Saurabh Kumar (SK)’s Post

Check out this short interview with insideHPC and our Co-founder and CEO, Bjorn Kolbeck, and learn how our software-defined scale-out storage solution brings its HPC and hyperscaler heritage to AI workloads: https://buff.ly/4bWUd76 #AIStorage #AIWorkloads #HPC #Hyperscaler #Quobyte #SoftwareDefinedStorage #ScaleOutStorage

Powerful article in StorageNewsletter regarding our Hyperscale NAS and GPU Direct Certification announcement...call me and my team 💻 Jeremy Cornelius and Alexander Cheap if you would like a deeper dive discussion... #ai #filesystem #globaldataenvironment #dataorchestration#hyperscaleNAS #tapestorage #gpudirect #nvidia

StorageNewsletter highlights the evolution to Hammerspace's Hyperscale NAS. Check out this thorough overview of our recent announcements - Hyperscale NAS, GPUDirect Certification, and integration of Tape Storage into the Global Data Environment. Storage Newsletter COMMENTS: "This news from Hammerspace is the result of several years of development and maturation and confirms that the company is a key player in file data access methods offering a wide variety of mechanisms. We know that the company offers GDE, pNFS, proprietary parallel mode with RozoFS and more recently S3-to-Tape.... it means that today the firm offers one of the most flexible choices on the market to access file data with de-facto or industry standards." Excellent write up and commentary from Philippe NICOLAS! #ai #filesystem #globaldataenvironment #dataorchestration #hyperscaleNAS #tapestorage #gpudirect #nvidia

Hyperscale NAS represents a significant leap forward in storage architecture, enabling efficient AI training, GPU computing, and other data-intensive workflows at scale. Its fusion of HPC performance and NAS simplicity makes it a powerful solution for modern enterprises.

"Leveraging #HPC Infrastructure for #AI Workloads – The Role of #Kubernetes in AI and HPC" by Jordi Blasco Pallares (CTO / HPCNow! – Do IT Now Alliance) Abstract: As the demand for AI workloads surges, there’s a pressing need to adapt existing High-Performance Computing infrastructures to accommodate the requirements of this new emerging user community. While many HPC users already run AI workloads, the influx of new AI users with little exposure to HPC, demands a more accessible and interactive platform. While Kubernetes is emerging as a solution for AI workloads, significant challenges arise as soon as the AI workloads need to scale. Addressing the needs of this growing user base goes beyond standard Kubernetes solutions. This talk delves into the challenges and solutions to integrating AI workloads with HPC infrastructures. It discusses leveraging existing HPC solutions, such as Warewulf for provisioning Kubernetes alongside Slurm clusters, enabling Slurm to balance the resources effectively for both Kubernetes and traditional HPC workloads based on the load. This talk will also introduce the work done by HPCNow! to enhance performance and efficiency for AI workloads on Kubernetes. Looking towards the future, the talk also examines the development of more mature and suitable solutions for AI and HPC on Kubernetes. #HPCKP24 #HPCKP

Curious about EXAScale and the future of high-performance storage? Check out our latest blog! Delve into the world of All-Flash Systems, comparing performance to traditional HDD-based systems. Outdo the competition with 30X performance per rack unit from DDN's cutting-edge technology! 🏆 Learn why efficiency and robustness at scale are crucial for data centers, especially in power-limited environments. Read here: https://lnkd.in/d4CYuc3c #EXAScale #StorageEfficiency #AI #HPC #DataCenter

#GenAI #Powerscale There is no better UDS= Unstructured Data Storage out there in the market then #Powerscale , I like the way Dell Technologies has positioned itself to meet the #GenAI workloads needs by leveraging Powerscale with stand out features like GPUDirect for ultra-high performance, Client driver for high throughput Ethernet support,Scale-out to scale up and down,Flexibility to support storage tiers

This is a great paper by Marc Staimer examining how the evolution of network attached storage has reached a tipping point driven by the surge in AI/ML workloads for unstructured data. He digs into the need for HPC-class parallel file system performance with the simplicity of Enterprise #nas, and how Hammerspace's Hyperscale NAS architecture enables IT teams to accomplish this and even accelerate the storage they already own. #hyperscalenas #hammerspace #dataorchestration

The new Dell AI/GenAI ready data platfom is out! 🎉 📣 PowerScale’s new NVMe all-flash line up - with GPU direct, non-disruptive scaling, multi-tenant capabilities, universal data access with multi-protocol support, federal-grade security and seamless interoperability with flexible public clouds. All in a 1U node supporting 30TB QLC SED drives! Changing the game for high-speed storage and enabling the most demanding AI & GenAI file workloads. "The true challenge lies not in reaching the peak, but in remaining at the top" #GartnerLeader #DellTechnologies #UnstructuredData #GenAIrevolution

AI will reshape this decade. 🧠🤖🚀 We're future-proofing our business with cutting-edge AI infrastructure. Dell's new PowerScale models are critical for handling the massive growth in data storage related to AI. Itzik Reich 🇮🇱 below provides an in-depth analysis.