Storage

December 16, 2024 by George Crump

The ROI of High-Performance HCI would be a compelling alternative to the high cost of dedicated All-Flash Arrays. However, as application performance demands increase, many HCI solutions struggle to deliver the required performance, scalability, and efficiency. Legacy HCI solutions struggle to keep pace with the demands of modern mainstream applications, let alone today’s high-performance applications. The result is that hyperconverged infrastructure (HCI), which has long promised simplification and cost savings, has never delivered on this promise.

The ROI of High-Performance HCI is achieved only when these platforms meet the demands of today’s workloads. Applications such as virtual desktops and databases require high performance and low latency—capabilities that many legacy HCI platforms fail to provide due to architectural inefficiencies.

Ultraconverged Infrastructure (UCI) is High-Performance HCI

UCI is the next evolution in data center software, designed to overcome the limitations of legacy HCI by deeply integrating virtualization, storage, and networking into a single codebase. Unlike traditional HCI, where storage is often a second-class citizen running as a virtual machine, UCI treats storage as a first-class service, ensuring optimal performance and resource utilization. By addressing these challenges head-on, UCI solutions like VergeOS deliver superior performance and unparalleled ROI.

Want to learn more? Register for VergeIO’s upcoming webinar to see these high-performance results live on cost-effective hardware:
👉 Live Demonstration: Break the Performance Shackles of HCI

Eliminating Expensive Storage Hardware

UCI improves the ROI of high-performance HCI by eliminating expensive dedicated storage controller hardware and vendor-marked-up storage media, which can cost up to 10 times more than off-the-shelf consumer-grade SSDs. These enterprise-class drives are designed with features like capacitors and error-correcting code (ECC) to ensure data integrity, but they significantly inflate infrastructure costs.

Legacy All-Flash Arrays (AFAs) and HCI solutions depend heavily on these specialized components because they rely on the hardware for data resiliency and verification. In contrast, UCI solutions like VergeOS integrate these functions into the software, ensuring data resiliency and integrity without depending on expensive, proprietary hardware. This software-first mentality also enables UCI platforms to mix hardware from different vendors and generations, providing greater flexibility for organizations to scale and upgrade their infrastructure without forklift overhauls.

Matching and Surpassing All-Flash Array Features

Maintaining UCI’s improvements to the ROI of high-performance HCI requires that UCI platforms like VergeOS match and surpass the feature set of traditional all-flash arrays. Enterprises have come to expect capabilities such as:

Unlimited snapshots without performance degradation: VergeOS supports instant snapshots without impacting IOPS or latency, allowing IT teams to back up and restore data seamlessly. VergeOS snapshots go beyond traditional snapshots because they are actually deduplicated clones, providing independence and scalability between snapshot generations.
Global Inline Deduplication: VergeOS performs global inline deduplication without slowing down I/O operations, enabling organizations to maximize storage efficiency without sacrificing performance. Unlike most deduplication technologies, which are added as an afterthought to the software and introduce latency, VergeOS’s deduplication was integrated from day one and has no noticeable impact on performance. Global deduplication also makes disaster recovery data transfers more efficient by reducing the amount of data that needs to be sent, particularly in many-to-one disaster recovery scenarios.
Replication to remote sites: Built-in replication ensures data can be efficiently copied to offsite locations, supporting robust disaster recovery strategies. When combined with VergeOS Virtual Data Centers, which encapsulate the entire data center, replication provides a simple and comprehensive solution for failover and recovery.

Traditionally associated with high-end storage arrays, these features are fully integrated into VergeOS’s UCI platform. By providing these capabilities within a consolidated, software-driven solution, VergeOS simplifies operations and delivers unmatched value.

High Performance with Low Latency

A critical challenge to improving the ROI of high-performance HCI is delivering high performance while maintaining low latency. Many HCI platforms struggle under demanding workloads because they must balance virtual machine and storage responsibilities within the same infrastructure. UCI, by contrast, is architected to ensure resource optimization and eliminate contention. When storage performance demands are extreme, VergeOS can even dedicate servers to specific functions, i.e., compute-only, GPU-only, and storage-only nodes.

By treating storage as a first-class service within its hypervisor, VergeOS achieves consistently low latency and sub-millisecond response times, even under heavy load. This deep integration allows VergeOS to allocate resources intelligently, ensuring that both compute and storage operations run smoothly without interference. Whether running I/O-intensive workloads or supporting mission-critical applications, VergeOS delivers the performance and responsiveness needed to keep pace with future demands.

