Dell Technologies World 2022: Liqid ThinkTank Simplifies CDI for AI and HPCMay 4, 2022
Liqid, the world’s leading software company delivering data center composability, announced it is showcasing Liqid ThinkTank, the first fully integrated composable disaggregated infrastructure (CDI) system for the world’s most challenging next-gen workloads, exclusively for IT leaders in attendance at Dell Technologies World 2022.
For data center technologists and decision makers who want to simplify their organizations’ adoption of CDI for artificial intelligence and machine learning (AI+ML), HPC, edge computing, and other applications that require maximum efficiency and performance, the GPU-centric solution offers turnkey simplicity, industry-leading performance, impressive total cost of ownership (TCO) savings, and meaningful reductions in carbon emissions.
“As organizations seek to incorporate AI+ML as part of their overall digital strategy and operationalize it day-to-day, IT teams have been quick to realize that traditional infrastructures often lack the flexibility to consistently support proof-of-concept deployments, much less scale to align with an organization’s vision for AI in the enterprise,” said Luke Quick, Vice President, OEM Sales, Liqid.
“Fortunately, organizations that wish to accelerate time-to-value for their AI+ML programs can now hit the ground running with Liqid ThinkTank. Our newest, adaptive CDI solution is built for GPU-intensive computing to significantly accelerate time-to-value for AI-driven business solutions, while serving as a powerful backbone in a greater AI ecosystem that is well positioned to organically adapt to tomorrow’s challenges.”
Liqid ThinkTank Solutions Deliver Form, Function, and Flexibility for Today’s AI and Future Innovation
According to Gartner, AI use in the enterprise has tripled in the past two years, and by 2025 AI will be the top driver of infrastructure decision making as the AI market matures, resulting in a 10x growth in compute requirements over that same period.
The deployment of accelerator resources such as GPU, FPGA, high-performance NVMe storage, persistent memory solutions such as Intel Optane, and incremental advances in applications and industry standards can only partially meet demand associated with the required infrastructure expansion.
Liqid Matrix CDI software and intelligent fibers, Liqid delivers the flexibility to quickly compose precise amounts of accelerator resource into host servers. Underutilized GPU, FPGA, NVMe, persistent memory, and other accelerator technologies can be pooled and directed to servers in the exact quantities required to satisfy changing workload needs, enabling never-before-seen levels of efficiency in the datacenter. Once a task is complete, resources are released back into the pool for use by other applications.
In addition, with 2-3x improvements in data center utilization rates compared to traditional static architectures, Liqid ThinkTank and other solutions based on Liqid Matrix CDI software can be incorporated as a fundamental element of a more sustainable data center ecosystem with the potential to significantly reduce energy consumption and saving millions of gallons of water and tons of global CO2 emissions per year.
Liqid ThinkTank is available today.