Pliops and DapuStor Join Forces to Advance AI and Data Acceleration
DapuStor SSDs and Pliops LightningAI KV-Cache Solution Combine to Deliver Higher Performance and Scalability for Enterprise AI Workloads
Pliops, a pacesetter in knowledge acceleration know-how, immediately introduced a collaboration with DapuStor to combine its high-performance SSDs into Pliops’ LightningAI KV-Cache offloading answer. This partnership strengthens Pliops’ ecosystem and reinforces its dedication to delivering high-performance, scalable options for AI and data-intensive workloads.
Early enterprise deployments of the built-in answer have demonstrated enhanced efficiency and effectivity, offering measurable worth to clients. By combining Pliops’ KV-Cache offload acceleration know-how with DapuStor’s enterprise-grade SSDs, the collaboration delivers a strong, dependable, and scalable answer for AI and knowledge functions. DapuStor’s robust APAC market presence, together with its succesful gross sales and help groups, additional amplifies the partnership’s influence – enabling quicker adoption and improved time-to-value for patrons globally. In parallel, the Pliops XDP LightningAI Accelerator card has been efficiently validated with DapuStor R6101 NVMe SSDs.
“This validation underscores DapuStor’s dedication to delivering high-performance storage options that completely complement cutting-edge accelerated AI workloads,” stated Alfred Chase Hui, VP of International Business at DapuStor. “The rigorous testing between DapuStor NVMe SSDs and the Pliops LightningAI answer proves their reliability and effectivity in enterprise-grade AI inference environments, such because the vLLM structure.”
DapuStor Corporation (DapuStor), based in April 2016, is a number one skilled in high-end enterprise SSDs, SOC, and edge computing-related merchandise. With world-class R&D energy, it has complete capabilities from chip design and product improvement to mass manufacturing. The firm’s merchandise have been extensively utilized in servers, telecom operators, and knowledge facilities.
As a part of this validation, Pliops examined its accelerated vLLM inference structure utilizing the LightningAI framework, leveraging the Llama-3.1-8B-Instruct mannequin and the vLLM multi-turn benchmark. The outcomes confirmed full compatibility and readiness of the Pliops XDP answer with DapuStor SSDs for AI and inference workloads.
“This collaboration displays our continued give attention to ecosystem collaboration and delivering real-world worth to our clients,” stated Ido Bukspan, CEO of Pliops. “By working intently with companions like DapuStor, we’re enabling next-generation infrastructure that meets the calls for of contemporary AI and knowledge workloads.”
Live SC25 Demos Highlight Performance Gains
Next week at SC25 (Booth #4608), Pliops will host reside demos and “Memory Minutes” – bite-sized classes that spotlight LightningAI’s structure, deployment success tales, and roadmap for rising workloads. The demos will function options optimized for each NVIDIA and AMD GPUs, underscoring LightningAI’s flexibility and efficiency throughout various compute environments.
For extra details about Pliops LightningAI and its rising ecosystem of companions, go to www.pliops.com.
The submit Pliops and DapuStor Join Forces to Advance AI and Data Acceleration first appeared on AI-Tech Park.
