Cloudera Sets Out the Essential Conditions for a Data and AI Platform, Cutting Deployment From Two Months to One Hour
IT DAILY ·
✦ AI Summary
Cloudera presented hybrid and an integrated data fabric as essential conditions for a data and AI platform at EVOLVE26 in Singapore.
Through a demonstration of Cloudera Anywhere Cloud, Sergio Gago CTO and Leo Brunnick CPO showed processes for natural-language analysis, impact assessment, data migration, and drafting a notification email.
The company said it had cut the average time from hardware preparation and internal review to data service operation from 60 days to under 60 minutes.
SINGAPORE — At Cloudera's annual data and AI conference, EVOLVE26 Singapore, held on the 20th local time at the Marina Bay Sands Convention Center in Singapore, Cloudera presented hybrid as an essential condition for a data and AI platform.
At the same event, Cloudera also presented an integrated data fabric as another essential condition for a data and AI platform.
Cloudera then shared the results of applying those conditions, saying deployment time had been cut from nearly two months to under 1 hour. The photo showed Chief Technology Officer Sergio Gago and Chief Product Officer Leo Brunnick in attendance.
Cloudera CTO Sergio Gago and CPO Leo Brunnick gave a demonstration of the new platform, Cloudera Anywhere Cloud, assuming a scenario in which a data center was paralyzed by a cyberattack. Cloudera Anywhere Cloud is a platform for building, deploying, and scaling production-grade AI applications across multi-cloud and on-premises environments.
In the demonstration, when Gago CTO instructed an agent in natural language to analyze the cause, the agent moved between the data center, the cloud, and the database to identify the source of the incident. The agent then identified and reported the financial impact, as well as identified and reported affected customers.
It then moved data and other assets to another cloud instead of the cloud where the outage occurred, and also drafted an email to notify affected customers. Brunnick CPO said such a response was possible only with a data and AI system, data fabric, and guardrails applied with full governance across multiple environments.
As conditions needed when designing a data and AI platform, Cloudera presented seven items: true hybrid, an integrated data fabric, AI-built-in services, modularity, scalability, agentic-first, and faster time to value.
The hybrid requirement means the platform must work the same way across on-premises environments, air-gapped environments, sovereign environments, and public clouds. To do so, code written once must be deployable unchanged across environments, and a self-contained system is also required.
The modularity requirement means individual components must be replaceable or addable without changing the entire platform. In the past, changing a specific service required sequential upgrades to management tools and the data platform, but modularity removes that burden. Brunnick CPO explained that Spark 4 can be downloaded from the Anywhere Cloud marketplace, and that it can also be operated in parallel with existing Spark 3-based workloads.
The company said it cut the average time from hardware preparation and internal review to data service operation from 60 days to under 60 minutes.
Gago CTO then pointed out that while anyone can deploy an on-premises service itself, it is not easy to deploy on-premises with high availability, disaster recovery, autoscaling, and multitenancy. He said the hardest task is building the data fabric that ties everything together, and explained that Cloudera is the one that made data fabric implementation happen.
Brunnick CPO said that one year ago, on this stage, he mentioned that Cloudera would deliver a 2030 data and AI platform in 2026, and that he also set a goal of installation time under 1 hour at the time. He said the reaction then was one of skepticism, but now, one year later, the company has reached its target.
Source: IT DAILY · Yang Seung-gap
Original: https://www.itdaily.kr/news/articleView.html?idxno=241103
References
This article was produced with the help of an automated content generation algorithm.
Source: IT DAILY
View originalThis article was summarized and organized by BizCrush based on the original article from IT DAILY. For exact quotations and full details, please refer to the original article.