SQream Blue’s New Snowflake Connector Boosts Processing Speeds and Reduces Costs by Integrating Data and ML Tasks

Snowflake users can now easily integrate SQream’s GPU parallelizing technology that’s proven to perform 2x faster and at 1/2 the cost with seamless integration to existing data flows

SQream, a leading accelerated data processing platform, today announced the launch of its native connector that brings its cloud data solution, SQream Blue, to the Snowflake environment. This follows SQream Blue’s performance benchmarks that shattered existing performance benchmarks for speed and cost efficiency.

Also Read: HPC-AI Tech Secures $50 Million in Series A Funding to Enhance Video Generation AI and GPU Platforms

The nonprofit Transaction Processing Performance Council (TPC) developed the TPCx-BB as a benchmark for objectively comparing Big Data Analytics System (BDAS) solutions. During the comparison portion of the TPCx-BB analysis, SQream Blue dramatically outperformed Snowflake and its X-Large Virtual Warehouse, handling 30 TB of data 2X faster and at 1/2 of the cost, with some tasks exceeding this even further with a 5X performance improvement.

In conjunction with these results, SQream has introduced a native Snowflake connector that allows SaaS big data projects to seamlessly integrate SQream Blue’s unique GPU parallelizing solution into Snowflake’s data warehouse platform to leverage the best of both platforms.

The launch of the Snowflake connector from SQream enables users to unlock huge cost-performance savings by offloading data loads to SQream Blue without exporting data or migrating information from existing Snowflake workflows.

The new connector uses a massive Snowflake library to establish a direct and optimized connection between SQream Blue and Snowflake. This eliminates the need for intermediate data transfers or complex ETL processes, improves query performance, and reduces operational overhead.

Additional benefits include:

• Seamless Integration: Treat Snowflake tables as native objects within SQream Blue, simplifying data access and management.
• Unmatched Performance: No data export is needed, allowing users to easily take advantage of SQream Blue’s cost-performance over Snowflake-managed data to witness the force of GPU-acceleration.
• Reduce Costs: Optimize resource utilization by offloading specific costly workloads from Snowflake, lowering one’s overall cloud bill.
• Enhance Flexibility: Gain greater control over data movement and processing by leveraging the strengths of multiple platforms.
• Data Efficiency: Minimize redundant data read from Snowflake via a pushdown filter mechanism.
• Available on AWS and GCP Marketplace as a native SaaS solution

“Cost-performance is the new metric for big data analytics in 2024, and SQream Blue’s proprietary GPU based technology has shown that processing and analyzing high volume, unstructured datasets can be done at speed while reducing budget,” said Matan Libis, VP Product at SQream.

“Now, enterprise users can leverage the benefits of Snowflake’s data warehouse platform and complement these with the force of GPU processing thanks to the launch of the native Snowflake plug-in connector,” added Libis.

In its State of Big Data Analytics Report published in June of 2024, SQream found that 92% of companies surveyed are actively aiming to reduce cloud spend on analytics, 71% regularly experience ‘bill shock’, and 41% list high costs as the primary big data challenge.

SQream Blue proved its ability to address this major challenge when processing 30TB of data at breakneck speed during the TPCx-BB analysis, equivalent to reading 25,000 copies of the Oxford English Dictionary in less than an hour.

This highlights how SQream Blue can address the cost-performance challenges associated with modern data analytics and complement other big data solutions in the market for an even higher long-term ROI.

Also Read: Using Generative AI for Decision Intelligence With Pyramid Analytics

[To share your insights with us as part of editorial or sponsored content, please write to psen@itechseries.com]