StorageReview.com

By StorageReview.comTechnology

StorageReview.com episodes

October 26, 2025Podcast #141: Graid Delivers Enterprise RAID Functionality to AI Workloads
Gain insight into improving RAID performance for AI workloads with Graid SupremeRAID AE.
Storagereview has worked with Graid on numerous projects, including our most recent review, Storage at GPU Speed, and they continue to impress. In this podcast, Brian chats with his friend, Kelley Osburn, Senior Director of OEM and Channel Business Development at Graid Technology.

Kelley is responsible for developing and executing the OEM and channel sales strategy, managing key partnerships and relationships, and expanding the company’s market presence and revenue. He leverages his expertise in flash memory, data storage, virtualization, cloud, DevOps, and containers/Kubernetes to deliver value-added solutions to our customers and partners.
Graid Technology delivers enterprise-grade RAID functionality for AI workloads with minimal infrastructure changes. Instead of a dedicated hardware RAID card or custom appliance, AE is offered as a software license and uses only a small share of an existing NVIDIA GPU. This enables organizations to achieve enterprise-grade storage performance and reliability without consuming additional PCIe slots, requiring infrastructure changes, or significantly impacting GPU performance for training and inference workloads.
Brian and Kelley dive into the functionality of the Graid solution, why it’s critical in today’s AI landscape, and what the future holds.
If you are interested in improving RAID performance for AI workloads without adding more GPUs to your existing racks, this podcast delivers.
If you don’t have time to listen to the entire podcast, we have broken it down into five-minute segments so you can easily hop around to get the information you need.
Listener’s Guide
Why Graid? Problem Solved! [00–05]
Traditional RAID/MD-RAID hit a wall with NVMe and PCIe Gen4/5/6—hard x16-lane bottlenecks.
Unsafe stopgaps: RAID 0 + snapshots/checkpoints increase risk and admin overhead.
CPU-bound math in MD-RAID diverts cycles from apps; GPUs excel at parallel parity/EC math.
Graid offloads to the GPU and uses peer‑to‑peer paths, so data bypasses the card (no x16 choke).
Anecdote: 16–44 drive servers magnify the physical lane mismatch of legacy HW RAID.
Outrunning “x16 limits” [05-10]
Affordable GPUs (A2000/A400) act as a traffic cop, not a data ferry—peer‑to‑peer NVMe routing.
Real builds: ~200 GB/s with ~24 drives; new paths push ~300 GB/s/server.
Design tradeoffs: ~24 x4 drives for peak perf; 40+ drives at x2 favors capacity/IOPS.
Small form factor: Low‑cost edition supports up to ~12 drives—ideal for AI desktops.
Banter: “How can it be faster than x16?”—because the data doesn’t traverse the GPU.
AI Edition: Use existing compute GPUs [12–15]
Run Graid on H100/Blackwell, you already own; no extra slot for a RAID GPU.
Tiny footprint: ~6/144 SMs on H100; “drop out” when idle to free resources.
GPU Direct Storage moves data from NVMe to GPU memory directly (lower latency/hops).
Keep mid‑riser slots for 400/800G NICs and other accelerators.
Anecdote: “Feed the Beast.” Keep GPUs from starving on IO.
Lab impact and edge patterns [15–20]
Modest tokens/sec impact during inference; storage and AI seldom peak at once.
Edge deployments: Single 2U with 2 GPUs for capture, light train, and inference at scale.
Preserve IO slots by sharing existing GPUs with Graid.
Examples: YOLO-style wildlife, retail, and other ready-to-use inference models.
Tip: Size storage to avoid starvation under bursty pipelines.
Integrity, namespaces, and scale [20–25]
RAID is more relevant at the edge: Resilience keeps GPUs productive during faults.
Transient errors: Detect bad reads, rebuild from on-the-fly parity, and relocate data.
Big namespaces: 61/122/256 TB-class drives make large protected pools valuable.
Scale via NVMe‑oF JBOFs (RoCE/RDMA); 24–96+ raw devices still require software RAID.
Roadmap: Higher drive counts + erasure coding for flexible protection domains.
Beyond AI: Databases, streaming, analytics [26–30]
Databases/HFT: Redis, Aerospike, and Oracle benefit from fast, consistent writes.
Splunk/log ingest: Sustained throughput prevents drops and backpressure.
Media: Multi-8K stream servers can reduce the total server count for a workload.
Density/power constraints favor higher perf per RU in colocations.
Rebuilds: Parity flows through GPU; slight latency increase, rarely app‑visible.
Roadmap, PCIe Gen6, how to engage [30-35]
Software-first: moving Gen4 to Gen5 boosts perf since Graid isn’t in the data path.
Gen6 flash (~28 GB/s x4) magnifies the need to remove chokepoints.
Peer-to-peer + GPU-offloaded math unlocks new PCIe generations without HW churn.
How to buy: Dell catalog, Supermicro OEM, channels (CDW), and Amazon.
Meet Graid: An NVIDIA Inception partner with booths at GTC Washington & San Jose.
The post Podcast #141: Graid Delivers Enterprise RAID Functionality to AI Workloads appeared first on StorageReview.com.
36 min
September 27, 2025Podcast #140: Inside HPE SimpliVity with Morpheus VM Essentials
HPE integrates Simplivity Hyperconverged HCI plus Morpheus VM Essentials stack. Get the details in this podcast.
Brian and the team traveled to Spring, TX, to get an up-close look at what HPE is working on. After the tour, Brian sat down with HPE’s Vinay Jonnakuti, Director, Product Management for HCI and Ecosystems, to unpack the newly integrated SimpliVity plus Morpheus VM Essentials stack. They discuss why integrated HCI is important now, the benefits of two-node high availability at the edge, cost savings with per-socket licensing, and how Morpheus combines hybrid cloud and automation. It also includes real-world retail and AI at the edge examples.
Brian and Vinay’s friendship goes way back, making this podcast interactive and fun. They do a bit of reminiscing, talk about the Silicon Valley TV show from 2016, and the fact that Vinay still has the Pied Piper t-shirt.
Then, it’s time to get down to business. HPE is strengthening its vertically integrated HCI stack by combining SimpliVity’s efficient data platform with Morpheus VM Essentials as the control plane. The result is a fully supported, single-vendor solution covering servers, networking, hypervisor, and management. Customers benefit from two-node HA, integrated backup and deduplication, automated deployments and upgrades, and per-socket licensing that lowers TCO, especially for SLED, retail, and distributed enterprises. Morpheus adds hybrid cloud orchestration to the platform, while ProLiant hardware accelerates the adoption of modern NVMe, CPUs, and NICs. Expect regular release schedules and more ISV-backed solutions for the retail, healthcare, and manufacturing sectors.
There is a lot to learn from this podcast. And it’s only 30 minutes! But if you don’t have the time to watch it in one sitting, we have broken it down into five-minute segments, highlighting the topics covered in each segment.