Real-World Proof: VergeOS in Action

The advantages of UCI ROI versus the ROI of high-performance HCI are best demonstrated through real-world testing. Recently, VergeIO published a benchmark showcasing the performance of VergeOS under demanding workloads. The test utilized eight servers costing less than $1,500 per node, demonstrating that high performance does not require high-cost infrastructure. Key results included:

1.5 Million+ Read IOPS using 64K blocks
24 GB/s of Write Throughput on a 25 GB/s network

It’s worth noting that VergeOS achieved these results using 64K block sizes, which provide a more realistic representation of enterprise workloads compared to the more commonly benchmarked 4K blocks. Again, these tests were performed on off-the-shelf servers configured well below the typical data center-class server.

Want to see these results live? Join VergeIO’s upcoming webinar to learn how VergeOS delivers high performance on affordable hardware:
👉 Want to See UCI Outperform an AFA? Join us for a Live Demonstration

Conclusion: UCI Delivers The ROI of High-performance HCI

With UCI solutions like VergeOS, the ROI of High-Performance HCI is no longer a theoretical concept but a practical reality. By eliminating the dependency on expensive hardware, matching and surpassing the capabilities of traditional storage arrays, and delivering consistent low-latency performance, VergeOS enables organizations to meet the demands of modern workloads without breaking their budgets.

UCI represents the logical evolution of infrastructure software, combining the simplicity of HCI with the performance and flexibility enterprises require. For organizations looking to consolidate operations, reduce costs, and scale efficiently, UCI is the path forward.

Filed Under: HCI Tagged With: HCI, Storage

December 10, 2024 by George Crump

IT professionals who’ve attempted to architect a high-performance vSAN often hit a brick wall. No matter how many nodes and how much network bandwidth they build into their design, they face poor utilization and high latency. As a result, hyperconverged infrastructure (HCI) is pigeonholed into specific use cases, like virtual desktop infrastructure (VDI), and fails to be deployed broadly across the enterprise. This siloing of HCI defeats its primary purpose: consolidating and simplifying data center operations. A high vSAN performance is essential for organizations to realize the promise of HCI.

The challenge is further compounded because high-performance storage I/O is no longer limited to specialized use cases. It has become a requirement for modern workloads. What was once considered high performance is now the baseline, pushing organizations to scale storage I/O performance to meet evolving application demands. Yet, HCI struggles to deliver, forcing businesses to rely on expensive dedicated storage arrays even for workloads that require mainstream performance. A dedicated storage array is 10X the price of the equivalent capacity in a vSAN, if the vSAN can deliver the required performance. Ultraconverged Infrastructure (UCI) bridges this gap, delivering the promised simplicity of HCI while addressing its limitations.

Why HCI Can’t Deliver a High-Performance vSAN

HCI promises to simplify IT infrastructure by combining virtualization, storage, and networking into a single solution. However, beneath the surface, it relies on separate software components hidden behind a graphical user interface. This fragmented design introduces inefficiencies, with storage operating as a virtual machine (VM) within the hypervisor. As a result, storage competes for resources with other VMs, leading to higher latency, lower efficiency, and underutilization of available bandwidth.

HCI comes with strict hardware requirements, further limiting its flexibility. Strict hardware requirements lead to higher upfront hardware costs and make scaling the infrastructure more complex because nodes have to match or be close in configuration to the previous set of nodes. Over time, it becomes difficult to find modern server hardware that matches four- or five-year-old servers. The result is that customers must refresh the entire cluster in one big forklift upgrade—despite HCI initially promising “no more forklift upgrades.”

How UCI Solves the High-Performance vSAN Problem

Ultraconverged Infrastructure (UCI) addresses the inherent inefficiencies of HCI by integrating storage, virtualization, and networking into a unified architecture. Unlike HCI, which depends on separate components, UCI leverages a single, optimized codebase that eliminates resource contention and improves performance.

VergeOS, the leading UCI platform, delivers consistent, scalable performance without requiring proprietary hardware or additional storage arrays. During our upcoming webinar, we will demonstrate this high-performance live so you can see it yourself. VergeOS provides complete hardware flexibility, enabling organizations to mix nodes with different processing and storage capacities. This flexibility allows businesses to scale incrementally, avoiding the costly and disruptive forklift upgrades associated with HCI.

Proof that UCI Delivers

In tests done with Solidigm, VergeIO was able to prove the capabilities of its vSAN:

1 Million+ IOPS at 30 GB/s Throughput: Using 64K block sizes, VergeOS demonstrated over one million random read IOPS with sub-millisecond latency, addressing the needs of performance-critical workloads.
12 GB/s Write Throughput: VergeOS delivered 195,000 random write IOPS, maximizing the bandwidth of a 25Gbps network connection while maintaining low response times.
An eight-node cluster built using $1,500 servers achieved over a million IOPS at a cost per IOPS of just 0.67 cents, making high-performance storage accessible to organizations of all sizes.

