Asset sprawl, siloed data and CloudQuery’s search for unified cloud governance

Obtaining vision – ultimately, visions – in the origins of the cloud for institutions grow more difficult.

Cloud real estate is sprawling and retailer, and the inventory capabilities in the current tools can be narrow and intuitive, and separate elements such as cost and safety data into non -lonely flexible platforms.

Cloud Journal Company Cloudquiery plays itself to address this problem through the centralization of cloud assets, descriptive data for security and cost in one place, and make it accessible through easy and integrated SQL information and reports. The company takes an approach to the first developers of cloud governance, and withdraw data from 60 sources-including AWS, GCP, Azure, OKTA and WIZ-to one data warehouse that can be inquired.

The company now announces a $ 16 million financing round led by Partec to increase the expansion of its cloud vision.

“The biggest challenge in the current tools is that they are horrific – one for safety, the other for cost, and the other for asset stocks – which makes it difficult to get a uniform vision across fields,” said CQ YEVGENY PATS in Venturebeat. “Even simple questions like” What is the size of EBS associated with EC2 that is stopped? It is difficult to answer without assembling multiple tools. ”

(Editor’s note: A discussion episode on “Gen AI’s data quality and beyond – The basis for confidence and performance” at VB Transform this month. Record today.)

Cloudquness under the cover

CloudQuey uses two main techniques within Hood: Data Votation, Clickhouse and Apache Arrow to develop data analysis applications.

The structure of the integrated high -performance program in Go to applications such as AWS, Azure and Google Cloud Platform (GCP) and many other platforms, withdrawing the composition, safety and descriptive data of the cost. The statute continuously synchronizes data from dozens of service providers and cloud services to a central central asset stock.

“We are focusing strongly on the accuracy and maturity of the data, and our adherence to a great frequency to ensure the work of the teams with the most reliable and modern information,” Patz said.

He explained that the data is scientifically organized with the SQL engine from Power Cloudquiery and compact reports, so that the difference can enjoy full flexibility without relying on the black box tools.

Pats said the company “is also selectively uses” large language models (LLMS) to inquire about the natural language and SQL recommendations and its recommendations “but always on the basis of accurate and transparent data.” He pointed out that since artificial intelligence understands SQL well, tools such as Claude and Openai can create custom reports and normal English analyzes.

Patz said that following the first developer approach is very important, because the developers are eventually those who build, operate and secure the cloud infrastructure today. However, many cloud vision tools for governance were built from top to bottom, not for people already in the trenches.

He said: “When you put developers first, with accessible data, flexible and original applications such as SQL, you enable them to move faster, pick up problems early and secure construction.”

Customers find ways to use Cloudqury outside the asset stock. “Many people start clearly, and then grow quickly in using cases such as monitoring compliance, managing the security situation, and improving the cost, all from the same basic platform,” Patz said.

How Hexagon built a data lake without a server for all its cloud stores

One institution that sees the results already is hexagon. The CCOE Center of the Software Company (CCOE) of the software company had a building lake building without a server that could collect data from all its cloud accounts and store it in one lake.

They also wanted the ability to inquire about this data using SQL, portray them with familiar tools (such as Aws Quicksight), and explore the history of its cloud formation over time.

The team built a non -server data pipeline using Cloudqury to collect data from all accounts and store it in the S3. Aws Glue and then accommodates data in Glue DB format in format that Amazon Athena can inquire about, which is what Athens is doing and perceived in Quicksight.

“The existence of a complete server solution was an important demand,” wrote Peter Furgurio, the expert of the clouds, Herman Schaf in a blog post. “This decision brought a lot of benefits because there is no need for a long time and almost zero maintenance.”

They had to overcome some challenges, especially with Amazon S3 support. The CCOE team was among the first to tried Cloudquiery features in the S3 destination and presented visions that lead to new features. These include:

• Parquet Support: Initially support the Cloudquiery file, CSV and JSON data format. The mistakes in JSON interpretations led to the addition of parquet support.
• Data division: The additional component of the CloudQuery file now allows the division to initial writing (not previously available, which leads to additional additional steps).
• ATHENA Resources Offer: Cloudquiery was initially provided only by AWS supplier compatible with Postgres. However, Athens did not support this, so Cloudquier added a function that can recover a list of all tables to create or update the resource display method.

The Figueireo Cloudquier team used to replace the AWS (IPAM) IP headline manager – which has been called expensive and limited in that it does not cover other cloud services providers.

In the end, he runs his Cloudqury in “Data Lake” mode using “very cheap infrastructure” including AWS S3, ECS, Glue, ATHENA and Lambda.

“We can quickly inquire about any IP via the painting and find those who are the owners,” Figuaredo said. “We are now able to collect everything we need at a very low cost with maintenance near zero. This is the sacred cup of our team.”