00:00–05:00
Introduction: Brian and Vinay set the scene at HPE in Spring/Houston. Brian introduces Vinay and asks him to describe his role leading product management for HPE SimpliVity and recall the pre-acquisition days and pop-culture tie-ins (Silicon Valley/Pied Piper bezel).
The situation: Transition from nostalgia to today’s focus on HPE’s vertically integrated stack, SimpliVity HCI, and HPE Morpheus VM Essentials as part of HPE’s end-to-end story.
05:00–10:00
Market dynamics and alternatives: Discussing VMware cost pressures, especially for SLED and smaller organizations, and exploring the landscape of alternatives, including support gaps in Proxmox, Nutanix costs, Azure Stack options, and other niche players.
HPE’s integrated differentiation: HPE’s unique “soup to nuts” approach, with servers, networking, hypervisor, and control plane, delivering a fully integrated, single-vendor solution with SimpliVity + Morpheus VM Essentials.
Morpheus scope: Morpheus Enterprise’s broad automation/customization capabilities and how VM Essentials provides a focused subset tailored to HCI management.
10:00–15:00
Operational simplicity and cost: Automated deployment/upgrades through the VM Essentials Manager; positive beta feedback; per-socket licensing helps TCO (no per-core tax), aligning well with budget-constrained sectors.
Feature depth retained: Built-in migration tooling, integrated backups, data efficiency (dedupe/compression), and day-one parity for core SimpliVity capabilities with a roadmap to expand features.
Two-node advantage: Unique two-node HA architecture (with lightweight witness) maintaining two copies of data, enabling predictable performance, resilience, and lower cost versus 3–4 node minimums elsewhere.
15:00–20:00
Portfolio fit and scale: Clear positioning for SimpliVity at the edge, distributed enterprise, and smaller clusters, with options to scale up through HPE’s broader storage portfolio (Electra MP/B10K) and similar management experiences.
Cadence and alignment: Tight release cadence: Morpheus VM Essentials shipping frequently (monthly to ~6-weekly) and SimpliVity targeting multiple releases per year, adding features, hardware support, and deep GreenLake integrations.
Unified platform vision: GreenLake + OpsRamp + Morpheus + ProLiant compute form the backbone for hybrid cloud management, observability, and orchestration with SimpliVity as a key edge component.
20:00–25:00
Hybrid cloud reality: The “world is hybrid” message—Morpheus can manage multiple hypervisors and public clouds alongside SimpliVity for a unified control plane and visibility.
Hardware matters (again): Emphasis on modern ProLiant-based platforms enabling faster adoption of NVMe, larger drives, faster CPUs/DRAM/NICs, and how moving off custom accelerators increases agility.
Edge AI use cases: Retail-edge scenarios (video analytics, loss prevention, restock, shopper intent, touchless/self-checkout) benefit from local inference; SimpliVity nodes with NVIDIA support can host AI/analytics VMs for real-time insights.
25:00–30:00
Vertical solutions and ISV ecosystem: HPE partnering with ISVs in retail, healthcare, and manufacturing to deliver validated joint solutions on SimpliVity with Morpheus control, combining resilience and data protection with domain apps.
Customer engagement: Houston Customer Briefing Center demos (rack of HPE gear + SimpliVity node), upcoming collateral and videos, and guidance to engage HPE partners/field for evaluations.
Wrap-up: Closing remarks and availability timing for the integrated SimpliVity + VM Essentials solution. Look for links to a deeper technical dive video and white paper accompanying the podcast.
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
The post Podcast #140: Inside HPE SimpliVity with Morpheus VM Essentials appeared first on StorageReview.com.
0 min
July 24, 2025Podcast#139: Immersion-Cooling, The Valvoline Way
Following a tour of the facillities, Valvoline’s VP of R&D, sat down with Brian to go deep on immersion-cooling.
Brian and the StorageReview team traveled to Lexington, Kentucky, to meet with Valvoline’s immersion-cooling pit crew to examine the practical, technical, and environmental aspects of immersion and direct-to-chip cooling. They highlighted the significance of material compatibility, the development of industry standards, and Valvoline’s ongoing R&D efforts to create fluids that are more sustainable and effective for the data center industry.
The hosts graciously gave a tour of the Valvoline HQ, which included some unique historical insights, followed by a look at race cars on display. Brian, a true motor racing fanatic, was in heaven.
Brian was pulled from the race cars and sat down for a conversation with Dr. Z. George Zhang, Vice President of R&D at Valvoline Global, about current and future trends in immersion cooling.
Dr. Zhang has been with Valvoline for over 26 years and is highly respected within the liquid-cooling community. He recently joined the Advisory Board for the North America Lubricant Expo & the Bearing Show.
He is a certified lubrication specialist (CLS) and oil monitoring analyst (OMA I). Dr. Zhang is an inventor or co-inventor on 8 U.S. patents and has authored 49 peer-reviewed publications.