The Importance of 64K Block Sizes

Testing with 64K block sizes provides a more accurate representation of real-world virtualized environments, where larger block sizes are commonly used for storage I/O. Traditional 4K block tests measure raw storage device performance, but they fail to capture the demands of virtualized workloads, which include virtual machines, databases, and large-file applications. By utilizing 64K blocks in testing, VergeOS showcases its ability to handle these real-world demands while maintaining exceptional performance metrics, making it an ideal solution for enterprise environments.

Conclusion: a High-Performance vSAN You Can Afford

Modern applications require infrastructure that scales both performance and affordability. While HCI struggles with the demands of today’s workloads, UCI fulfills the promise of a unified, high-performance platform. VergeOS combines optimized performance with flexibility, enabling organizations to handle any mainstream or high-performance workload without expensive, proprietary hardware.

With VergeOS, IT leaders can consolidate their infrastructure, reduce costs, and scale their performance to meet growing demands. To see VergeOS in action and learn more about these results, join our upcoming webinar for a live demonstration. It’s time to move beyond the limitations of HCI and unlock the full potential of UCI.

Filed Under: Storage Tagged With: Storage, UCI

December 10, 2024 by George Crump

Real-World HCI IOPS Results: 1 Million+ IOPS Using 64K Blocks

Ann Arbor, Michigan – December 10, 2024 – VergeIO, a leader in ultra-converged infrastructure (UCI), today announced the release of VergeOS Version 4.13, a breakthrough update developed in partnership with Solidigm, a leading provider of innovative NAND flash memory solutions. This release sets new benchmarks in hyperconverged infrastructure (HCI) performance, scalability, and affordability for enterprise environments.

Testing with 64K block sizes provides a more accurate representation of real-world virtualized environments that often use larger block sizes for storage I/O. Unlike traditional 4K block testing, which primarily evaluates raw storage performance, 64K blocks better reflect the demands of modern virtualized workloads, including virtual machines, databases, and large-file applications. These tests demonstrate how VergeOS and Solidigm can collaborate to deliver meaningful performance improvements for enterprise environments.

All About Storage Performance

VergeOS Version 4.13 focuses on delivering unparalleled storage performance through advanced networking optimizations that reduce latency and improve throughput.

Extreme Performance Powered by Solidigm

The Extreme Performance test for VergeOS 4.13 was conducted by Solidigm, showcasing the full potential of Solidigm’s technology. Key results included:

A 6-node cluster with mainstream dual Gold CPUs and Solidigm Gen 5 NVMe SSDs surpassed 1 million random read IOPS using 64K blocks, a common configuration in virtualized infrastructures, all while maintaining sub-millisecond response times.
In 64K random write performance, the configuration achieved 485K IOPS at more than 30 GB/s throughput, again with sub-millisecond response times, demonstrating exceptional efficiency and reliability in HCI environments.
The raw performance of the Solidigm SSDs using 4K blocks reached 17 million IOPS, pushing the boundaries of storage technology while maintaining sub-millisecond response times.
VergeOS’ data protection and deduplication features were fully active during these tests, highlighting VergeOS 4.13’s ability to maintain peak performance while delivering critical enterprise-grade capabilities.

Response time measurements were taken by deploying separate VMs outside the testing cluster to monitor latency under load. This method ensured real-world accuracy and demonstrated that sub-millisecond response times were consistently achieved, even under noisy neighbor conditions.

“Solidigm’s tests of VergeOS 4.13 demonstrate the unmatched performance and efficiency that our SSDs deliver in demanding workloads,” said Roger Corell, Director of Leadership Marketing at Solidigm. “The ability to achieve over 1 million IOPS with 64K blocks, coupled with sub-millisecond latency and enterprise-grade data protection, highlights the power of our collaboration with VergeIO to redefine hyperconverged infrastructure.”

Affordable Scalability Without Compromise

VergeIO demonstrated the cost-efficiency of VergeOS 4.13 with an eight-node cluster built in its labs using $1,500 servers equipped with consumer-class AMD Ryzen 9 7940HX CPUs, 96GB RAM, and 25Gbps Ethernet connectivity. This affordability test, conducted in VergeIO’s labs, delivered the following results:

1.5 million random read IOPS using 64K blocks at a total cost of $10,000, equating to a cost of just 0.67 cents per IOPS, with sub-millisecond response times, setting a new benchmark for cost-efficiency in HCI solutions.
195,000 random write IOPS using 64K blocks, achieving 12 GB/s throughput, effectively utilizing the network’s 25Gbps bandwidth, with sub-millisecond response times.

