Secure Scalable IoT Computing Platform


Infocomm - Cloud Computing


The widespread adoption of the Internet of Things (IoT) resulted in large scale deployments of connected sensors and controllers. A disparate array of options were available to make the connected systems work. But this also introduced new problems in building any real production level system that is secure and scalable at the same time. It has been reported that the majority of the IoT systems today are highly vulnerable to cyber-attacks. The fast growth and adoption of public cloud systems offered new possibilities. The multitude of sensor types that need to be connected and the complexities of architecting the cloud services has slowed down the transition. Security and privacy are primary concerns when data leaves customers premises to a cloud platform. Scalability is essential for the capacity growth of a service platform in terms of an increasing number of customers and growing customers data. Availability of system round the clock throughout the year while maintaining cost-effectiveness necessitates the need for the right choice of redundancies and application capabilities.

IoT platform primarily based on public cloud services to simplify the process of onboarding any IoT device or system. The platform makes use of existing public cloud services, on-prem private cloud infrastructures, the micro-services architecture and our experience in identifying the most common patterns for specific industrial applications. Security is considered as an intrinsic component of the ecosystem in the design and we have taken thorough steps to make sure that there are no single weak failure points in the complete system.



The IoT platform consists of the server-based services, support for edge devices, tools for log analysis, reporting modules and frameworks to ease end-user application development. The platform can be easily configured through a web portal to meet the requirements of specific customer use cases.

The core services for the platform are composed of various cloud services, keeping a balance between performance, scalability, cost, maintainability and portability. In most cases managed cloud services will be used, even if proprietary, and we prepare a strategy for migration to other providers or to an on-premise system depending on customer requirements.

We offer several options for edge devices, from custom low-power micro-controller based gateways to micro-services based modular platform for Linux based operating systems. These edge options are well integrated with application modules to easily configure, update and even extend the features of the edge devices with close to zero downtime. To maintain security and to simplify key management few bulk provisioning approaches will be available. We also provide SDKs and sample codes to ease the development process.

For application development, the common interfaces are provided as REST and GraphQL APIs. For real-time systems, direct subscriptions to data streams will be provided. Authentication, authorization and enterprise integration will be managed through standard protocols like OAuth and SAML.

 Data Management

Different database services are available in the system that can be attached to different data sources depending on the type of data and processing requirements. For storage of continuously growing data like events or time-series sensor data, a NoSQL (eg, DynamoDB, MongoDB) or Timeseries (eg, InfluxDB) database can be selected. All transaction-related data, where complex queries are involved, can be stored in relational style databases. Static files like Images/CSV/PDF or data for analytics engines will be stored in a flat filesystem, object storage or Content Delivery Networks.

Much of the design follows a stateless approach or a clear separation between stateful and stateless components of the system. This approach along with proper database designs and replication services provides the infrastructure for multi-region deployment.

Serverless, Container-based

Function as a Service (FaaS) is a widely used service in public clouds like Google, Microsoft and other cloud providers or even on modern private cloud systems. These serverless frameworks make it easy to develop the event processing modules as well-defined packages and simplifies the overall maintenance efforts. Many of the core services in our platform is developed as FaaS and new modules can be installed to extend the current features.

Where FaaS is not suitable, a container-based, Kubernetes managed system is available for application deployments. With Kubernetes, it is possible to automate continuous delivery of incremental updates or feature additions to the running production system with minimal disruption.

Log analysis and Alerts

A key aspect of keeping the system reliable and secure is continuous monitoring and analysis of the log messages. With a number of services running at the same time, identifying errors and intervening timely are critical for complex systems. We have built log monitoring and analysis solutions that can trigger alerts on critical errors and easily locate the error points. The solution also includes intuitive visualization and analytics to identify the patterns and to provide the right solutions.










Others which needs service-based platforms.

Market Trends & Opportunities

Considering 100K SME's & large scale solution accounted for 60% of the market size for such technology.

Opportunities to create:

  • Scalable solution
  • Cost-effective compared to current infrastructure
  • A cloud platform maintained individually 


Secure Scalable Solution All-in-One Platform

  1. No Maintenance Cost
  2. No Hardware failure or replacement cost
  3. No Manpower cost
Non-invasive Diagnosis of Internal Combustion Engine Performance
Unified Platform with AI Modules for Management of Smart Estate
Operational Intelligence Platform For Optimization of Water Treatment
Portable AI Food Analyser Detects Food Contamination in Real-Time
Zero Trust Cybersecurity for Enterprise and Cloud
Ising Modelling over Digital Annealer instead of Quantum Computer
Natural Language Process, Machine Learning & Data Visualization