With all the buzz around liquid cooling, this podcast is a must-listen for anyone interested in immersion cooling technology. If you’re strapped for time, we’ve broken it down into five-minute segments to help you navigate this important topic.
0:00 – 5:00
Introductions & Backgrounds: Brian introduces Dr. George Zhang from Valvoline, highlighting Valvoline’s long history in automotive lubricants and innovation (like high-mileage oils and MaxLife).
Motor Oil as Engineering: Discussing how people often overlook the complexity of motor oil. It’s not only about lubrication but also cleaning, sealing, and heat transfer.
Heat Transfer Parallels: Dr. Zhang explains how expertise in automotive heat management translates to data center cooling, especially as data centers face massive heat loads from high-power GPUs.
Data Center Energy Use: Fun fact: ChatGPT’s 2024 energy use is comparable to the entire country of Ireland, highlighting the scale of heat management required.
5:00 – 10:00
Mission-Critical Differences: Data centers operate 24/7, unlike cars, so uptime and stability are crucial; however, the environment is less chemically harsh than that of automotive engines.
Material Compatibility: Data center hardware utilizes various metals, plastics, and labels, differing from those used in cars, making fluid compatibility a key concern. Labels and plastics can break down in immersion cooling environments.
Cleanliness & Purity: In direct-to-chip cooling, fluid purity is critical because tiny channels in cold plates can clog easily, impacting cooling performance.
Cooling Segments: Air cooling is still dominant, but liquid cooling (direct-to-chip and immersion) is growing. Valvoline is active in both, with a focus on their PG 25 fluid for direct-to-chip.
10:00 – 15:00
Technical Differentiation: Not all PG 25 fluids are the same. Valvoline’s patented formula utilizes low-solids and liquid-based additives to prevent deposits and salt bridges.
Condition Monitoring: Valvoline leverages its automotive expertise to help customers monitor the health of their fluids and system conditions, supporting global supply chains.
Ecosystem Complexity: Liquid cooling involves numerous partners, including server vendors, hose and cold plate manufacturers, quick-connect suppliers, and more. Collaboration and communication are key to avoiding finger-pointing when issues arise.
Water Quality & Sensors: Data centers often lack expertise in water quality, so Valvoline is collaborating with sensor manufacturers to enable real-time monitoring of fluid properties, such as turbidity and conductivity.
15:00 – 20:00
Sensor Integration: The future will include more sensors in cooling loops for early detection of issues, potentially even at the connector or cold plate level.
Market Adoption: Most US data centers are adopting cold plate/direct-to-chip cooling because it can be retrofitted, while immersion cooling requires new infrastructure.
Immersion Cooling Advantages: Immersion can cool 100% of components and is vastly more efficient at heat transfer than air, but requires careful fluid movement to avoid hot spots.
Thermal Conductivity: Immersion fluids (hydrocarbons) are chosen for their low electrical conductivity (dielectric properties), allowing for the safe submersion of electronics.
20:00–25:00
Immersion Cooling in Practice: Discussing different immersion cooling setups, such as tank immersion (servers dunked vertically) and their impact on power efficiency (PUE as low as 1.03–1.04, significantly better than air cooling).
Component Reliability: Immersion cooling can significantly reduce failure rates for components such as RAM, storage, CPUs, and GPUs. However, it introduces new operational challenges, such as handling servers with cranes and dealing with oil drips.
Fluid Chemistry: Valvoline uses pure hydrocarbon fluids for immersion—no fluorine or chlorine, just carbon and hydrogen. This ensures chemical inertness, similar to air, and avoids reactivity with server components.
Material Compatibility: Early lessons included issues with cables, labels, and plastics not designed for use with oil, such as color leaching, labels falling off, and plastics becoming brittle. Compatibility testing is now a significant focus.
25:00–30:00
Industry Standards & Collaboration: Valvoline collaborates with OCP (Open Compute Project) and ASHRAE to develop and adhere to material compatibility lists for both immersion and direct-to-chip fluids. These lists help customers choose the right materials for liquid cooling.
Networking & Signal Integrity: Concerns exist about high-speed networking and light transmission through oil; however, most hardware manufacturers now relocate switches out of the fluid. Valvoline hasn’t seen major signal issues but can adjust fluid dielectric properties if needed.
Industry Perception: Valvoline is often considered “the car guys” at IT events, but they emphasize their expertise in heat transfer and chemistry, which is directly relevant to data center cooling.
OCP Working Groups: OCP has separate groups for immersion and direct-to-chip cooling, showing the industry’s growing seriousness about liquid cooling.
30:00–35:00
Alternative Immersion Designs: Discussion of “precision immersion,” like Isotope’s approach, where fluid is sprayed directly onto hot components in a rack, using less fluid and allowing for easier retrofits. These designs can be more energy-efficient and fit standard racks.
Component Coverage: Precision immersion ensures that all components, including those that do not generate significant heat, such as capacitors and transistors, are effectively cooled, potentially improving overall reliability.
Multiphase Cooling: Multiphase (two-phase) cooling utilizes the heat of vaporization for enhanced efficiency; however, the fluids used are often PFAS (“forever chemicals”), which are environmentally problematic. Valvoline avoids these for sustainability reasons.
System Complexity: Multiphase systems require sealed environments to prevent gas escape, adding complexity and risk.
35:00–37:00 (End)
Fluid R&D and Future Directions: For immersion fluids, Valvoline is researching ways to reduce its carbon footprint, including bio-derived hydrocarbons, although these present challenges due to the presence of double bonds and higher costs.
PG 25 (Direct-to-Chip) Sustainability: PG 25 is currently derived from petroleum, but plant-based options are available, although they are pricier. Valvoline is exploring ways to make these more cost-effective and sustainable.
Chemistry Matters: Chemistry knowledge is increasingly essential for IT professionals working with advanced cooling systems, compatibility, fluid properties, and environmental impact, all of which hinge on a solid understanding of chemistry.
Lab Testing: Valvoline does extensive in-house testing for compatibility, performance, and system efficiency, both for immersion and closed-loop cooling systems.
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
The post Podcast#139: Immersion-Cooling, The Valvoline Way appeared first on StorageReview.com.
40 min
June 06, 2025Podcast #138: Solidigm Liquid-Cooled SSDs
Providing a look at the latest in liquid-cooled SSDs from Solidigm.
Brian invited Cody Noland, Solidigm’s Sr Thermal Mechanical Engineer, to the show to discuss the latest in liquid-cooled SSDs. The Solidigm liquid-cooled D7-PS1010 E1.S, the industry’s first cold-plate-cooled SSD. The drive was introduced during GTC 2025 and garnered massive attention during the event and after. Solidigm’s proprietary E1.S SSD case design efficiently channels heat from the non-contact side through the case to the side that interfaces with the cold plate. This promotes efficient cooling on all sides of the drive, improving thermal performance and reliability during demanding workloads.
Cody has been working as a thermal mechanical design engineer for most of his career, designing solutions for companies such as Intel and CoolIT. He performs complex thermal analysis on PCBs with multiple power sources influencing board placement and optimizing conduction paths. Cody is skilled in CFD, CAD, Solidworks, DFMEA, DFM, DFA, CNC, CAM, Thermal data acquisition, thermal analysis, data analysis, server thermal analysis, mechanical tolerance analysis, and HVM design. It is apparent where his interest lies.
Cody’s hobbies included using his CNC Router, woodworking, and welding.