“These results demonstrate our commitment to making enterprise-grade performance accessible to organizations of all sizes,” said Greg Campbell, Founder and CTO of VergeIO. “Our affordability test shows that you don’t need expensive hardware to achieve remarkable results. With VergeOS 4.13, customers get a high-performance, scalable solution that fits within their budgets.”

Live Storage Migration: Critical for Next-Gen Storage Technologies

VergeOS 4.13 also introduces live storage migration for virtual machines, an essential feature in the era of advanced storage technologies like Solidigm’s 122TB QLC NVMe drives. These high-density drives, while delivering exceptional capacity, present unique challenges as they are integrated into existing environments.

Live storage migration allows organizations to dynamically move workloads between different storage tiers, optimizing performance, ensuring seamless continuity, and extending the life of storage media. During the December 17 webinar, VergeIO will demonstrate the live storage tiering capabilities of VergeOS 4.13, showcasing how the platform bridges high-performance and high-density storage seamlessly.

“VergeIO and Solidigm’s latest collaboration demonstrates how modern infrastructure can address the challenges of performance, scalability, and density,” said Marc Staimer, President of Dragon Slayer Consulting. “The results of both tests highlight the incredible synergy between VergeOS and Solidigm’s high-density SSDs. Combined with live storage migration, these capabilities empower organizations to adopt next-generation storage technologies without sacrificing performance or reliability, all while reducing costs and operational complexity.”

Experience It Live – December 17th Webinar

VergeIO invites IT professionals and decision-makers to experience VergeOS 4.13 in action during a live webinar on December 17, 2024. The event will feature a live demonstration of the platform’s capabilities and an in-depth discussion on how VergeIO and Solidigm are reshaping hyperconverged infrastructure. Register Here

About VergeIO

VergeIO is the future of virtualization and infrastructure. It is the ideal choice for those seeking an alternative to VMware. VergeIO is a leading provider of ultra-converged infrastructure (UCI) solutions, integrating virtualization, storage, and networking into a single, easy-to-manage platform. VergeIO’s software enables organizations to reduce costs, simplify IT operations, and achieve unmatched performance.

For more information on VergeOS Version 4.13 or to register for the webinar, visit www.vergeio.com.

About Solidigm

Solidigm is a leading global provider of innovative NAND flash memory solutions. Solidigm technology unlocks data’s unlimited potential for customers, enabling them to fuel human advancement. Originating from the sale of Intel’s NAND and SSD business, Solidigm became a standalone U.S. subsidiary of semiconductor leader SK hynix in December 2021. Headquartered in Rancho Cordova, California, Solidigm is powered by the inventiveness of team members in 13 locations around the world. For more information, please visit solidigm.com and follow us on Twitter and LinkedIn.

Media Contact:

Judy Smith, JPR Communications

Email: [email protected]

Filed Under: Press Release Tagged With: HCI, Storage, ultraconverged

October 30, 2024 by George Crump

As IT professionals seek VMware alternatives, they often encounter hyperconverged infrastructure (HCI) solutions, but these systems can’t deliver the media and node flexibility of Ultraconverged Infrastructure (UCI). UCI solutions like VergeIO provide businesses with enhanced adaptability to support diverse storage media and node types. This approach better aligns with real-world demands and long-term infrastructure goals.

What is Ultraconverged Infrastructure?

Unlike traditional HCI or three-tier architectures, UCI integrates storage and networking directly into the hypervisor, running as services rather than virtual machines (VMs). Traditional three-tier systems rely on separate networking, virtualization, and storage hardware components. At the same time, HCI typically bundles these functions but still operates them as independent layers, each running as independent VMs.

With UCI, these critical functions are embedded within the hypervisor, improving efficiency and higher performance. This architectural shift also delivers greater flexibility in choosing media and server (node) types, allowing IT teams to scale infrastructure resources precisely according to their specific workload demands. VergeIO’s implementation of UCI is VergeOS.

The Limitations of Traditional HCI in Mixing Media and Node Types

Traditional hyperconverged infrastructures have rigid configurations requiring identical nodes for computing and storage. Organizations must scale both resources equally, which may not meet their needs. Additionally, traditional HCI solutions can’t support multiple storage types in the same environment, like flash and HDDs. These limitations force businesses to overprovision resources and spend unnecessarily on high-performance storage not aligned with their workloads.

Ultraconverged Infrastructure (UCI) addresses these challenges by enabling independent scaling of compute and storage through a mixed-node approach. It supports various storage media types, allowing IT teams to use high-density QLC flash, high-endurance TLC flash, and HDDs for optimized performance. This flexibility lets organizations assign workloads to the best resources for cost efficiency and improved performance.

