DigiFarm creates a sustainable future for agriculture and achieves 30% cost savings with OCI

Authored by Kellsey Ruppel, principal product marketing director at Oracle.

The author wants to thank Nils Helset, CEO and founder of DigiFarm, for his contributions.

An image of a combine in a field.
Figure 1: DigiFarm helps farmers get more from their fields.

 

Technology, such as precision agriculture and crop monitoring, can vastly improve outputs and take the guesswork out of when and where to put seeds, fertilizer, and other inputs, but it can’t work if farmers don’t have a clear view of their land.

Precision agriculture and other farming tech are critical to producing enough food to support the world’s growing population. Crop production needs to increase between 60–100% to feed the extra two billion people who might live on the planet by 2050, according to the Institute on the Environment.

To tackle the field boundary piece of this puzzle, DigiFarm’s platform uses neural network models to automatically detect field boundaries using high-resolution satellite data.

Goals for cloud migration

A 15th-generation farmer in Norway, Nils Helset took over his family farm eight years ago and saw firsthand how usable land changes over time because of flooding, erosion, and other factors and how those difficult-to-predict variables affect crop management and production. To manage those variables, Helset created a technology platform to optimize his own crop production, and it quickly evolved into DigiFarm, a business to help other farmers, beginning with a focus on field boundaries.

“Precision agriculture services and in-field analytics must start with accurate field boundary data, whether you’re talking about crop inputs, seeds, crop protection, fertilizer, yield prediction, and so on,” says Helset. “But at the time, existing field boundary data was either not available, inaccurate, or outdated. For example, the US Department of Agriculture drew 32 million parcels of land by hand 13 years ago and then digitized them, and they haven’t been updated since then.”

To provide accurate field boundary data, DigiFarm turned to Oracle and runs its cloud native, precision agriculture platform on Oracle Cloud Infrastructure (OCI), helping farmers and other agribusinesses determine seeded field acre boundaries.

Suite of Oracle products that DigiFarm used

OCI includes all the services needed to migrate, build, and run IT in the cloud, from existing enterprise workloads to new cloud native applications and data platforms. DigiFarm used the following OCI services and technologies:

• Region: An OCI region is a localized geographic area that contains one or more data centers, called availability domains. Regions are independent of other regions, and vast distances can separate them, across countries or even continents.
• Compartment: Compartments are crossregion logical partitions within an OCI tenancy. Use compartments to organize your resources in OCI, control access to the resources, and set usage quotas. To control access to the resources in each compartment, you define policies that specify who can access the resources and what actions they can perform.
• Identity and Access Management (IAM): OCI Identity and Access Management (IAM) is the access control plane for OCI and Oracle Cloud Applications. The IAM API and the user interface enable you to manage identity domains and the resources within the identity domain. Each OCI IAM identity domain represents a standalone identity and access management solution or a different user population.
• Availability domain: Availability domains are standalone, independent data centers within a region. The physical resources in each availability domain are isolated from the resources in the other availability domains, which provides fault tolerance. Availability domains don’t share infrastructure, such as power or cooling, or the internal availability domain network. So, a failure at one availability domain is unlikely to affect the other availability domains in the region.
• Fault domain: A fault domain is a grouping of hardware and infrastructure within an availability domain. Each availability domain has three fault domains with independent power and hardware. When you distribute resources across multiple fault domains, your applications can tolerate physical server failure, system maintenance, and power failures inside a fault domain.
• Virtual cloud network (VCN) and subnets: A VCN is a customizable, software-defined network that you set up in an OCI region. Like traditional data center networks, VCNs give you complete control over your network environment. A VCN can have multiple non-overlapping CIDR blocks that you can change after you create the VCN. You can segment a VCN into subnets, which can be scoped to a region or to an availability domain. Each subnet consists of a contiguous range of addresses that don't overlap with the other subnets in the VCN. You can change the size of a subnet after creation. A subnet can be public or private.
• Security list: For each subnet, you can create security rules that specify the source, destination, and type of traffic that must be allowed in and out of the subnet.
• Dynamic routing gateway (DRG): The DRG is a virtual router that provides a path for private network traffic between VCNs in the same region, between a VCN and a network outside the region, such as a VCN in another OCI region, an on-premises network, or a network in another cloud provider.
• Network address translation (NAT) gateway: A NAT gateway enables private resources in a VCN to access hosts on the internet, without exposing those resources to incoming internet connections.
• Internet gateway: The internet gateway allows traffic between the public subnets in a VCN and the public internet.
• API Gateway: Oracle API Gateway enables you to publish APIs with private endpoints that are accessible from within your network and which you can expose to the public internet if required. The endpoints support API validation, request and response transformation, CORS, authentication and authorization, and request limiting.
• Bastion host: The bastion host is a Compute instance that serves as a secure, controlled entry point to the topology from outside the cloud. The bastion host is provisioned typically in a demilitarized zone (DMZ). It enables you to protect sensitive resources by placing them in private networks that can’t be accessed directly from outside the cloud. The topology has a single known entry point that you can monitor and audit regularly. So, you can avoid exposing the more sensitive components of the topology without compromising access to them.
• Virtual machine (VM): The OCI Compute service enables you to provision and manage compute hosts in the cloud. You can launch Compute instances with shapes that meet your resource requirements for CPU, memory, network bandwidth, and storage. After creating a compute instance, you can access it securely, restart it, attach and detach volumes, and terminate it when you no longer need it.
• Functions: OCI Functions is a fully managed, multitenant, highly scalable, on-demand, Functions-as-a-Service (FaaS) platform, powered by the Fn Project open source engine. Functions enables you to deploy your code and either call it directly or trigger it in response to events. Oracle Functions uses Docker containers hosted in OCI Registry.
• Database system: The database system lets you easily build, scale, and secure Oracle databases with license-included pricing in OCI. You can also use OCI to manage Oracle databases in your data center alongside your cloud databases.
• Object Storage: OCI Object Storage provides quick access to large amounts of structured and unstructured data of any content type, including database backups, analytic data, and rich content such as images and videos. You can safely and securely store and then retrieve data directly from the internet or from within the cloud platform. You can seamlessly scale storage without experiencing any degradation in performance or service reliability. Use standard storage for “hot” storage that you need to access quickly, immediately, and frequently. Use archive storage for “cold” storage that you retain for long periods of time and seldom or rarely access.
• Block volume: With block storage volumes, you can create, attach, connect, and move storage volumes, and change volume performance to meet your storage, performance, and application requirements. After you attach and connect a volume to an instance, you can use the volume like a regular hard drive. You can also disconnect a volume and attach it to another instance without losing data.