This is a technical discussion on cooling and thermal design. This is a shorter-than-normal podcast, so if you can spare 30 minutes, it’s worth your time. If not, we have broken it down into five-minute segments, so you can find the topics that interest you.
0–7 min:
Why Liquid Cool SSDs? (And Who Let the Engineers Out?)
Introducing Cody from Solidigm
Highlighting Solidigm’s GTC demo: liquid-cooled SSDs
Pairing cold plates and storage, usually reserved for CPUs and GPUs
What inspired Solidigm to take the plunge
Data centers are evolving from air to hybrid to full liquid cooling
Pushing the market forward by showing that storage can (and maybe should) be part of the liquid cooling revolution
The goal: keep SSDs cool without sacrificing serviceability or requiring fancy tools
Reminiscing about dunking Solidigm drives in Castrol fluid for a previous video, highlighting the humor (and mess) of hot-swapping SSDs in oil
Can we cool the whole system, including storage
7–11 min:
The One-Sided Cold Plate
Solidigm’s demo utilized E1.s drives, which are popular with hyperscalers
Good fit for liquid cooling experiments
The cold plate only contacts one side of the SSD, saving space and adding more drives
Are there changes to the PCB
Some internal tweaks to optimize heat transfer, but the core drive remains familiar
11–17 min:
Performance, Power, and the Next Gen of SSDs
Does liquid cooling just match air cooling, or does it actually boost performance
Liquid cooling is a game-changer, pushing SSD power and performance beyond what’s possible with air cooling
Power management
What does the future hold for Gen6 SSDs
Power is going up, and connector limitations are a factor
Pushing the boundaries for specific applications
Will Solidigm patent this, or will it be open-sourced
17–22 min:
From Labs to Parking Lots, The Wild World of Cooling
The explosion of liquid cooling vendors and the rise of immersion cooling
Wild deployments like DUG’s 440+ immersion tanks
It’s an exciting time to be a thermal engineer
A significant surge in cooling innovation
Solidigm is ensuring its drives can handle air, immersion, and direct liquid cooling
Immersion cooling can lower failure rates, but will liquid cooling make SSDs even more reliable
22–27 min:
GTC Demo Reactions: “What Is Going On Here?”
What did people think of the GTC demo
The most common reaction was, “What is going on?”
The demo sparked a lot of industry chatter and got people thinking about 100% liquid-cooled systems and new ways to design dense storage.
Wrap-up:
The Future of Liquid-Cooled Storage
Wrap up, noting that liquid cooling is here, especially for high-density AI servers
Hats off to Solidigm for showing off new tech before it’s even on the market
Thanks to Cody for escaping the thermal lab to share his insights
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
The post Podcast #138: Solidigm Liquid-Cooled SSDs appeared first on StorageReview.com.
30 min
April 25, 2025Podcast #137: AI Is Forcing Data Centers to Go Liquid
Explore how 2MW liquid cooling is transforming AI data centers with new levels of efficiency, scalability, and thermal management.
If you attended Data Center World 2025 in Washington, DC, it would have been challenging to walk far before finding another liquid cooling solution. Each is unique with a special sauce. We have been following the innovations to get more from the compute side, and most solutions point to liquid cooling. One of the companies we have worked with is CoolIT. They just released the CHx2000 at DCW, and we were there to talk to Luca Cutrone, Director of Business Development at CoolIT.

Luca’s LinkedIn profile describes him as a “fun-loving technophile” with over a decade of leadership experience. Following a seven-month break, Luca joined CoolIT in January 2024, and it appears to be a great fit. Luca’s focus throughout his career has been on customer success, and he carries that mission forward in his work at CoolIT.
Brian caught up with Luka Cabrone in the CoolIT booth at Data Center World to discuss the latest innovations in liquid cooling. The conversation focused on CoolIT’s new product launches, the evolving data center cooling landscape, and the practical challenges of implementing advanced cooling solutions in today’s high-density environments. It was recorded live on the show floor.
More information about CoolIT can be found in Brian’s review of their direct liquid-cooled architecture.
If you are interested in liquid cooling technology, this podcast is for you. A lot of ground is covered in this short podcast. However, if you don’t have time to watch it in one sitting, we’ve broken down the conversation highlights into five-minute segments so that you can skip around.
00:00–05:00: Introduction
CoolIT’s new 2000 liquid CDU is capable of rejecting up to 2 megawatts of heat.
The growing need for liquid cooling in high-density chips and AI is discussed.
Educating the market on transitioning to liquid cooling.
05:00–10:00: Innovations
The real-world benefits of liquid cooling include power savings and reduced fan usage.
A University of Chicago case study shows a 37% power savings in a lab environment.
Quieter, cooler labs are a benefit of liquid-cooled environments.
Optimizing power delivery is essential as rack densities increase.
10:00–15:00: Immersion cooling
The chemistry behind CoolIT’s cooling systems— using a propylene glycol-water mix.
Emphasis on filtration and system cleanliness for cold plate microchannels.
Single-phase vs. two-phase cooling, with CoolIT focused on single-phase.
Immersion cooling and the engineering behind CoolIT’s 4000-watt cold plate.
15:00–20:00: How to acquire CoolIT solutions
The buying process for CoolIT’s solutions and integration into operations is explained.
CoolIT works with OEMs for factory integration, and facilities teams manage the infrastructure.
OCP is pushing for standardization, but challenges remain around warranty and compatibility.
20:00–End: Challenges
The future challenges of managing increasing heat loads are discussed.
The demand for innovative, smart operational tools, such as leak detection and AI-driven monitoring.
Industry collaboration on standards is essential.
CoolIT is committed to innovation and supporting the data center ecosystem.
As we were walking through the exhibit floor, it was clear that liquid cooling is no longer a niche technology, but a critical enabler for the next generation of data centers.
CoolIT
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
The post Podcast #137: AI Is Forcing Data Centers to Go Liquid appeared first on StorageReview.com.
0 min
January 17, 2025Podcast #135: HPE Alletra Storage MP Sanjay Jagad
Brian talks with HPE’s VP of Product Management for Cloud Data Infrastructure while on-site at the HPE facility in Spring, TX. Listen as they discuss the Alletra Storage MP system.
Brian recently visited HPE’s headquarters in beautiful downtown Spring, TX. He invited Sanjay Jagad, a good friend of Storagereview.com, to join him for a detailed discussion about storage and HPE Alletra Storage MP. They discussed the factors driving the storage industry today and how MP has impacted storage system construction.
We have a full write-up and long-form video on YouTube to get up to speed on the platform.