Comparing HCI and UCI

The following table summarizes key differences between HCI and UCI, emphasizing how UCI overcomes many of the limitations faced by traditional HCI:

Feature	Hyperconverged Infrastructure (HCI)	Ultraconverged Infrastructure (UCI)
Node Flexibility	Requires identical nodes with balanced compute and storage resources	Supports mixed nodes (compute-heavy, storage-heavy, GPU-heavy), allowing independent scaling
Media Flexibility	Allows independent scaling, adding only storage or computing as needed	Supports a wide range of media types (TLC, QLC, HDD) tailored to workload requirements
Scalability	Must add identical nodes, scaling compute and storage equally	Allows independent scaling, adding only storage or compute as needed
Cost Efficiency	Higher costs due to forced resource overprovisioning	Reduced costs by scaling based on actual workload needs
Resource Allocation	Limited flexibility, requires additional hardware to meet diverse workloads	Flexible resource allocation across different node types for varied workloads
Performance	Often limited by storage and compute configuration; may not fully utilize advanced hardware	Maximizes performance by optimizing workload placement and storage tiering
Data Placement	Typically lacks fine-grained control, with limited storage tiering	Supports advanced data placement and storage tiering, utilizing high-density QLC, TLC, and HDD
Use Cases	Suitable for basic virtualization needs, with uniform resource requirements	Supports diverse workloads (VDI, ML, AI, data lakes, backup) by adjusting to specific needs
High Availability and Recovery	Basic high availability, often requires more servers to maintain stability	Enhanced high availability with efficient recovery, requiring fewer servers
ROI and Resource Utilization	Lower ROI due to higher hardware costs and limited resource optimization	High ROI, optimized resource use through flexible node and media support
Hardware Refresh	All servers must be refreshed at once	Servers can be refreshed gradually, one at a time, as needs change

Leveraging Mixed Storage Media: TLC, QLC, and HDDs

The Media and Node Flexibility of Ultraconverged

A key strength of UCI is its ability to support a variety of storage media, including TLC (Triple-Level Cell) flash, QLC (Quad-Level Cell) flash, and traditional HDDs. Each storage type offers unique benefits, and UCI enables IT teams to assign workloads to the most appropriate media, optimizing cost and performance without compromise.

TLC NVMe Flash: High-performance, high-endurance TLC flash is ideal for applications requiring frequent access to data, such as real-time analytics or transactional databases. UCI platforms allocate TLC flash where speed is critical.
QLC NVMe Flash: Cost-effective and high-density, QLC flash can store large datasets with minimal expense. QLC media, like Solidigm’s new 60TB+ QLC drives, is optimal for workloads with significant storage needs but lower performance requirements.
HDDs: HDDs remain a cost-effective choice for archival storage and backup, as they offer high capacity without the expense of flash storage. UCI allows organizations to assign archival or backup data to HDDs, reducing costs and freeing up flash resources for more demanding tasks.

In a recent evaluation, StorageReview verified VergeOS’s multi-media support, showcasing its flexibility to handle diverse storage types within a single environment. Readers can watch our on-demand webinar, in which VergeIO, StorageReview, and Solidigm discuss the test results and how these media options enhance the platform’s performance. Click here to register for the on-demand session.

Scaling Storage and Compute Independently with Mixed Node Types

UCI supports mixed node types, enabling independent scaling of compute and storage resources. Traditional HCI solutions require identical nodes for expansion, which is inefficient for businesses with unequal compute and storage demands. For example, data-intensive applications may need more storage without extra compute, whereas HPC tasks might require more compute with less storage.

In UCI, storage-heavy nodes or compute-heavy nodes can be added independently within the same instance, enabling organizations to scale up only what they need. This flexibility offers significant advantages for specific workloads:

Data Lakes and Analytics: Storage-heavy nodes provide the capacity required for large data lakes, while compute-heavy nodes and GPU-heavy nodes can seamlessly access the storage, creating a powerful path to analytics, machine learning (ML), and AI workloads—all supported by the media and node flexibility of Ultraconverged Infrastructure (UCI).
Virtual Desktop Infrastructure (VDI): Compute-heavy nodes can handle the CPU resources needed for VDI. In contrast, fewer storage-heavy nodes are used for backend storage, ensuring cost-effective scaling without excess.
Backup and Archival: Storage-heavy nodes offer the necessary space for long-term backup and archival data without requiring additional compute resources. When paired with GPU-heavy nodes, this configuration provides a high-capacity, cost-efficient foundation supporting AI-driven data analysis or data mining when needed.

