical infrastructure.
Inference
See AI Inference. The production phase of machine learning where trained models process new data to generate predictions. Distinct from training, which requires significantly more compute and is typically batch-oriented.
ISO 27001
An international standard for information security management systems (ISMS). Certification demonstrates that an organization has systematic controls for managing sensitive data and maintaining confidentiality, integrity, and availability.
Kubernetes
An open-source container orchestration platform originally developed by Google. Automates deployment, scaling, and management of containerized applications. The de facto standard for running microservices at scale in cloud environments.
Latency
The time delay between a request and the beginning of a response. Measured in milliseconds, latency is critical for real-time applications. Sub-12ms inference latency enables near-instantaneous AI responses for users across Africa.
LLM
Large Language Model. An AI model trained on massive text datasets to understand and generate human language. LLMs like GPT, LLaMA, and Mistral power chatbots, code generation, and content creation applications.
Load Balancer
A device or software that distributes incoming network traffic across multiple servers to ensure no single server is overwhelmed. Essential for high availability and fault tolerance in distributed systems.
Multi-Tenant
A cloud architecture where multiple customers (tenants) share the same physical infrastructure while maintaining logical isolation. Multi-tenancy reduces costs but requires robust security controls to prevent data leakage between tenants.
Microservices
An architectural approach where an application is composed of small, independent services that communicate via APIs. Each service can be developed, deployed, and scaled independently, enabling faster iteration and fault isolation.
On-Premises
IT infrastructure that is physically located within an organization's facilities rather than hosted in a cloud provider's data center. Offers maximum control and security but requires significant capital investment and maintenance.
OpenAPI
A specification for defining RESTful APIs in a machine-readable format (YAML/JSON). Enables automatic generation of documentation, client SDKs, and server stubs. Formerly known as the Swagger Specification.
PaaS
Platform as a Service. A cloud computing model where the provider manages the runtime environment — OS, middleware, and infrastructure — while developers focus on application code. Reduces operational overhead but limits customization.
PEM Electrolysis
Proton Exchange Membrane electrolysis. A method of producing green hydrogen by using electricity (preferably from renewable sources) to split water into hydrogen and oxygen. More efficient than alkaline electrolysis at smaller scales and faster to ramp up/down.
REST API
Representational State Transfer API. An architectural style for web services that uses HTTP methods (GET, POST, PUT, DELETE) to perform CRUD operations on resources. The most common API pattern for web and cloud services.
RBAC
Role-Based Access Control. A security model that restricts system access based on the roles assigned to individual users. RBAC simplifies permission management by grouping permissions into roles rather than assigning them individually.
Renewable Energy
Energy generated from naturally replenishing sources — solar, wind, hydroelectric, and geothermal. Harch Corp's 2GW+ Renewable Pipeline powers green data centers and green hydrogen production across North Africa.
SaaS
Software as a Service. A cloud computing model where applications are hosted by a provider and accessed via the internet on a subscription basis. Eliminates the need for local installation and maintenance.
SOC 2
Service Organization Control 2. An auditing standard developed by AICPA that evaluates a service organization's controls for security, availability, processing integrity, confidentiality, and privacy. SOC 2 Type II is the gold standard for cloud providers.
Sovereign Cloud
Cloud infrastructure that ensures all data processing and storage occurs within a specific country's borders, subject to that nation's laws. Essential for governments and critical industries that cannot risk foreign jurisdiction over their data.
SENSE/THINK/ACT
HarchOS's three-layer operational pipeline. SENSE ingests real-time data at scale (10M events/sec), THINK processes inference and decision-making, and ACT automates infrastructure responses. Together they form the intelligence loop for sovereign operations.
Terraform
An open-source Infrastructure-as-Code tool by HashiCorp that enables declarative provisioning of cloud resources. Terraform manages infrastructure through configuration files rather than manual processes, enabling version control and reproducible deployments.
Throughput
The amount of data processed or work completed in a given time period. Measured in requests per second, FLOPS, or tokens per second for AI workloads. High throughput is essential for batch processing and cost-efficient inference.
Vertical Integration
A business strategy where a company controls multiple stages of its supply chain. Harch Corp's vertical integration — from renewable energy generation through data center operations to AI compute delivery — eliminates intermediary costs and ensures quality.
Virtualization
The creation of virtual versions of computing resources — servers, storage, networks — using software rather than physical hardware. Enables multiple virtual machines to run on a single physical machine, improving utilization and reducing costs.
Water Desalination
The process of removing salt and minerals from seawater to produce fresh water. Reverse osmosis and thermal distillation are the primary methods. Harch Water operates 200M m³/year desalination capacity across North and West Africa.
Webhook
An HTTP callback that sends real-time notifications when a specific event occurs. Webhooks enable event-driven architectures by pushing data to subscribed endpoints, eliminating the need for polling.
Documentation
Go Deeper
The glossary covers the concepts. The documentation covers the implementation. Explore guides, API references, and quickstarts for HarchOS, SENSE, and ACT.