Distributed systems modelling

Distributed systems

ICT systems ranging from complex SoCs to networks to cryptocurrency blockchains all share the characteristics of being distributed: they comprise physically separated semi-autonomous subsystems that share resources and exchange information. Physical distance imposes inevitable delays in exchanging information; and sharing resources, such as communication channels, brings further constraints including potential exceptions and failures.

A key problem in developing time-critical distributed systems is knowing whether the required performance will be delivered at acceptable cost. Resolving this problem demands an approach that supports both prospective validation and post-facto verification throughout the System Development Life-Cycle. PNSol has developed a novel approach to prospective/post-facto performance and resource consumption validation called the ∆Q framework.

Performance Engineering

Expensive post-deployment upgrades and support headaches can be avoided by a disciplined approach to engineering the performance of the system. PNSol are the world’s only ex-ante network performance engineers, able to work with development teams and system integrators to:
  • Identify which components and code pathways are performance-critical – and which are not;
  • Expose implicit assumptions that would otherwise become performance hazards;
  • Establish performance norms so that customer and end-user expectations can be set correctly, avoiding post-deployment disagreement and disappointment;
  • Quantify tradeoffs between cost and performance;
  • Identify scalability limits.

Benefits for network operators

The inability to reliably predict UX prevents operators and their suppliers from optimising the cost of existing services, and makes the development of new ones expensive and risky. When creating new services, it is essential to know with confidence whether the service is feasible and economic.

In other industries, such as construction, disciplines like structural engineering and quantity surveying reduce risk. We offer comparable capabilities that are missing from the mainstream of network engineering.

By using our modelling approach, we can reduce the cost, uncertainty and risk associated with creating new services or making network structural changes.

Need help with modelling system performance and resource costs? Contact us!