This combination of mixed nodes enables organizations to flexibly support a wide range of workloads, from storage-intensive tasks to GPU-powered analytics and AI applications, all while optimizing resource use and reducing overprovisioning.

ioOptimize: Maximizing Efficiency in Mixed-Node and Mixed-Media Environments

In VergeIO’s UCI implementation, VergeFS boosts efficiency by allowing data to be placed across various media and optimizing computing and storage use. IT can allocate performance-critical data to TLC flash and assign archival data to QLC flash or HDDs. Additionally, IT can direct high-performance workloads to compute-heavy nodes, freeing storage-heavy nodes for data-intensive applications. This management prevents resource bottlenecks and maximizes ROI throughout the infrastructure.

The Advantages of UCI’s Flexibility in Media and Node Types

Combining mixed storage media and mixed node types allows UCI to deliver several essential benefits for modern data centers:

Cost Efficiency: Organizations can optimize storage costs by matching storage media to workload requirements without compromising performance. High-density QLC and HDDs help reduce expenses, while high-performance TLC flash is allocated to applications that truly need it.
Scalability: Mixed node types allow organizations to scale only the resources they need, adding storage or compute independently for greater scalability and control over infrastructure growth.
Enhanced Flexibility: The media and node flexibility of Ultraconverged Infrastructure (UCI) allows businesses to fine-tune infrastructure according to specific workload requirements, reducing waste and maximizing resource utilization.
Future-Proofing: UCI supports a broad range of storage and compute configurations, allowing businesses to adopt new storage technologies and accommodate changing needs over time, ensuring the infrastructure remains resilient and adaptable.

Conclusion: Ultraconverged Infrastructure for True Flexibility and Efficiency

Flexibility is essential in today’s complex IT landscape. Ultraconverged Infrastructure (UCI) provides media and node flexibility unmatched by traditional HCI, supporting high-performance TLC flash, high-capacity QLC flash, and cost-effective HDDs. UCI allows businesses to mix compute-heavy and storage-heavy nodes, scaling resources to meet real-world demands and reduce costs.

Solutions like ioOptimize enhance UCI’s effectiveness by optimally placing data and maximizing resource efficiency across mixed-node environments. By adopting UCI, businesses attain a future-ready infrastructure that scales flexibly, aligns with workload needs, and minimizes overprovisioning—ideal for organizations transitioning from VMware to a more adaptable, cost-effective platform.

Filed Under: Storage Tagged With: HCI, Storage, UCI

October 15, 2024 by George Crump

The StorageReview VergeIO lab results show that VergeOS meets the demands of IT professionals looking for a VMware Alternative. StorageReview’s evaluation revealed that VergeIO not only matches VMware’s performance; it surpasses it. Their testing shows that VergeIO is a viable choice for those seeking a more efficient, high-performance solution at a lower cost.

The VergeIO Review Process

StorageReview’s in-depth evaluation used an advanced hardware setup that included Solidigm SSDs, AMD EPYC processors, and Giga Computing’s dense liquid-cooled nodes. The lab environment was designed to replicate the real-world demands of enterprise IT infrastructure, emphasizing high-performance workloads like virtualization, VDI, and database management.

The goal was to assess VergeIO’s ability to deliver seamless performance while evaluating its built-in disaster recovery features and ease of use.

Highlights: StorageReview VergeIO lab results

1000 VM Bootstorm in 71 Seconds
VergeIO was able to boot 1,000 VMs in just over a minute—an impressive feat that highlights its ability to handle high-demand workloads efficiently. This benefits enterprises deploying large VDI environments or managing dynamic workloads where quick spin-up times are critical.
6.9 GB/s Write Performance
VergeIO’s platform achieved 6.9 GB/s write speeds during testing. For organizations dealing with large data volumes, this means more efficient data transfers and reduced latency for write-intensive applications. Solidigm’s SSDs were crucial in achieving these results, demonstrating the synergy between VergeIO’s software and modern hardware.

StorageReview VergeIO Lab Results: Scalability and Cost Efficiency:

IT professionals need a high-performance alternative to VMware that offers speed, scalability, and cost-effectiveness. VMware’s licensing can be intricate and expensive, particularly for large organizations. In comparison, VergeIO’s per-node licensing is more straightforward and predictable. VergeIO’s base license encompasses all advanced features, including storage tiering and disaster recovery, without incurring extra costs for add-ons.

VergeIO’s model is at least 50% cheaper than VMware’s, providing enterprises a straightforward way to lower infrastructure costs without sacrificing performance. This pricing transparency is a significant advantage for companies looking to scale efficiently.

