The 10 Coolest Enterprise AI Platforms Of 2021 (So Far)
As demand surges for enterprise AI software, we’ve rounded some of the top platforms from vendors including AWS, DataRobot, Google Cloud, Hewlett Packard Enterprise, IBM and Microsoft.
Enterprise AI Platforms
As Ritu Jyoti, program vice president for AI research at research firm IDC, puts it, the pandemic of 2020 and 2021 has pushed artificial intelligence “to the top of the corporate agenda.” And that’s leading to a surge in demand for enterprise AI platforms, IDC reports. For 2021, revenue in the global AI market is expected to jump by 16.4 percent, to reach $327.5 billion, according to the research firm. And AI software represents the vast majority—88 percent—of total AI market revenues, IDC says.
“AI is becoming ubiquitous across all the functional areas of a business,” Jyoti said in a news release announcing IDC’s AI market projections. “Advancements in Machine Learning, Conversational AI, and Computer Vision AI are at the forefront of AI software innovations, architecting converged business and IT process optimizations, predictions and recommendations, and enabling transformative customer and employee experiences.”
At CRN, we’ve been tracking some of the top enterprise AI and machine learning platforms, from vendors including AWS, Databricks, DataRobot, Google Cloud, Hewlett Packard Enterprise, IBM, Microsoft and SAS.
What follows is our roundup of the 10 coolest enterprise AI platforms of 2021 so far.
For more of the biggest startups, products and news stories of 2021 so far, click here.
The flagship machine-learning service from AWS, Amazon SageMaker enables developers and data scientists to rapidly build and train machine-learning models and deploy them into production environments. SageMaker offers tools for each step of the machine-learning development lifecycle, including labeling, data preparation, feature engineering, statistical bias detection, auto ML, training, running, hosting, explainability, monitoring and workflows. It provides an integrated Jupyter authoring notebook instance for easy access to data sources for exploration and analysis; common machine learning algorithms optimized to run efficiently against massive data sets in distributed environments; and native support for bring-your-own algorithms and frameworks for flexible distributed training options.
Databricks Unified Data Platform
With an emphasis on scalability, the Databricks Unified Data Platform covers data science, machine learning, analytics and data engineering, and is available on multiple clouds. Databricks allows customers to rapidly experiment and train their models, and then enables quick scaling of the models, as well. The platform offers automanaged and scalable CPU and GPU clusters on multiple cloud platforms, preconfigured with popular machine learning frameworks with built-in optimizations.
Dataiku Data Science Studio
Dataiku’s Data Science Studio provides a single platform for all data science and machine learning tasks, with a focus on multidisciplinary data science teams, collaboration and ease of use. Dataiku caters to customers that have a need for performance metrics that go beyond model accuracy, providing the ability to create custom business metrics optimized to deliver a particular business benefit and to monitor concept drift.
DataRobot Enterprise AI Platform
DataRobot says that its Enterprise AI Platform “democratizes” data science with end-to-end automation for deploying AI applications within an organization. The platform is used to prepare data for machine learning and AI applications; automate the creation of machine learning and time series models; and centrally deploy, monitor, manage and govern production machine learning models. The DataRobot Enterprise AI Platform is available on multiple cloud platforms, as well as in on-premises environments or as fully managed service.
Google Cloud Vertex AI
Google Cloud‘s Vertex AI platform is designed to help developers more easily build, deploy and scale machine learning models with pre-trained and custom tooling within a unified artificial intelligence platform. The managed platform brings together AutoML and AI Platform into a unified API, client library and user interface. It requires almost 80 percent fewer lines of code to train a model versus competitive cloud providers’ platforms, according to Google Cloud. The product allows data scientists and ML engineers across ability levels to implement MLOps to build and manage ML projects throughout a development lifecycle.
H2O Driverless AI
Leveraging automation to rapidly accomplish machine-learning tasks, H20.ai‘s Driverless AI offering is an Automatic Machine Learning (AutoML) platform that aims to provide a full suite of data science capabilities for the enterprise. Key capabilities include automatic feature engineering, which taps into a library of algorithms and to automatically create useful new features for a dataset. Other core capabilities for the H2O Driverless AI platform include model validation, tuning, selection and deployment--as well as time-series, “bring your own recipe” and machine learning interpretability.
Included in Hewlett Packard Enterprise’s Ezmeral software portfolio are several solutions for enabling enhanced AI and machine learning for customers. The HPE Ezmeral ML Ops solution provides “DevOps-like” agility and speed to machine learning workflows, according to HPE. The solution provides support for all stages of the machine learning lifecycle--across sandbox experimentation, model training, deployment and tracking. Meanwhile, the HPE Ezmeral Data Fabric platform (formerly the MapR Data Platform) brings together technologies for data management and data processing to enable data science, analytics and other advanced enterprise data needs.
IBM Watson Studio
IBM Watson Studio, available in the IBM Cloud Pak for Data offering, is a modular platform offering a wide array of enterprise AI capabilities powered by IBM’s Watson technology. The platform is ideal for enterprises that are seeking to run and manage AI models with greater efficiency, while also simplifying their lifecycle management for AI deployments, according to IBM. Ultimately, IBM Watson Studio is a modern solution that “leverages [IBM’s] roots in SPSS, ILOG CPLEX and other earlier products, complemented by a stream of innovations from IBM research.”
Microsoft Azure Machine Learning
With the aim of enabling enterprises to build and deploy models more quickly, Microsoft’s Azure Machine Learning offering is a solution for “all skill levels” using the Jupyter Notebook open document format, drag-and-drop functionality and automated machine learning capabilities. Azure Machine Learning offers “end-to-end” MLOps for creation and deployment of models, leveraging automated workflows, as well as support for open-source technologies including MLflow, Kubeflow, ONNX, PyTorch, TensorFlow and Python. Recent enhancements have included the introduction of Azure Machine Learning managed endpoints, which automates the creation and management of the necessary compute infrastructure.
SAS Visual Data Mining and Machine Learning
As SAS’ integrated solution for solving complicated analytical problems, the SAS Visual Data Mining and Machine Learning offering is a “comprehensive” visual and programming interface that leverages the company’s cloud-enabled, in-memory Viya analytics engine. SAS Visual Data Mining and Machine Learning enables machine learning processes and data mining for users across skill levels, with capabilities including automatically generated insights for identifying common variables across models.