This 40-minute podcast covers a wide range of topics, and it is well worth your time if you can spare it. Otherwise, we have time-stamped the recording so you can skip around as needed.
0:00-5:00 Introduction
HPE Campus Visit & Introduction to Alletra MP Storage
Brian visits HPE’s headquarters in Texas and introduces the Alletra Storage MP.
Discuss HPE’s universal storage platform, which separates hardware from operating environments. Provides flexibility and high availability.
Overview of business benefits and future of storage from HPE’s perspective.
Evolution of Storage
Sanjay explains the traditional two-controller storage system and how the market is evolving to meet modern demands, such as increased data and distributed applications.
5:00 -11:00 Introduction to Alletra Storage MP
HPE’s Reimagined Storage Approach
Introduction to Alletra MP, HPE’s reimagined storage solution.
Focus on simplifying hardware designs with modular building blocks and embracing software-defined storage (SDS) for improved scalability.
Decoupling Hardware and Software
HPE’s strategy of decoupling hardware from software to stay adaptable to new technologies (e.g., NVMe fabrics, CPUs).
This allows for independent scaling of compute and storage, giving customers more flexibility.
11:00 -15:00 Scale and Efficiency
Granular Scalability and Cost Efficiency
Granular scalability allows customers to add storage or controllers incrementally, preventing overprovisioning and optimizing costs.
This OPEX-based approach enables customers to pay only for the resources they need at a given time, improving TCO (Total Cost of Ownership).
Supply Chain Efficiency and Modular Hardware
HPE’s standardized hardware platform, using non-proprietary components (e.g., AMD CPUs, OCP NICs), ensures quicker adoption of new technologies like Gen 5 SSDs.
The modular, Lego-like design allows customers to integrate new tech without replacing entire systems.
15:00 -22:00 Availability and Performance
High Availability and Performance Improvements
HPE’s stateless controllers ensure seamless failover, offering high availability and performance improvements via the NVMe fabric.
Customers benefit from dynamic scaling and minimal downtime, crucial for mission-critical workloads.
Investment Protection through Continuous Upgrades
Timeless program: HPE ensures that as newer CPUs, memory, and I/O modules are released, they can be integrated seamlessly without requiring a complete system replacement.
This offers customers future-proofing and investment protection for their infrastructure.
22:00 -26:00 Ability to adapt new technologies
Adapting to New Technologies and AI-Driven Management
HPE’s architecture enables rapid adoption of new technologies (e.g., Gen 5 SSDs, DDR5, CXL).
AI Ops simplifies operations by automating routine tasks such as provisioning, monitoring, and optimizing workloads. This reduces manual effort and speeds up processes.
Dynamic Workload Management
The system leverages AI to dynamically adjust infrastructure resources and Kubernetes policies based on workload needs, ensuring SLA compliance and consistent performance.
26:00 -30:00 Unified Hybrid Cloud Control
Seamless Hybrid Cloud Management
HPE’s architecture supports seamless hybrid cloud management, allowing customers to control both on-premises and cloud-based resources from a single platform.
This provides true hybrid cloud flexibility, enabling consistent governance and security across environments.
Operational Simplicity and Time Savings
The architecture reduces the time required to provision and manage storage from weeks to minutes.
With fewer classically trained storage admins, HPE’s platform is easy to use for IT teams, making management more efficient.
30:00 -34:00 Integrating GreenLake
GreenLake Integration and Service Expansion
HPE’s GreenLake Cloud Console enables customers to expand their storage services easily, starting with block storage and adding more services like data protection or AI services as needed.
GreenLake offers a scalable and flexible management platform for customers to adapt their hybrid cloud strategies.
Reimagining Enterprise Storage
HPE’s shift from traditional SAN to a universal platform offers greater flexibility and efficiency.
This approach caters to traditional IT workloads and modern cloud-native applications, meeting a wide range of enterprise needs.
34:00 -40:00 Wrap Up
Customer Success and Scalability
Strong customer adoption of Alletra MP, especially among large enterprises looking to standardize on HPE’s platform.
HPE’s system’s modularity allows for seamless scaling, whether for compute or storage, as customer needs evolve.
Future Innovation and Flexibility
HPE continues to innovate with a robust roadmap, promising more functionality and scalability in the near future.
The focus remains on future-proofing customer infrastructure and addressing dynamic, cloud-native workloads.
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
The post Podcast #135: HPE Alletra Storage MP Sanjay Jagad appeared first on StorageReview.com.
42 min
December 09, 2024Podcast #136: HPE XP8 Goes Ten Years With No Downtime
Brian visits HPE in northern California and has a chance to sit down with Rupin Mohan, Sr. Director (GM) Primary OEM Business at HPE.
This week, Brian welcomes a true veteran of the storage industry. Rupin Mohan is the Sr. Director (GM) Primary Storage OEM Business at HPE and recently met with Brian at one of HPE’s northern California locations to discuss the HPE XP8 Storage system. Naturally, the main topic centers around the fact that the XP8 has achieved an uptime of ten years without a single outage.
 Rupin’s background within the storage industry is extensive. He has spent over 26 years with Hewlett-Packard Enterprise and is focused on “results-driven” business outcomes. He possesses a deep technical knowledge of flash memory, Fibre Channel, enterprise storage, and many other facets of storage technologies.
Rupin is also a member of the Board of Directors and Marketing Chairman of the Fibre Channel Industry Association (FCIA) and a member of the Board of Directors for the Storage Networking Industry Association (SNIA).
This is an engaging discussion about how Rupin and his team hit a ten-year milestone of zero downtime on the XP8. Rupin describes his team’s effort to ensure customers’ 100% availability.
