What is Softadastra?
Softadastra is a computing research and technology company working toward a long-term goal: making computing infrastructure-independent.
Can software be separated from the infrastructure that executes it?
Softadastra is a computing research and technology company working toward infrastructure-independent computing.
We write and maintain Vix.cpp, an open-source C++ developer platform for building modern systems.
Vix.cpp is where much of our developer infrastructure exists today. It gives us a practical foundation for exploring developer tooling, runtime integration and future programming models.
We are now building the Softadastra Runtime, the adaptive execution layer at the center of our long-term computing architecture.
Software should describe what it needs to accomplish.
Infrastructure should determine how to satisfy it.
The Softadastra Runtime is being developed as the adaptive execution layer between software and computing infrastructure.
It observes application requirements, state and available resources, then progressively takes responsibility for infrastructure decisions that applications currently have to make themselves.
github.com/softadastra/runtimeInfrastructure decisions progressively move out of application code and into the Runtime.
The capabilities below describe the technical direction of the Runtime. They are being implemented and validated progressively.
| Runtime area | Observes | Responsibility |
|---|---|---|
| Placement | latency · locality · trust · capabilities | Determine where computation should execute across available infrastructure. |
| Scheduling | priority · load · resources · constraints | Schedule workloads according to application requirements and current resource conditions. |
| Distributed state | durability · consistency · availability | Manage application state independently from any single machine or location. |
| Replication | state · availability · locality | Determine when and where state should be replicated across the computing fabric. |
| Recovery | failure · availability · application state | Recover or resume execution when machines, networks or infrastructure become unavailable. |
| Reconciliation | divergent state · connectivity · consistency | Reconcile state after partitions, offline operation or infrastructure reconnection. |
| Trust | identity · policy · security | Restrict execution and state placement to infrastructure permitted by application policy. |
| Hardware | CPU · GPU · accelerators · capabilities | Select compatible computing resources without permanently coupling the application to specific hardware. |
| Adaptation | network · resources · failures · changing conditions | Re-evaluate execution decisions while the application and its environment change. |
Future reference hardware for infrastructure-independent computing, powered by the Softadastra Runtime.
concept reference hardware
A small, low-power computer designed to run local applications and state, participate in a Softadastra computing fabric, and continue operating when Internet connectivity is unavailable.
Target specification. The first Softadastra Node is a planned reference system. Final hardware may change as prototypes and the Runtime are validated.
| node v1 reference target | |
|---|---|
| Purpose | Reference hardware for infrastructure-independent computing |
| Runtime | Softadastra Runtime |
| Architecture | x86-64 initially |
| CPU class | Low-power 4-core processor |
| Prototype target | Intel N150 class or equivalent |
| Memory | 16 GB |
| Storage | 512 GB NVMe SSD |
| Storage model | Replaceable local storage |
| Networking | 2.5 GbE |
| Wireless | Wi-Fi 6 |
| Internet required | No |
| Offline operation | Local computation + local state |
| Reconnection | Synchronization + reconciliation through the Runtime |
| Fabric role | Local compute node |
| Acceleration | Integrated graphics initially |
| External compute | Edge · cloud · GPU nodes · future accelerators |
| Power input | 12 V DC target |
| System power | < 20 W typical target |
| Power adapter | ≈ 36 W class |
| Cooling | Passive / fanless preferred |
| Noise | 0 dB cooling target if passive design is validated |
| Operating system | Minimal Linux-based Softadastra Node system |
| Security | Secure boot · encrypted state · device identity |
| Runtime memory | < 2 GB base software stack target |
| Form factor | ≈ 120 × 120 × 45 mm target |
| Weight | < 500 g target |
| Display | Headless by default |
| USB | At least 2 × USB 3 |
| Deployment | Desk · wall · office · remote site · edge |
| Price target | ≤ $250 before taxes and shipping |
| Status | PLANNED |
One infrastructure option. Never an infrastructure dependency.
Softadastra is interested in people who want to work on fundamental computing problems, from programming models and runtimes to distributed systems and computing infrastructure.
Many of the questions are difficult. Some are still unresolved. That is precisely why they are worth working on.
If infrastructure-independent computing interests you, work on it with us.
Softadastra is a computing research and technology company working toward a long-term goal: making computing infrastructure-independent.
It means software should increasingly describe what it needs to accomplish instead of being permanently designed around specific machines, cloud providers, networks, locations or hardware.
Modern software is deeply coupled to infrastructure. Developers often have to decide where computation runs, where state lives, how systems recover, how data is replicated and which infrastructure or hardware must be used. Softadastra is researching how more of those decisions can move into the computing system itself.
The Softadastra Runtime is the core execution technology behind Softadastra. It is an intent-driven adaptive runtime designed to progressively connect application requirements, state and constraints to available computing infrastructure.
The Runtime is intended to progressively handle decisions such as computation placement, scheduling, distributed state, replication, recovery, reconciliation, trust and adaptation as infrastructure conditions change.
The Computing Fabric is the pool of computing capabilities available to the Runtime. It may include local machines, edge systems, cloud infrastructure, CPUs, GPUs, accelerators and future computing resources.
No. Infrastructure still exists. The goal is to make more of it an implementation detail of the computing system rather than something every application must be permanently designed around.
No. Cloud infrastructure is only one possible environment. The broader direction includes local computing, edge infrastructure, cloud resources, heterogeneous hardware and future infrastructure.
Softadastra Node is a planned hardware direction: reference computing hardware designed to run the Softadastra Runtime and participate in a Softadastra computing fabric.
No. The Node is intended to be one infrastructure option, not an infrastructure dependency. The Runtime should be able to operate across compatible Softadastra hardware, existing machines, edge systems, cloud infrastructure and future computing resources.
Not as a commercial hardware product. It represents a future hardware direction. Its final architecture, specifications, manufacturing model and availability have not been fixed.
It is Softadastra’s central technological direction. Software expresses intent, state and constraints, while an adaptive runtime interprets those requirements and determines how available computing resources can satisfy them.
Vix.cpp is Softadastra’s open-source C++ developer platform. It provides developer tooling and a foundation for building modern systems while also serving as one path for experimenting with ideas related to Softadastra’s broader computing direction.
Softadastra Engine explores runtime foundations for resilient and offline-first applications. It focuses on systems that remain useful when connectivity is unreliable or unavailable and contributes toward the broader infrastructure-independent computing direction.
Not in the complete form Softadastra is pursuing. It is a long-term research direction. The individual programming models, runtime capabilities, distributed systems and infrastructure abstractions must be developed and validated progressively.
Important parts of Softadastra’s developer infrastructure are developed in the open. Vix.cpp is open source today, and additional components may become open as the research and technology develop.
You can explore the open-source work, follow the research and contribute through GitHub. Softadastra is especially interested in systems programming, distributed systems, runtimes, operating systems, compilers, programming languages, networking, reliability, security and heterogeneous computing.