StorageReview VergeIO Lab Results: vSAN with Storage Tiering

VergeIO offers storage tiering as part of its base package, allowing IT teams to optimize data placement based on performance needs. Solidigm SSDs were used in the lab to create two distinct storage pools—Tier 1 with high-performance TLC SSDs and Tier 2 with larger, more cost-effective QLC SSDs. VergeIO’s ability to seamlessly manage different storage types within the same platform simplifies operations while ensuring that performance-critical applications can access the fastest available storage.

VergeIO’s vSAN includes no additional licensing fees for consumed capacity, setting it apart from competitors like VMware, which requires higher-tier licenses for high-capacity environments.

StorageReview VergeIO Lab Results: Migration

One of the biggest concerns for organizations migrating away from VMware is the complexity of the migration process. However, StorageReview found that VergeIO makes this transition seamless. The platform includes native integration with vCenter and ESXi APIs, allowing IT teams to sync and migrate VMs with minimal disruption. The ability to retain critical settings like MAC addresses and storage tiers ensures a smooth and efficient transition to VergeIO. Enroll in our hands-on lab to perform a VMware Migration in real life.

StorageReview VergeIO Lab Results: Data Protection and Recovery

Beyond performance, IT professionals require solutions that ensure a critical workload’s protection and recoverability. VergeIO’s built-in disaster recovery capabilities provide a clear advantage. During testing, the platform’s high availability feature demonstrated quick recovery, with VMs becoming available again on another node in 138 seconds after a node failure. When coupled with IOGuardian, even this short time can be potentially eliminated thanks to its ability to provide real-time recovery from multiple simultaneous drive or server failures.

For data protection IOClone, based snapshots enable independent snapshots that aren’t reliant on previous clones, improving recovery times. This independence means that IT can have thousands of active snapshots and retain them indefinitely without impacting performance. VergeOS provides a robust scheduling capability for managing snapshot retention. Enroll in our hands-on lab to learn the full potential of VergeOS Snapshots.

By incorporating disaster recovery directly into the platform, VergeIO eliminates the need for separate solutions, reducing complexity and cost. VergeIO can replicate an entire data center in minutes. This replication includes data and the complete configuration (VM, network settings), enabling fast and comprehensive recovery even on different hardware.

VergeIO simplifies backup and recovery processes with native snapshot and replication features. Its disaster recovery integration within the central platform ensures that backups can be completed quickly, minimizing the risk of data loss and reducing downtime.

For organizations where uptime and data protection are paramount, VergeIO’s ability to seamlessly integrate these capabilities into its core offering—without additional licensing fees—offers significant operational advantages.

The Bottom Line: A VMware Alternative That’s Easier, Faster, and More Affordable

VergeIO’s performance, scalability, and built-in disaster recovery features make it a compelling alternative to VMware. The StorageReview VergeIO Lab Results show that VergeIO performs exceptionally well under real-world conditions, simplifies operations, and reduces costs.

With 50% lower licensing costs, simplified management, and a platform built to handle complex enterprise workloads, VergeIO offers a solution that meets the needs of IT professionals looking to reduce overhead without compromising on performance or protection.

If you’re exploring VMware alternatives, VergeIO is worth serious consideration. To learn more about how VergeIO performed in the lab, register for our webinar on October 23rd with StorageReview and Solidigm. We’ll demonstrate these results firsthand.

For the full StorageReview report, visit StorageReview’s Lab Evaluation.

Filed Under: Storage Tagged With: Alternative, Performance, Storage, VMware

November 1, 2023 by George Crump

Exiting VMware to Eliminate High Storage Costs

Storage, an integral part of a VMware infrastructure, often consumes a hefty chunk of the IT budget, and exiting VMware to eliminate high storage costs may be the best strategy. Exiting VMware can reduce primary storage costs and investments in secondary storage like backup and archive. A VMware alternative can provide better ransomware resiliency and data protection, reducing backup software investments.

Most IT professionals think that high storage costs come from being forced to invest in high-performance dedicated all-flash arrays (AFA). VMware’s vSAN and Nutanix’s hyperconverged infrastructure (HCI) products were supposed to lower storage costs by enabling IT to use off-the-shelf storage media while maintaining enterprise-class features, performance, and availability. Both products fell well short of the goal and, in most cases, are as expensive as the legacy three-tier architecture.

Why Does VMware Have High Storage Costs?

So, what should storage realistically cost? A 15.3 TB NVMe SSD is priced at less than $1,500, meaning 300 TB of high-performance storage comes in at about $30,000.

Why, then, is it well over 5X to 10X that amount to add 300TB of flash capacity to a VMware environment? Dedicated arrays have to cover the cost of the components already in the environment (CPU power, networking, memory). HCI hasn’t lived up to expectations, either. These vendors still charge too much for their software, which doesn’t deliver the enterprise-class performance or capabilities that IT demands, so IT must overcompensate with more hardware and additional drives.

