Analytics is everywhere and the enterprise is starting to adopt Artificial intelligence (AI). Multiple business functions across industries, from banking and finance to healthcare and media, have deployed ML. The promise of efficiency and profitability gains from faster access to insights from large amounts of data that continues to be collected and stored guarantees that AI implementations will accelerate.
However, the application of AI does come with development and implementation challenges. Cloudera is a company that hopes to remove those obstacles to AI deployment by recently launching a new software-as-a-service-based (SaaS) data lakehouse offering called Cloudera Data Platform One (CDP). In this article, I evaluate that claim against the common obstacles in AI deployment.
SaaS is the key to simplicity
Regular readers will know I am a proponent of SaaS as the easiest “on-ramp” to the cloud. CDP One is a SaaS implementation of Cloudera’s CDP platform, which includes various services, including data ingestion, governance, preparation, lakehouse/streaming analytics, and machine learning.
With a SaaS delivery model, Cloudera manages all the technical issues, which means customers don’t need to call on in-house expertise, which might be in short supply. It relieves the infrastructure burden of sizing, configuring, provisioning, and monitoring clusters and nodes.
SaaS enables companies to migrate data workloads to a cloud-based architecture and take advantage of low-code and streaming analytics tools and machine learning models.
No need for specialized operations
CDP One includes DevOps, SecOps, and CloudOps as part of the service. The term “zero-ops” means no specialized operations required to perform self-service analytics on any data. Zero-ops translates to time and resource savings by not having to navigate, orchestrate, and monitor a highly complex architecture.
Operations carried out in the background include upgrades, infrastructure provisioning, and configuration and monitoring.
Fast and easy open lakehouse
A data lake consists of raw data; the purpose is not yet defined. A data warehouse contains structured, filtered data already processed for a specific purpose.
We all know that data are growing at an incredible rate. Having data distributed across multiple locations, including warehouses and data lakes, adds complexity and cost. The data lakehouse unifies data across data lakes and warehouses to allow data analysis without worrying about the underlying storage format.
CDP One is a data lakehouse SaaS offering with built-in cloud computing, cloud storage, machine learning (ML), streaming analytics, and enterprise-grade security.
Analytic pipelines that won’t slow you down
Analytic pipelines, a set of actions that ingest raw data from disparate sources and move the data to a destination for storage and analysis, contain highly sensitive data. CDP One includes built-in comprehensive security and governance capabilities allowing centrally managed data policies to be consistently applied in line with corporate policy and regulatory requirements.
CDP One also has several security capabilities such as intrusion detection, prevention, and intelligent anomaly detection to ensure that all data is kept secure.
Everything in a single user interface
Data management and machine learning platforms typically have disparate interfaces for managing data at every process step. There could be multiple interfaces, one for adding new data connections, another to build out data transformations, and others for security. Additionally, data governance and machine learning/AI results in several systems that are not linked.
CDP One puts this all under a single UI that allows the end user to manage every process step from one place, reducing complexity.
Get more value from data quicker
There is always a need to access new data sets. Insights gained from AI/ML, visualization, and analytics drive the need to onboard new data into the analytics process. The process traditionally starts with onboarding and transforming the datasets, followed by building analytical models, which can take weeks or months. Setting up proper security, configuring transformation jobs, and moving the models into production is time-consuming.
With CDP One, the process is reduced to one day, potentially a competitive advantage. Users can now be faster and more agile in responding to new ideas and market changes and use data to drive business decisions.
Driving economies of scale
I rarely find customers who have saved money by migrating to the cloud. That said, Cloudera is claiming “20%-35% TCO savings on average as compared to ‘platform-as-a-service’ or legacy on-premises”.
Your mileage may vary, but savings from integrated, out-of-the-box, and fully managed cloud-native data services that include all cloud, security, and monitoring operations seem reasonable.
Other factors include architectural simplification, reduced DevOps overhead, infrastructure costs, additional software licensing fees, automated workload, and infrastructure rightsizing with ongoing support and continuous upgrades and improvement.
But the overriding factor is the value of data. CDP One enables companies to produce products that offer data and analytics to more end users and enables expert developers and low-code data analysts. Users can now build and onboard data sources and analytics with zero-ops.
Finding people with the proper skill set, such as machine learning (ML) modelers and data scientists, can be challenging and expensive. I often hear from companies that lack of internal knowledge is a significant challenge preventing or slowing companies from adopting AI technology.
With CDP One, Cloudera has done an excellent job in removing obstacles to AI adoption. CDP One is simple to use with no operational skills necessary, along with built-in end-to-end cloud-based security.
CDP One is available on request now, with general availability planned for later in the year. The first version is coming out on AWS but based on the company’s relationship with Google and Azure, I’m fully expecting that to be an option soon.
CDP One will appeal to a broad set of companies, including midsized companies or companies that don’t have the necessary engineering resources. Within those new companies, there is also appeal to new personas, such as executives, data science leaders, and data platform owners.
CDP One could be Cloudera’s most significant announcement in the company’s history.
Note: Moor Insights & Strategy writers and editors may have contributed to this article.