Migration path

After evaluating several vendors, Helset and his team selected OCI to develop DigiFarm’s precision agriculture platform because OCI’s bare metal graphic processor units (GPUs) are vital to the kind of machine learning he runs.

“We selected Oracle because of the affordability and performance of the GPUs combined with Oracle’s extensive cloud footprint,” Helset says. “GPUs are very important for training deep neural network models. The higher the GPU performance, the better our models. And because we work in several different countries and regions, we needed the infrastructure to support that.”

After DigiFarm built their model, they can apply that model to any images from any place in the world and it works well using OCI. To make their platform work, DigiFarm first ingests optical satellite images from the European Space Agency from a public bucket in Amazon Web Services (AWS). Those images are then brought into an instance on OCI and placed into OCI Object Storage. DigiFarm next trains its deep neural network model to classify and enhance the resolution of those images in a private subnet on seven Oracle bare metal servers. Using a combination of NVIDIA Tesla A100, V100, and P100 GPUs, DigiFarm applies its inference models to analyze the enhanced images and determine the precise boundaries of a farmer’s seeded acres. Using Oracle bare metal GPUs this way has helped DigiFarm improve performance and reduce costs.

“Our training and inference models are resource-intensive and need powerful GPU and CPU instances,” says Rohit Shetty, head of engineering and infrastructure at DigiFarm. “A typical virtual machine requires software that includes a 10–15% performance tax. GPUs don’t require that underlying software. Our processes run anywhere from 24–30 hours at each stage, so even a 10% improvement in performance means that we’re saving a couple of hours. We’re a small startup, so that time is crucial from both a delivery and a cost perspective.”

A reference architecture diagram for DigiFarm.
Figure 2: DigiFarm’s reference architecture.

 

When the boundaries are delineated, the images are stored in OCI Block Storage, turned into georeferenced maps, and made available to DigiFarm clients. DigiFarm's clients can access their georeferenced maps using Oracle API Gateway, which instructs OCI Functions to pull the delineated images from Block Storage.

So far, DigiFarm has helped approximately 500,000 farmers in 30 countries delineate their seeded field acre boundaries with 92% accuracy. The new images are 40 times more accurate than the existing models.

Results

Since its founding in 2019, DigiFarm has grown from two employees to 58, raised more than €7 million in funding, and delineated more than 450 million hectares of land.

Working with Oracle has enabled Helset to keep costs in check as the business grows. DigiFarm is saving 30–40% running their bare metal GPUs on OCI compared to other vendors’ solutions. That represents $10,000–$12,000 every month, which is significant savings for them.

OCI has also enabled performance improvements because deploying new instances in their region, which has improved efficiency, is easy for DigiFarm. They’ve seen that issue create a bit of a challenge with other vendors because of a lack of hardware or GPU provisions.

The DigiFarm team recently tested its field boundary system outside of Norway and achieved the same high level of accuracy in the US, France, and Brazil.

Next steps

Predicting what the future might bring in terms of food demand and production is difficult, but companies like DigiFarm have the potential to help farmers make data-driven decisions, reduce uncertainty, minimize production costs, and increase crop yields. So far, DigiFarm clients have seen yield potentials. The yield of a cultivar when grown in environments to which it’s adapted, with unlimited nutrients and water and stresses effectively controlled, increase up to 10% and input costs decrease 15%.

The next product in DigiFarm’s pipeline is a crop classification and zoning model that can help farmers implement variable seeding rates, spraying, and yield predictions. These plans can also improve sustainability because an estimated 40% of the world’s fields are currently overfertilized.

DigiFarm believes that using technology to revolutionize how farmers identify, manage, and update field boundaries on a large scale is the most tangible way to build a sustainable future.

For more information on DigiFarm and Oracle Cloud Infrastructure, see the following resources:

• DigiFarm farmers improve crop production with OCI
• Explore other technical case studies from OCI
• Try Oracle Cloud for yourself