This shorter-than-usual podcast lasts only 32 minutes, so try to take it all in if you have the time. If you are strapped for time, as most of us are, we have broken the podcast down into 5-minute intervals. So hop around if you need to.
Other StorageReview resources:
HPE XP8 On-site video
XP8 Technical paper
00:00 Welcome and introduction
Introducing the HP XP8
Ten Years of Uptime
Mission-Critical Environments
Scalability, Performance, Reliability
04:40 -10:00
Design For Availability
First Disaggregated System
Minimum Configuration
Customization
Hybrid options
Uptime and Optimal Performance.
10:00 – 15:00
XP8 in Mainframe and Open Systems
HP NonStop Systems Support
XP8’s Position in HP’s Portfolio
Legacy System Support
Lounging Under the GreenLake Umbrella
15:00 – 20:00
Hybrid Strategy
Keeping It On-premises
Legacy Systems Support
MP Personas
Built For Diverse Workloads
20:00 – 25:00
Data Migration
XP8 External Storage Feature
Simplifying IT Admin Tasks
Intuitive Management
100% Service Attach Rate
Ensuring Proper Installation
25:00 – 30:00
Financial Impact of Downtime
XP8 Market Stability
High-End Market Is Not Going Away
Keeping Businesses Alive
30:00 – End
Collaboration Across Division
Alletra MP Discussion
Wrap-up
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
The post Podcast #136: HPE XP8 Goes Ten Years With No Downtime appeared first on StorageReview.com.
24 min
November 25, 2024Podcast #135: Modern DaaS with Amazon Thin Clients
Amazon WorkSpaces delivers a compelling alternative to traditional desktops.
This week, Brian follows up on a recent review for Amazon WorkSpaces posted in mid-October with an interesting conversation with Amazon’s Director of Product, End-User Computing, Melissa Stein.
Melissa has spent the past four+ years with AWS and specializes in setting product vision and strategy, management, storage, software, and databases, to name a few. Melissa will be speaking at the upcoming AWS re:Invent, covering contact center efficiency with Amazon WorkSpaces (EUC206: Boost contact center efficiency with Amazon WorkSpaces and Thin Client), complimenting this podcast.
For our subscribers interested in thin client technology, AWS, and cloud computing, give this podcast some attention. If you need to hop around, we have broken down the discussion in five-minute increments to help you navigate where you need to be.
00:00 – 05:00 – Introduction
The podcast covers topics from Amazon’s thin clients, desktop as a service, and changes in the industry.
New Solutions for End User Computing
Amazon WorkSpaces
AppStream
Developing thin clients
Design goals
Keep the costs down
Repurpose devices to reduce turnover and improve security
Security
Imposed restrictions for thin clients
Keep data off the thin client
Encrypt everything
No access for sideloading applications or connecting removable media
05:00 – 10:00
Continue discussion around security
The boot process prevents unauthorized access
Streamlined fulfillment
Pandemic-inspired device availability
Simplicity
10:00 – 15:00
Thin clients are shipped with a video and brochure
No need for help desk support
Simple for hybrid and remote workers
Desktop as a service
Reminiscent of “dumb terminals”
Lag and poor performance are mitigated
Amazon WorkSpaces
Scalable
Easy to manage, easy to use
15:00 – 20:00
Further discussion about scalability
Addressing high turnover
Shifting computing power to the data center
Addressing inefficiencies
Leverage cloud-based solutions
Reducing costs
Catering to seasonal workloads (tax season, open enrollment)
Thin clients for smaller call centers
Amazon WorkSpaces for persistent desktop needs
20:00 – 26:00
WorkSpaces Pools
Cost-effective
Great for users who don’t require persistent desktops
Close integration between hardware and software
Complying with hardware regulations
Energy-efficient
Efficiencies extend to data centers
Reports of a 75% reduction in endpoint expenses
26:00 – 30:00
Customers appreciate:
Cost-savings
Reduced maintenance
Eliminate VPN requirements
Thin client strategy is not limited to a single device
Future developments with a focus on broader use cases
30:00 – 36:00
What’s happening at AWS re:Invent
showcase thin clients
Amazon WorkSpaces
Amazon Connect
AI-driven analytics
Testing to ensure performance and reliability
Secure browsers
Optimize user experience
Wrap-up
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
The post Podcast #135: Modern DaaS with Amazon Thin Clients appeared first on StorageReview.com.
37 min
September 25, 2024 Podcast #134: Leak-Proof Liquid Cooling
Chilldyne Podcast – liquid cooling solutions for data centers focusing on leak-proof design and efficiency for high-performance servers.
Brian rarely gets to sit down to a podcast with a doctor. However, today, he is joined by Dr. Steve Harrington, CEO of Chilldyne. We have been keenly interested in liquid cooling technologies, and there are more popping up all the time. But we have a soft spot for Chilldyne.
Some background on Dr. Steve Harrington. He is the CTO of Chilldyne and the founder of Flometrics. He is an expert and inventor in Fluid Dynamics and thermodynamics, designing pumps, valves, nozzles, flowmeters, aircraft cooling systems, rocket fuel pumps, rocket test stands, turbine flow measurement systems, medical ventilators, air/oxygen mixers, respiratory humidifiers, CPAP machines, spirometers, heat exchangers, vacuum cleaners, oxygen concentrators, motorcycle fairings, infusion pumps, electronics cooling systems, wave machines, data acquisition systems etc. But wait, there’s more. His expertise extends to electronics, programming, optics, nuclear physics, biology, and physiology.
Steve has more than 29 years of experience in fluid dynamics and thermodynamics. He has consulted for aerospace, semiconductor, medical device, racing, electronic cooling, and other industries. He has over 25 patents and has completed projects for NASA, DARPA, SOCOM, and USACE.
When he is not busy, he is a part-time faculty member at the University of California, San Diego, where he teaches an aerospace engineering senior design class where students instrument, build, and fly liquid rockets. He is also a surfer, a pilot, a scuba diver, a boat and car mechanic, an electrician, and a plumber.
There is much to learn about liquid cooling and we think this podcast will help answer some of the questions on the minds of IT and data center professionals everywhere.
