The announcement: On July 10, Domino Data Lab announced a partnership with SAS Analytics that will let Domino users run SAS Analytics for Containers in the public cloud on AWS while using Domino’s data science platform as the orchestration layer for the infrastructure provisioning and management. This partnership will allow SAS customers to use Domino as an orchestration layer to access multiple SAS environments for model building, deploy multiple SAS applications on AWS, track each SAS experiment in detail, while having reproducibility of prior work.
What does this mean?
Domino customers with SAS Analytics workloads currently running on-prem will now be able to deploy those workloads to the public cloud on AWS by using SAS Analytics for Containers via the Domino platform. Domino plans to follow up with support for Microsoft Azure and Google Cloud Platform to further enable enterprises to offload containerized SAS workloads in the cloud. By running SAS Analytics for Containers via Domino, Domino users will be able to track, provide feedback on, and reproduce their containerized SAS experiments the same way they do so with other experiments they’ve constructed using Python, R, or other tools within Domino.
This partnership was driven by multiple joint SAS and Domino customers that have well-established SAS Analytics workloads in production on-prem. As these workloads spike, spinning up additional on-prem resources is more of a pain point than spinning up similar resources in the cloud. Being able to push the workload up to the cloud provides more flexibility around load balancing, speed, and cost than on-prem servers can usually provide. Accessing these additional resources via Domino provides the added convenience of being able to do so from within a single data science platform environment, and permits data scientists to treat the containerized SAS Analytics like any other model.
Large enterprise clients will often have data science workloads distributed across multiple languages: Python, R, and SAS among them. Each language has its strengths and weaknesses – in particular, SAS code is frequently written in the context of regulated environments. With the above-mentioned SAS workloads in production, the goal is to provide cloud resources to support data scientists leveraging their language of choice. This partnership is intended for SAS customers to use Domino as a data science enabler in conjunction with their existing SAS investments.
In general, enterprises with established on-premises SAS workloads and working on modern analytic modeling and data science projects should consider Domino Data Lab. SAS-using Enterprises adopting Domino will be able to deploy their SAS Analytics workloads to the public cloud on AWS. Shifting this workload to cloud services provides more flexibility around speed and cost than on-prem servers can typically provide to support peak demand.
Amalgam Insights believes that data science platforms able to operationalize a variety of different languages and dedicated workloads will provide an advantage to companies needing to bridge gaps between traditional and modern systems. Large organizations in particular are likely to have departments with multiple language requirements. This Domino partnership represents a step in this direction, with its ability to support both traditional SAS workloads that are embedded into many large enterprises and modern Python and R being used in many modern analytics and data science projects. Given that SAS still has a significant portion of the analytics market workload, Domino supporting SAS in this manner demonstrates a mature approach that treats established SAS Analytics models as valuable and usable resources.
Domino’s partnership with SAS represents a full business partnership including product, engineering, and go-to-market efforts. Amalgam Insights believes that the Domino-SAS partnership represents an important step in providing scalability for existing on-premises SAS workloads and allows data science-savvy organizations with dedicated SAS workloads with the opportunity to integrate some of their most important enterprise analytics with modern data science approaches while providing consistent support for scale, lineage, and governance across all experiments.