Shared Everything

Shared Everything

By Nicole Hemsoth PrickettBusinessNewsTechnology
Download on the App Store

All Shared Everything episodes

  • The Hard Lessons Behind Computing’s Biggest Bets
    Some of the most valuable lessons in computing never make it into the success stories. In this episode, HPC and AI veteran Glenn Lockwood draws on firsthand experience across scientific computing, national labs and hyperscale AI (NERSC, SDSC and Microsoft on its Fairwater supercomputer) to revisit ambitious systems that broke new ground, while revealing the technologies, assumptions and architectural bets within them that didn’t work quite as expected. From attempts to make distributed computing complexity disappear, to a high-performance storage technology users largely chose not to use, to enormous AI infrastructure bets overtaken by changes in how frontier models are built, these are remarkably candid stories from inside consequential computing projects. The failures expose lessons about complexity, adoption, general-purpose versus specialized technology, and what happens when workloads evolve faster than infrastructure can.
    44 min
  • Five Ways to Think About the Future of the Datacenter
    In this special edition of Shared Everything, we've assembled some of our favorite conversations on the engineering challenges reshaping AI infrastructure. From power generation and rack density to cooling, networking, efficiency, and even the possibility of datacenters in space, these discussions explore how every layer of the modern AI stack is evolving.
    Featuring Renen Hallak, CEO and Cofounder of VAST Data, on why next-generation AI datacenters will be designed around power, including dedicated generation and nuclear energy; Jonathan Koomey, internationally recognized researcher on computing and energy, on the history of datacenter power consumption and why AI could ultimately drive greater efficiency; Ken Patchett, VP of Datacenter Infrastructure at Lambda, on building facilities that can keep pace with rapidly evolving AI hardware; Rich Miller, Founder and Editor of Data Center Frontier, on the architectural changes spanning power delivery, cooling, networking, and high-density AI deployments; and Dan Nishball, Research Analyst at SemiAnalysis, on the economics and total cost of ownership behind the emerging concept of space-based datacenters.
    24 min
  • The Infrastructure Around Kafka Is Ready for Reinvention
    Kafka has been the backbone of real-time data streaming for more than a decade, but AI is changing what modern event infrastructure needs to deliver. In this episode, VAST Data's Matt Mohr joins us to explore the evolution of Kafka, where traditional architectures are beginning to show their limits, and how VAST Event Broker brings streaming, storage, and real-time analytics together on a single platform. From financial services and observability to AI clouds and quantum computing, it's a technical conversation about where streaming is headed next.
    15 min
  • (Re)Engineering the Inference Stack
    In this episode, we're joined by Stephen Bates, AMD Fellow in the AI Group, and Alon Horev, Co-founder and CTO of VAST Data, to explore one of the biggest architectural shifts happening in AI today: the move from training to inference. Using AMD and VAST's expanded partnership as a starting point, the conversation examines why KV cache has become a critical part of AI infrastructure, how long-context and agentic AI are changing the way systems are designed, and why storage, networking, and open software are becoming just as important as the GPUs themselves in delivering fast, scalable inference.
    30 min
  • Encore Episode: TACC's Dan Stanzione on AI, Power, and the Future of Supercomputing

    TACC Executive Director Dan Stanzione, and VAST Data's Don Schulte discuss the evolution of the Texas Advanced Computing Center and its role in high-performance computing (HPC). Dan highlights TACC's history, including the transition from Stampede to Stampede 2 and the impact of AI on power consumption and cost. They discuss the upcoming Horizon system, which will replace Frontera, featuring 4,000 Nvidia GPUs, 900,000 CPU cores, and 0.5 exabytes of solid-state storage. The conversation also touches on the importance of data management, the shift from batch-oriented workflows to real-time data assimilation, and the potential of emerging technologies like photonics and quantum computing.

    31 min
  • Where's The Real Bottleneck in Enterprise AI?
    Everyone wants to talk about AI models, but most enterprise AI projects succeed or fail because of the data underneath them. In this episode, Abhas Ricky, Chief Business Officer at Cloudera, joins Jeff Denworth, co-founder of VAST Data, to explain why getting the right data to the right place at the right time is still the hardest problem in AI. They discuss how enterprise infrastructure has changed since the Hadoop era, why keeping GPUs busy requires much more than fast storage, and how companies are bringing together data, governance, metadata, and AI into a single architecture that can support real production workloads instead of impressive demos.
    27 min
  • The Data Mess the AI Era Exposes
    Howard Marks has spent more than four decades watching every major shift in enterprise computing, giving him a rare perspective on what AI is changing beneath the surface. In this episode, he explains why the AI boom has fundamentally altered the economics of infrastructure, forcing organizations to rethink long-term planning, data management, and security. The conversation explores why modern storage has evolved into an intelligent data platform, how enterprises can prepare their data for AI without massive forklift upgrades, and why governance and access controls must become native parts of the infrastructure itself.
    21 min
  • AI Is Accelerating Cybersecurity Faster Than Enterprises Can Adapt
    On this episode of the Shared Everything podcast, Nicole sits down with cybersecurity analyst and longtime practitioner Adrian Sanabria to explore how AI is reshaping enterprise security faster than organizations can adapt. They discuss why AI has accelerated vulnerability discovery, the growing challenge of prioritizing risk instead of simply patching everything, how prompt injection and retrieval poisoning expose fundamental weaknesses in today's large language models, and why supply chain security remains one of the industry's biggest unresolved problems.
    25 min
  • Can We Build the AI Infrastructure We Keep Promising?
    On today's episode of the Shared Everything podcast, we're joined by Rich Miller, one of the datacenter industry's longest-serving journalists, to examine what's really driving the AI infrastructure boom. We discuss the growing gap between ambitious announcements and what can actually be built, why power and cooling have become the industry's biggest constraints, how investors are evaluating the next wave of projects, and whether technologies like advanced liquid cooling, small modular reactors, and even space-based datacenters can help meet exploding demand. Rich also shares insights from more than 25 years covering the evolution of the modern data center and where the industry is headed next.
    27 min
  • Productivity Is Replacing Performance as the Real Metric
    On this episode of the Shared Everything podcast, Nicole Hemsoth sits down with Sven Breuner, Field CTO at VAST Data and one of the pioneers of modern parallel file systems, to discuss how AI is reshaping data infrastructure. Drawing on two decades of experience in HPC storage, Sven explains why the industry is moving beyond a singular focus on performance toward making data easier to access, automate, and extract value from. They explore the convergence of AI and traditional scientific computing, the growing importance of intelligent data platforms, workflow automation, and why the future belongs to organizations that can turn ever-growing volumes of data into meaningful insight.
    21 min