Decoding the High Storage Costs

The above mentioned, $30,000 should be the entire cost to add 300TBs of high-performance storage. Most customers have plenty of excess storage bays in their physical hosts, so they should be able to add these twenty drives without additional hardware costs.

If HCI vendors offered common sense licensing, there would be no additional software charge. The problem is that vSAN and Nutanix don’t have common sense licensing; they constantly charge for add-ons. From a technical perspective, they can’t easily handle adding a mixture of different storage device types to existing servers; they can’t deliver the hundreds of thousands of IOPS potential of the storage, and these solutions often lack the enterprise features that customers expect.

The combination of the failure of HCI to cut costs and the subsequent need for a high-end all-flash array (AFA), balloon the VMware storage budget. The cost of dedicated storage is negatively impacted by:

Markup Pricing: Despite some genuine costs involved in their strategy, dedicated storage array vendors charge exorbitantly for their capacity.
Performance Requirements: Their software inefficiencies necessitate powerful processors to maintain the array’s pace with the performance potential of the media they insert into it.
Feature Implementation: Functions such as deduplication, snapshots, and data protection, while beneficial, are also inefficient and strain performance, demanding even more RAM and processing power.
Unnecessary Server Expenses: Often, organizations have servers with empty drive bays. Despite this, they purchase and maintain separate powerful “storage servers,” incurring extra costs.

The High Cost of Data Protection

Exiting VMware to eliminate high storage costs may also allow you to reduce the high cost of protecting VMware. Because of the inefficiencies in the HCI and dedicated storage array software, customers must also invest heavily in backup and recovery architectures, which include the backup software itself, a backup storage target, and a long-term object storage-based system to immutably store backup data so that it is protected from a ransomware attack.

Ironically, all of the investment in backup infrastructure has done little to stem the tide of ransomware. According to a recent VergeIO survey and study:

It takes most customers 3+ weeks to recover from a ransomware attack
It takes most customers 3+ days to realize they are under attack.

If the infrastructure can’t warn you it is under attack and takes days to realize it is happening, then almost every snapshot contains encrypted data, as do most backups. Pulling the correct data out of these backup copies is manual and time-consuming, which is why most recovery efforts take more than three weeks.

VergeIO’s Ultra-Converged Infrastructure: Eliminating High Storage Costs

VergeIO introduced VergeOS, an ultra-converged infrastructure solution to address the high storage costs dilemma. Boasting a unified, efficient code base, VergeOS promises:

Cost Efficiency: Customers can install server-class media like 15.3TB SSDs in their existing servers for a fraction of typical HCI / dedicated array costs while gaining over 300TB of storage capacity.
Enhanced Performance: With these drives, users can potentially achieve over a million IOPS with improved data protection.
Deduplication: Integrated into its core, VergeOS’s global inline deduplication incurs minimal CPU or RAM overhead.
Snapshot Technology: Powered by IOclone technology, VergeIO’s snapshots act more like data clones, making them a viable backup solution. Add to that, IOfortify, and you have infrastructure-wide ransomware protection.
Virtual Data Centers: VergeIO’s unique Virtual Data Center (VDC) technology encapsulates the entire data center and simplifies tasks like patching, disaster recovery, and ransomware resiliency.

The Verdict: Should You Turn Off Your Array?

For those using HCI, the benefits of VergeOS – from superior performance to reduced costs – make the switch compelling. VergeOS installs on the existing HCI hardware, breathing new life into it while improving performance, data protection, and ransomware resiliency. VergeIO features common sense licensing. VergeOS is licensed per physical host, not the number of processors, cores, RAM, or storage capacity. One license per physical server, and it includes all of the capabilities.

Even dedicated storage array customers will find it more cost-effective to change when weighing the high maintenance costs against VergeOS’s offerings. Most AFA customers tell us they can switch to VergeOS for considerably less than the cost of their current maintenance contract. Combine that with the cost savings of exiting VMware; these customers typically report a 50% reduction in TCO. As one customer said, “I’ll save money just by turning my AFA off.”

In conclusion, making informed decisions is crucial as the IT landscape evolves and storage costs skyrocket. Exiting VMware and considering alternatives like VergeOS can extend the savings beyond the VMware license. Many customers find their storage cost savings justify the migration while paving the way for enhanced performance and greater data resiliency.

Exiting VMware to eliminate high storage costs is one of many reasons to consider a VMware alternative, but you need to take a step-by-step approach. Join us for our webinar, “How to Exit VMware Step-by-Step,” for practical advice on how to develop a VMware exit strategy at a pace that makes sense for your organization.

Filed Under: Storage Tagged With: Storage, VMware