That should be enough to make you interested in listening to this entire podcast. However, if you are strapped for time, we have broken the pod down into five-minute segments so you can hop around as needed.
Get the podcast on video!
00:00 – 05:30 Introduction
“Hot” New Technology: Liquid Cooling
Brian opens with a pun about liquid cooling being a “hot” technology in data centers. Liquid cooling has come full circle—once abandoned with the advent of CMOS, it’s now making a resurgence due to the intense heat generated by modern processors.
From Supercomputers to Jet Engines
Steve provides some personal history on his journey to liquid cooling, starting with cooling supercomputers in the 1980s. Fun fact: His expertise in cooling rocket engines and laser systems translated well to modern data center liquid cooling.
Why Did Liquid Cooling Leave?
Liquid cooling took a break because CMOS technology was thought to have solved the power problem. Surprise! Power demands are back with a vengeance.
From Aerospace to Data Centers
In aerospace, liquid cooling isn’t just about cooling; it’s about reliability over time—think planes, rockets, and lasers. Data centers, on the other hand, need uptime and longevity, adding unique challenges to Steve’s transition from aerospace.
ARPA-E Grant and the 2 Kilowatt Chip
Steve’s foresight included partnering with ARPA-E to develop a cold plate for a two-kilowatt chip. This foresight is paying off, as more data centers are bidding for projects with these high-power chips.
05:30 – 10:24 Reassuring the CFO
Liquid Cooling: A CFO’s Nightmare?
Testing liquid cooling systems can cost millions. Convincing CFOs to sign off on such “experiments” is no easy task, especially since they don’t fit neatly into the financial spreadsheet. It’s like buying a $2 million test drive—it sounds fun but is risky.
Sharing Secrets: The Meta Paradox
Meta and other commercial giants tend to lock up their liquid cooling secrets tighter than the company’s algorithm. Sharing is caring, but not when there’s competition involved.
Chilldyne’s Leak-Proof, Negative Pressure System
Chilldyne’s secret sauce: a negative pressure system that’s essentially leak-free. If a leak happens, air enters instead of coolant spilling, keeping the chips safe from a soggy fate.
Handling Leaks: When Not to Panic
Even with leaks, Chilldyne’s system keeps running.
Liquid Cooling Data: The Missing Chapter
Industry-wide data on the impact of liquid cooling on wear and tear is sparse. Liquid cooling can feel like more of a mysterious “beer conference” topic than an open discussion.
10:24 – 14:55 Keep the bacteria out of the water
The Biodiversity of Data Center Coolant
Did you know that the local bacteria in your water supply can mess with your liquid cooling system? Different regions have different microbes, which can wreak havoc on data center coolant systems, leading to clogged plates and overheating GPUs.
Coolant Chemistry 101
Chilldyne has a built-in chemistry lab to monitor coolant quality. Forget your sterile water dreams; this is a battle between biology and technology.
Coolant Additives: A Budget Dilemma
Don’t cheap out on your coolant chemistry unless you enjoy emergency maintenance.
PG 25: Gamers’ Friend, Data Centers’ Enemy
PG 25 is great for gamers—it doesn’t freeze and prevents bacteria growth. However, it attacks seals, leading to leaks over time, making it less ideal for long-term data center use.
Chemistry’s Role in Data Center Maintenance
Data centers often forget that liquid cooling isn’t a “set it and forget it” solution. It requires regular monitoring and maintenance.
14:55 – 20:16 The need for low-toxicity additives
The Additives That Keep the Cool
Chilldyne uses low-toxicity additives: a splash of antibacterial and anti-corrosion chemicals.
Say No to PG 25 (Sometimes)
PG 25 is a “don’t freeze” lifesaver for gamers shipping their liquid-cooled rigs but less so for data centers.
Facility Water is Just the Start
When installing a CDU (cooling distribution unit), Chilldyne starts with distilled or reverse osmosis-filtered water.
The Cold Plate Clean-Up Crew
One customer refused to use recommended additives, leading to a clogged, hot mess in their GPUs. Chilldyne stepped in with a chemical cleanse, but if a cold plate gets too gummed up, sometimes it’s game over—time for replacement.
Liquid Cooling ≠ Set It and Forget It
Electronics may run smoothly for years, but liquid cooling? That’s a different beast.
20:16 – 26:03 Water is still the best for cooling
Water: The MVP of Cooling Fluids
Why stick with good ol’ water? It’s cheap, non-toxic, and performs well—especially in single-phase systems.
Immersion Cooling: The Slow Burner
Immersion cooling with engineered oils has some potential, especially in edge zones or moderate power servers, but it’s just not efficient enough for today’s two-kilowatt monsters.
When Pipes Get Silly Big
Watch for logistical nightmares as cooling systems scale up.
Electrical Limits: The Real Bottleneck
While liquid cooling systems can scale, there’s a limit to how much electrical current a chip can handle.
Don’t Forget the Chiller
While cooling towers are ideal in many places, some regions require chillers due to water scarcity. Go with a cooling tower.
26:03 – 29:44 Is that a leak?
The Leaks Won’t Tell You
Unlike server components, liquid cooling systems aren’t yet smart enough to give you advanced warnings about leaks.
The Problem with Positive Pressure
Positive pressure systems that detect leaks with special tape are reactive—they shut down servers when something goes wrong. Negative pressure systems like Chilldyne’s, however, keep the servers running even with minor leaks, avoiding costly downtime.
Compatibility Chaos
The issue with buying parts from multiple vendors is compatibility problems.
Plumbing Isn’t an IT Skill
Data center operators are skilled in networking, cybersecurity, and power management—but they’re not plumbers or chemists. That’s where specialized vendors come in.
Switch-Over Valves: The Fail-Safe
Chilldyne uses switch-over valves to provide redundancy in cooling systems. These valves work like airplane safety mechanisms—if one system fails, the other kicks in without the servers even noticing.
29:44 – 35:22 Make sure liquid cooling is what you need
Scale Matters: The 100kW Threshold
Steve suggests that liquid cooling isn’t worth the hassle unless you’re dealing with over 100kW of compute power.
Fortune 500 and Beyond
Big companies are already there, but even mid-tier enterprises are starting to feel the heat. The power consumption of GPU-packed servers means that liquid cooling will soon be a necessity for many organizations.
The Rack Scale CDU That Wasn’t
Chilldyne has a design for a 50-100kW rack scale CDU, but no takers yet.
The Four-Inch Pipe Problem
Liquid cooling systems max out at around 1-2MW before the plumbing gets unwieldy. Keeping the system manageable is key.
Electrical vs. Cooling Limits
We’re approaching the point where electrical limits are more problematic than cooling capacity.
35:22 – 40:16 Is it hot in here?
The Heat Disposal Conundrum
It’s not enough to cool the gear—you’ve got to deal with all that heat.
The Chip Temperature Tango
There’s a balance between cooling efficiency and chip performance. Design systems that can go colder if needed because the next-gen chips might perform 20% faster with lower temperatures.
Data Center Math is Getting Hard
Factor in cooling tower efficiency, GPU performance, fan speeds, and a whole lot more. The math behind today’s cooling solutions is critical for optimization.
HVAC and IT: Strange Bedfellows
HVAC engineers and IT teams used to operate separately, but liquid cooling is bringing them together.
Keeping the IT Guys in the Loop
Encourage collaboration between hardware providers and cooling experts to ensure the solution won’t melt under pressure.
40:16 – 43:35 Wrap-up
Get Started with Liquid Cooling Now
Steve recommends starting small, but start now. Get a system, run it, and learn from it before you’re in too deep.
The Danger of Overconfidence
Some companies are planning massive liquid-cooled data centers without ever testing smaller systems.
Lead Time: It’s Real
Lead times for liquid cooling systems can range from 16 to 52 weeks. So, better get those orders in early!
Spec Confusion with New Racks
NVIDIA’s new MVL racks are on the horizon, but specific water temperature and chemistry guidelines are still unclear.
Home Lab Mentality: Get Your Reps In
Steve encourages enterprises to take the “home lab” approach: start small, mess around with it, and learn.
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
The post Podcast #134: Leak-Proof Liquid Cooling appeared first on StorageReview.com.
45 min
September 12, 2024Podcast #133: Effortless Liquid Cooling with JetCool
Closed-loop liquid cooling from JetCool delivers power efficiencies and increased CPU/GPU performance.
Liquid cooling continues to gain traction in the enterprise, especially with the intense compute requirements associated with AI. CPUs and GPUs are running hotter, and traditional air cooling struggles to keep those temperatures down. We have covered several liquid cooling systems over the past few years, with JetCool being a standout player.
After running into JetCool during SuperComputing 2023, we brought their technology into our Ohio lab, nestled inside an R760. To say we were impressed would be an understatement. To gain more insight into this technology, Brian invited JetCool founder and CEO Bernie Malouin for this podcast.
Bernie describes himself as a technical professional with demonstrated experience from concept studies through deployment. By the way, Bernie also holds a Doctor of Philosophy (Ph.D.) in Mechanical Engineering and a Bachelor’s Degree in Mechanical and Aeronautical Engineering. That’s impressive, but before he founded JetCool, Bernie spent eight years at MIT Lincoln Labs. He was the technical and programmatic lead for a multi-sponsor research project to implement advanced cooling technology in high-power RF devices for advanced electronics applications.
Although this podcast is less than 40 minutes long, Brian and Bernie cover a lot of ground. It’s worth viewing or listening in its entirety, but if you are short on time or need specifics on a particular topic, we have broken down the conversation into five-minute chunks so you can skip around.
 00:00 Introduction
The landscape
It’s AI’s fault – or is it hype
GPUs and CPUs need to cool it
It’s about efficiency
What’s the power budget?
Wait! Liquid-Assisted?
06:00 Clarify Liquid Assist, please
Getting technical
Cold plates, closed-loop, and radiators, oh my
Efficiencies in cooling with JETCOOL
Arrays of tiny fluid jets
Surgically removing heat
Now, where are those heat nodes?
Profiling the CPU
10:00 Relationships with the chip makers
Staying ahead of the curve
Chip makers – partners in design
Details on where to cool
Tiny pumps
micro DC brushless pumps
circulating fluid throughout the server
15:00 Let’s talk about pumps, cold plates and radiators
Radiator technology
Modified technology in Formula One racing
Proven reliability
Cold plates
Originally developed for aerospace applications
What about the liquid?
JETCOOL uses PG 25 (Propylene glycol)
Benefits
non-corrosive
natural antifreeze
non-toxic
ready to use
Top off the fluid
can you go to a jiffy lube?
22:00 What happens when the hose breaks
Leak tests
Overpressure tests
Only 8-10 Oz of fluid in the system
Small puncture
faulty seal
might experience a few drops, but they evaporate
the telltale sign that there is a problem
CPU temperature rises
Leak detection mechanisms
25:00 Looking for the signals
OCP is dedicating two days for liquid cooling for their next conference
Partners can do a better job of educating customers about liquid cooling
Upgrade through the channel
driving the pull-through
30:00 Yeah, but how much does it cost?
An upgrade is about ten percent of the system price
ROI
within about six months
Suitable for co-location services
gaining efficiencies in co-lo facilities
On-premises deployments
limited space
limited budget
power constraints
Heavy lifting performed by the Intern and Kevin
37:00 Silence is liquid gold
One of the best benefits
Simple to try
Wrap-up: Stop by JETCOOL at OCP in October!
Full Video on YouTube
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
The post Podcast #133: Effortless Liquid Cooling with JetCool appeared first on StorageReview.com.
40 min

About StorageReview.com

From the publisher's feed

StorageReview.com is a leading provider of news and reviews throughout the entire IT stack - from the datacenter to the edge, and all points in between.

Share StorageReview.com

Sign up to save your podcasts

StorageReview.com

StorageReview.com episodes

About StorageReview.com