Granica Named a 2023 Gartner® Cool Vendor
Get a demo

Jun 08, 2023

Fusing Fundamental Research with Cloud-Scale Systems Engineering


Cloud is the best place to build modern data centric applications, as it provides scalable and flexible infrastructure and specialized services to process, secure, manage and store data long term. The data itself is becoming more diverse in type and representation, and is getting more centralized in the cloud. Hence, it is quite common for that data to grow into Petabytes in size, with this trend only accelerating over time.

Now with this data growth, many challenges start growing. When using that data, processing time, latency, compute and data access costs all grow, and storage and compliance costs grow when data is at rest. To use the data with existing applications, it may need to be transformed or loaded into other cloud services like database or filesystem. The applications need to become scalable and may need rearchitecting to scale and leverage the right cloud resources. The data becomes more unknown or dark, which means it is less understood, and difficult to guarantee security and compliance.

To add to the above, the pace of innovation is very fast in the cloud, be it the cloud infrastructure, services, or applications such AI/ML, or in new ways of consuming or sharing data. This can make the architecture choices made a few years back, to become stale or inefficient. Custom solutions are being developed in-house or by cloud or enterprises for specific use cases such AI, analytics, database, logging, observability etc. Many of the challenges though are still the same.

At Granica, we are excited to work on these fundamental problems! Our vision is to build a great cloud data platform that fuses the best of research with an efficient and agile cloud native approach.

It starts with research treating the data not just bytes, but as information that is consumed at a huge scale by new applications. The difference between bytes and information is dependent on the application, and the volume of data often contains a lot of redundant information. The research guides on identifying this information and designing the platform that can make this information stored, secured and used in an efficient manner.

When designing the platform, we built a framework to apply continuous research and algorithms to the data, and use that to make the data platform become more efficient over time. The applications, scale of data, or its usage patterns can change and data can get more diverse, and our framework has to learn and evolve as that happens. To do this best, the system is always learning from the live data.

The cloud native platform is a distributed system that is compute efficient, highly available and scalable. With PBs of data, the platform scales to GB/s to 10GB/s of performance. The platform also moves fast with the advancements in cloud, which means great performance, easy to manage, observe, and cost efficiency.

For the application, the platform provides a solid foundation over the cloud resources and aims to reduce the complexities that they manage. It strives to do this in a transparent manner, so they don’t have to be rearchitected; and with better performance, their designs can be simplified.

At Granica, the fusion of research and platform continues to uncover a lot of value for a variety of unstructured data use cases, and we are excited to build on this and provide meaningful outcomes and customer delight!

Want to discuss? Share your comments/questions below: