MDS Micro Tech Field Day Presentation

Event: Tech Field Day 1

Appearance: MDS Micro Presents at Tech Field Day 1

Company: MDS Micro

Video Links:

Personnel: Karriem Adams

MDS Micro presents their Cloud Matrix architecture, featuring the Quad-V mini blade system, a highly modular and efficient server solution optimized for virtualization environments. Developed in close collaboration with VMware engineers, the Cloud Matrix is designed to offer enterprise-grade performance while reducing footprint, power consumption, and cost, addressing the evolving needs of data centers.

During the presentation at Tech Field Day 1, Karriem Adams of MDS Micro explained the origins and purpose of the Quad-V system, emphasizing its modular 2U chassis that houses four hot-swappable servers. Unlike traditional blade chassis requiring full racks regardless of actual demand, Quad-V allows enterprises to scale incrementally. Designed in collaboration with VMware architects, Quad-V was engineered to address both performance bottlenecks and space inefficiencies. The Cloud Matrix consists of two main components: the server layer built on these Quad-V blades, and the networking layer powered by Zeego switches providing high-bandwidth InfiniBand connectivity. This combination enabled data center setups, such as VMware’s 2009 VMworld event booth, to consolidate operations from 14 racks to just half of one.

Adams also highlighted the Cloud Matrix’s technical advantages, including integrated InfiniBand for high-speed I/O, support for multiple processor platforms within a single chassis, and a hot-swappable, low-power design with on-board virtualization support. The solution supports various storage configurations, from traditional drives to SSDs, while remaining hardware agnostic, ensuring seamless integration with existing data center infrastructure. With support for up to 384 GB of RAM and 48 cores per chassis, and software like SimpleView for unified IPMI-based management, Cloud Matrix delivers high density, scalability, and power efficiency. The system can fit 84 servers, up to 672 cores and over 8TB of RAM in a single 42U rack, making it a compelling option for enterprises looking to enhance their virtualization capabilities while maintaining flexibility and reducing TCO.


SolidFire Presents at Tech Field Day 8

Event: Tech Field Day 8

Appearance: SolidFire Presents at Tech Field Day 8

Company: SolidFire

Video Links:

Personnel: Dave Wright

Dave Wright, founder of SolidFire, presents the company’s architecture and vision at Tech Field Day 8. Drawing on his prior experience at Jungle Disk and Rackspace, Wright outlines the key storage challenges in cloud environments and how SolidFire built a solution from the ground up to meet the performance, efficiency, and automation needs of large-scale cloud service providers.

The presentation begins with Dave Wright recounting his background at Jungle Disk and Rackspace, where early exposure to cloud hosting challenges, especially around storage, served as the impetus for founding SolidFire. At Rackspace, he observed a lack of scalable, cost-effective, and high-performance storage solutions suitable for cloud infrastructures. These insights led SolidFire to focus specifically on building a platform designed for service providers, addressing the inadequacies of traditional enterprise storage systems that struggled with multi-tenancy, cost control, and consistent performance at cloud scale.

Wright explains that SolidFire’s value proposition is built on three foundational pillars: performance, efficiency, and automation. To handle multi-tenant cloud environments, SolidFire prioritizes consistent and guaranteed performance through a concept called “performance virtualization,” separating performance controls from capacity configuration. The system allows configuration of both IOPS and throughput per volume, with adjustable burst capabilities and enforcement mechanisms. This ensures predictable application behavior while mitigating noisy neighbor issues—a key capability lacking in traditional SANs. Equally important is efficiency: SolidFire’s scale-out architecture allows high utilization (up to 85% or more) and leverages real-time compression, deduplication, and thin provisioning to reduce unnecessary resource consumption.

Perhaps most distinctively, SolidFire embeds automation deeply into its DNA. Wright emphasizes that for service providers operating at cloud scale, human management does not scale. SolidFire circumvents this by leading with a comprehensive REST-based API for deployment, provisioning, and configuration. This enables rapid instrumentation of storage, akin to how compute is automated in data centers. Wright demonstrates that new nodes can be added or removed with minimal manual intervention, and SolidFire integrates natively with emerging cloud orchestration platforms like OpenStack and VMware vCloud.

From a technical perspective, SolidFire’s architecture diverges significantly from traditional RAID-based SAN technologies. Data is split into 4K blocks, deduplicated at ingest based on cryptographic hashes, and mapped to metadata volumes via a distributed system of BINs. The system utilizes standard x86 hardware with SSD-only drives, relying on replication instead of RAID for redundancy. This minimizes rebuild times—sometimes to mere seconds—while limiting the impact of failures to the smallest possible subset of the system. Storage snapshots and clones are created via metadata copies without performance degradation or data duplication, and the system is engineered to facilitate efficient backup and replication capabilities through its block identifier approach.

In conclusion, Wright underlines that SolidFire’s use of commodity hardware, all-flash design, and intelligent software stack allows a competitive price point—under $5 per gigabyte for effective, usable capacity. By combining high performance, cost-efficiency, granular performance guarantees, and true API-driven automation, SolidFire offers a compelling solution for cloud providers dealing with the complexities of multi-tenant environments and application performance SLAs. Wright’s presentation not only outlines a robust technical architecture but also makes a case for the alignment of SolidFire’s design principles with the real-world needs of next-generation cloud infrastructure providers.


Arista Presents at Tech Field Day 8

Event: Tech Field Day 8

Appearance: Arista Presents at Tech Field Day 8

Company: Arista

Video Links:

Personnel: Andre Pech, Jayshree Ullal

In this session, Arista explores its novel network software architecture, Extensible Operating System (EOS), emphasizing its unique design, extensibility, and integration with existing systems. The presentation, led primarily by software engineer Andre Pech and joined by CEO Jayshree Ullal, outlines how EOS allows for flexible, resilient, and developer-accessible network configuration and management. Arista also demonstrates real-world use cases and answers technical questions from industry experts.

Andre Pech begins the presentation by introducing Arista’s Extensible Operating System (EOS), highlighting its architectural design that draws lessons from the Unix paradigm. He emphasizes that EOS is based on an unmodified Linux kernel with separate, independent user-space processes handling different functionalities. Unlike traditional network operating systems that often rely on monolithic kernels susceptible to crashes, Arista’s architecture isolates failures, allowing individual processes to crash and restart independently without bringing down the whole system. This modular design facilitates better resilience, faster upgrades, and easier debugging, fostering a server-like robustness in networking hardware.

A focal point of the presentation is SysDB, Arista’s system database, which functions as the central state repository for EOS. SysDB avoids the need for complex message-passing systems by storing the entire network configuration and status data in a single place. Agents or applications mount parts of SysDB and are notified of relevant state changes. This architecture not only improves modularity and fault tolerance but also enables features such as in-service software upgrades (ISSU) by simply updating and restarting individual processes without interrupting the system’s operation. Pech illustrates this with a live demo involving the Spanning Tree process, showing that killing and restarting it results in no disruptions to forwarding behavior due to state persistence in SysDB.

The session also covers EOS’s extensibility at multiple levels, from native Linux access to scripting with the EOS CLI, and deeper integration through Python APIs that interface directly with SysDB. Pech demonstrates how engineers can use standard Linux tools like bash, TCPdump, and scripting languages to access, monitor, and extend switch functionalities—treating the switch almost like a regular Linux server. One highlight includes a demonstration of using XMPP (Jabber protocol) to create a chat-based interface where switches can exchange information, enabling group-based configuration, real-time queries, and management across the network.

Further discussion explores Arista’s operational tools and partnerships, including integration with VMware ESX through VM Tracer, which dynamically updates VLAN assignments and configurations based on virtual machine movements. Attendees inquire about support for L3 protocols, port mirroring, and debugging capabilities, revealing Arista’s robust feature set tailored for complex data center environments. Pech stresses the focus on open standards and the drive toward features like VXLAN and NVGRE for scalable virtual networking, keeping Arista at the forefront of network virtualization and multi-tenant cloud solutions.

CEO Jayshree Ullal joins the session toward the end to address business and industry trends, explaining how Arista’s strategy aligns with evolving data center needs without vendor lock-in. She emphasizes that the company’s focus remains on best-of-breed, standards-based, software-driven switching. Questions about multi-vendor environments, competition with Cisco and Dell, and future support for technologies like Trill and Fibre Channel over Ethernet are candidly addressed. Ullal insists that Arista’s software-centric approach, combined with transparent pricing and elite support from engineers themselves, positions them uniquely in a crowded networking market poised for next-generation growth.


Pure Storage LIVE at Tech Field Day 8: Silicon Valley

Event: Tech Field Day 8

Appearance: Pure Storage Presents at Tech Field Day 8

Company: Pure Storage

Video Links:

Personnel: John Colgrove, John Hayes, Michael Cornwell, Scott Dietzen

Pure Storage discussed their vision and technology at Tech Field Day 8, advocating for the widespread adoption of all-flash storage in the enterprise. The company outlined how their purpose-built solution delivers significant improvements in performance, power consumption, and cost-efficiency over traditional spinning disks and hybrid storage systems. The presentation delved into the limitations of 15K RPM drives and the need to move beyond performance disks by eliminating legacy architectures and leveraging flash memory’s capabilities.

During the presentation, CEO Scott Dietzen explained that while CPU performance has dramatically increased over the past 15 years, disk I/O has lagged, creating a major bottleneck. Current storage systems spend unnecessarily on underperforming disk drives, especially for high-performance demands. Existing all-flash solutions suffer from limiting costs and poor scalability in enterprise environments. The Pure Storage team emphasized solving this with innovation on two fronts: making flash as reliable and available as disk, and reducing the total cost of ownership through aggressive data reduction, dynamic RAID schemes, and flash optimization.

Cofounders John Colgrove and John Hayes, along with flash technologist Michael Cornwell, detailed the architecture behind the Pure Storage solution, including advanced inline data deduplication and compression, RAID 3D for increased reliability, and an operating environment called Purity, which virtualizes flash while supporting real-time data integrity and performance. The system supports sub-millisecond latency and scales across enterprise workloads. The team also showcased a high-density VMware demo, consolidating 800 VMs with high deduplication ratios and exceptional performance, positioning Pure Storage as an enterprise-class contender with innovative software-driven flash storage.


DataDirect Networks presents at Tech Field Day 8 (Part 2)

Event: Tech Field Day 8

Appearance: DDN Presents at Tech Field Day 8

Company: DataDirect Networks

Video Links:

Personnel: Jean-Luc Chatelain, Jeff Denworth

In this presentation from Tech Field Day 8, DataDirect Networks (DDN) demonstrates the deployment and management of their Web Object Scaler (WOS) system. The focus is on how WOS can quickly be installed and brought online at customer sites, providing object storage services through a streamlined setup process. The WOS architecture allows for scalable, policy-driven object storage across geographically distributed nodes, all managed through a single, intuitive interface.

The setup process begins by connecting new WOS nodes to the network and assigning IP addresses, after which an administrator can access the unified management interface via a browser. This UI provides centralized management for all nodes in a global cluster, regardless of location. The user can configure new zones that reflect physical data center locations and add nodes to these zones with simple drag-and-drop functionality. Once added, the cluster automatically integrates the nodes with minimal configuration and no service downtime. Policies drive the storage behavior, enabling administrators to define data replication rules for performance and reliability, including synchronous and asynchronous modes where replicas are written to local or distributed nodes.

To showcase the capabilities and resilience of WOS, the demo includes a latency test involving a simulated failure of a local node and automatic failover to a remote node without service interruption. This demonstrates WOS’s built-in data locality and latency optimization, which relies on each node’s dynamic in-memory latency map to route data access efficiently. The system ensures immediate availability and minimum latency by always selecting the nearest available replica of the data. DDN’s WOS offers a robust and flexible backend for object storage needs, delivering scalability and performance that are critical for modern enterprise and cloud environments.


DataDirect Networks presents at Tech Field Day 8 (Part 1)

Event: Tech Field Day 8

Appearance: DDN Presents at Tech Field Day 8

Company: DataDirect Networks

Video Links:

Personnel: Jean-Luc Chatelain, Jeff Denworth

DataDirect Networks (DDN) introduces their company, product philosophy, and key technologies during their presentation at Tech Field Day 8. In part one, Jeff Denworth and Jean-Luc Chatelain focus on the evolution of hyperscale data storage needs, the shortcomings of traditional storage infrastructures, and how DDN’s platforms are built to meet the demands of emerging data environments including cloud, high-performance computing, life sciences, government, and media.

During the presentation, Jeff Denworth establishes DDN’s identity as a $200+ million privately held company specialized in ultra-high-performance data solutions, targeting uncommon yet profound data challenges in industries like genomics, supercomputing, video surveillance, and cloud content services. He introduces DDN’s storage array architecture—SFA (Storage Fusion Architecture)—which supports high throughput and IOPS using a smaller number of SSDs, thereby offering efficiency and scale. Jeff then transitions to discuss WAN Object Scaler (WOS), DDN’s object storage platform specifically built to solve performance and scalability limits in traditional SAN and file systems when managing billions of data objects at petabyte scales.

Jean-Luc Chatelain takes over to delve into the limitations of file-based systems when applied to massive-scale data issues, such as inefficiencies in I/O performance, metadata overhead, configuration complexity, and rising total cost of ownership. He introduces WOS as a “noFS” (no file system) hyperscale object storage solution that enables distributed, geo-replicated, self-healing storage with flat namespace architecture, low overhead, and efficient metadata handling. With practical use cases from defense agencies to surveillance systems to global media companies, WOS demonstrates significant performance gains, architectural simplicity, and linear scalability, addressing the real-world needs of organizations facing exponential data growth in a distributed and latency-sensitive world.


Symantec Presents at Tech Field Day 8

Event: Tech Field Day 8

Appearance: Symantec Presents at Tech Field Day 8

Company: Symantec

Video Links:

Personnel: Don Angspatt, Jerry Gallin, Oscar Wahlberg, Ranga Rajagopalan, Saveen Pakala

At Tech Field Day 8, Symantec presented a comprehensive update on its Storage and Availability Management portfolio. Featuring the soon-to-be-released 6.0 version of Storage Foundation and Veritas Cluster Server, the focus was on supporting private cloud adoption, enhancing resiliency in multi-tiered applications, and delivering improved storage optimization. The event showcased new technologies geared toward helping enterprises transform their existing infrastructure into more agile, efficient, and resilient platforms for mission-critical applications.

During the presentation, Symantec executives and product managers emphasized the strategic importance of getting “the private cloud you want from the infrastructure you’ve already got.” This message was demonstrated through a series of sessions and live demos highlighting new features in version 6.0. These included file system-level deduplication and compression for primary storage, automation of recovery for multi-tier applications using Virtual Business Services (VBS), fast failover support for Windows environments, and enhanced integration between Application HA and Backup Exec for VM auto-restore. These enhancements targeted the challenges faced by enterprises amid mounting pressure to virtualize and evolve toward private cloud architectures without sacrificing the performance or stability of their mission-critical environments.

The event also included lively and candid discussions among the Tech Field Day delegates, many of whom questioned traditional enterprise assumptions embedded in Symantec’s messaging. Several attendees expressed skepticism about large-scale private cloud transitions or suggested that traditional “glass house” IT centers are no longer leading innovation. Symantec acknowledged that while cloud adoption isn’t always immediate, their solutions, like VBS and multi-platform clustering, offered practical bridges between legacy systems and modern cloud-readiness. The session concluded with strong interest in the technical depth of the demos and an open invitation to explore further, positioning Symantec’s 6.0 release as a significant step in merging enterprise reliability with the agility demanded by modern IT.


Nasuni Tech Field Day Presentation: The Cloud Inside the Storage Controller

Event: Tech Field Day 8

Appearance: Nasuni Presents at Tech Field Day 8

Company: Nasuni

Video Links:

Personnel: Andres Rodriguez

For years, the traditional storage controller has been a game of RAM and disk. Put the cloud inside and what you have is a third component.

In this clip from Tech Field Day 2011 Nasuni CEO, Andres Rodriguez, explains what the next generation of storage controllers looks like and how integrating the cloud changes everything.


Nutanix Presents at Tech Field Day 8

Event: Tech Field Day 8

Appearance: Nutanix Presents at Tech Field Day 8

Company: Fusion-io, Nutanix

Video Links:

Personnel: Ajeet Singh, Dheeraj Pandey, Mitch Crane, Mohit Aron

Nutanix CEO and co-founder Dheeraj Pandey presented at Tech Field Day 8, introducing the company’s converged infrastructure solution that merges compute and storage into a single scalable appliance designed for virtualized environments. By drawing inspiration from architectures used by hyperscale cloud providers and adapting them for enterprise needs, Nutanix offers simplified management and improved performance, aiming to eliminate the complexity and inefficiencies tied to traditional SAN-/NAS-based datacenter approaches.

In the presentation, Pandey used a humorous animation to illustrate the traditional virtualization stack’s struggles with SAN management and performance bottlenecks. This was contrasted with Nutanix’s approach of co-locating compute and storage within each node, eliminating the need for separate storage networks or devices. Nutanix’s architecture was described as highly cloud-inspired, with its foundation in commodity x86 hardware, leveraging SSDs and Fusion I/O for caching, and intelligently managing data through MapReduce for performance and availability. This hyper-converged model allows enterprises to enjoy cloud economics and scale-out simplicity in a turnkey solution—all while using standard protocols like NFS and iSCSI.

Throughout the session, the Nutanix team detailed the system’s underlying technology and highlighted its benefits for mid-sized enterprises and service providers, including simplified deployment, linear scaling, and a unified management interface. The appliance’s design supports high-performance virtualization and VDI using local storage tiered by access frequency, while offering enterprise features like snapshots, failover, and intelligent data locality management. Nutanix emphasized a generalist-friendly approach to infrastructure, aiming to empower virtualization and IT ops teams with a self-healing, low-maintenance, one-box platform, and concluded by demonstrating a live UI walkthrough and discussing real-world deployments in legal and financial firms.


Arkeia presents at Tech Field Day 8

Event: Tech Field Day 8

Appearance: Arkeia Presents at Tech Field Day 8

Company: Arkeia Software

Video Links:

Personnel: Bill Evans

Arkeia Software presented at Tech Field Day 8 with a focus on the company’s background, data deduplication technology, and strategies for offsite data protection. Bill Evans, representing product management, outlined the company’s mid-market focus and discussed their innovations in virtualization and backup methodologies. A significant portion of the presentation was dedicated to explaining Arkeia’s approach to deduplication, particularly their unique technology called “progressive deduplication,” which is designed to reduce bandwidth and accelerate data movement offsite for cloud backup solutions.

Arkeia Software, founded in 1996, specializes in network backup solutions for both physical and virtual platforms. They deliver backup software and appliances and have built a strong presence in North America and Europe. Their product strategy emphasizes virtualization, with announcements about support for platforms like VMware vSphere, Microsoft Hyper-V, Citrix, and more. A major technological advancement discussed was Arkeia’s progressive deduplication, acquired from Kadena Systems. This method improves upon traditional fixed and variable block deduplication by allowing overlapping blocks of a fixed size, enabling higher compression rates and faster data processing. It includes on-the-fly hashing and progressive matching to minimize workload and avoid redundant data storage during incremental backups.

Throughout the presentation, Evans stressed the importance of deduplication in reducing cloud bandwidth usage, especially as mid-market companies look to replicate data to the cloud rather than rely on traditional tape-based offsite storage. He highlighted how deduplication improves storage efficiency across time, systems, and applications, particularly in virtualized environments. The discussion included the challenges and benefits of source-side vs. target-side deduplication and considerations about restore points, agent usage in backups, and risks like rogue administrators or malicious deletions. The session concluded with insights on when to use tape versus cloud solutions, noting that while cloud is viable for small data volumes, larger deployments often still require physical transport due to bandwidth constraints.


Introducing Actifio with Ash Ashutosh at TFD4

Event: Tech Field Day 4

Appearance: Actifio Presents at Tech Field Day 4

Company: Actifio

Video Links:

Personnel: Ash Ashutosh

Ash Ashutosh, Founder, President, and CEO of Actifio, introduces the company and its technology. This video was recorded as part of Gestalt IT Tech Field Day 4 on November 12, 2010.

In his presentation at Tech Field Day 4, Ash Ashutosh introduced Actifio as a company aiming to revolutionize the way enterprises handle data management, particularly around the often redundant and fragmented process of making multiple data copies for backup, disaster recovery, analytics, compliance, and development. Ash outlined the problem with the current paradigm, explaining how most organizations spend multiple times more on managing data copies than on the original data, storing as many as 13 to 120 copies across different tools and use cases. Each of these tools—whether for backup, snapshots, replication, or test environments—creates and manages its own copy in isolation, resulting in inefficiency, complexity, and ballooning costs. Actifio’s approach is to unify these processes by using a single intelligent platform that virtualizes data and allows various applications to interact with a consistent, optimized data store.

Actifio’s core innovation is a system built around four fundamental primitives of data management: copy, store, move, and restore. The company’s value proposition is in consolidating these activities across traditionally siloed backup and storage systems into one “virtual data pipeline.” Through a new object-based file system and sophisticated metadata that includes application context and lifecycle management policies (SLAs), Actifio separates data management from the underlying storage infrastructure. This allows companies to use existing storage hardware or cloud storage providers while dramatically optimizing how data is copied and moved. By allowing instant restore and dramatically reducing the time and resources needed for backup and recovery, Actifio addresses persistent pain points such as long backup windows and slow recovery times. The system is designed to function transparently within existing IT environments, enabling incremental, non-disruptive adoption.

Ash emphasized usability and simplicity by aligning the user experience with modern paradigms, notably avoiding outdated storage concepts like “LUNs” and relying instead on SLA-based, policy-driven workflows. The platform provides seamless support for physical and virtual environments, enables restoration across varied infrastructures, and dramatically cuts network and storage overhead through deduplication and compression. Actifio’s design allows it to be applied in scenarios ranging from local data protection to remote disaster recovery and active-active multi-site environments without imposing a specific disk solution, reinforcing its storage-agnostic philosophy. Ash concluded by noting that Actifio invested significant engineering effort—backed by 16 patents filed—into making the platform robust and scalable while maintaining a clear focus on user needs, honed through collaboration with dozens of early customers in the Boston area before expanding its market presence.


Cisco UCS Roundtable

Event: Tech Field Day 2

Appearance: Cisco Presents at Tech Field Day 2

Company: Cisco Datacenter

Video Links:

Personnel: Bas Raayman, Ed Saipetch, Edward Haletky, Greg Ferro, Jason Boche, John Obeto, Omar Sultan, Scott Lowe, Simon Seagrave, Stephen Foskett

Cisco’s Unified Computing System (UCS) redefines the economics and operational model of data centers. It combines computing, networking, and storage infrastructure with management and virtualization in a single system to reduce total cost of ownership and increase agility. This roundtable session at Tech Field Day 2 featured a discussion of UCS technologies by leading industry experts and bloggers.

During the roundtable, participants discussed various defining attributes of Cisco UCS, including simplified management, scalability, and the utility of service profiles for hardware abstraction. These service profiles enable administrators to assign configurations and identities to blades which can be quickly swapped or moved between chassis with minimal disruption and no manual reconfiguration. This abstraction not only simplifies hardware replacement and upgrades but also facilitates operational consistency across deployments. For large enterprises, this ease of scalability and minimal intervention reduces provisioning time and administrative overhead, allowing departments to quickly deploy services without depending on multiple IT teams.

Another significant discussion focused on the secure multi-tenancy capabilities of UCS. While the term conveyed support for isolated environments for different internal business units through features like vFilers and quality of service (QoS) policies, several roundtable participants pointed out that this model does not fully address broader security concerns—particularly those involving data confidentiality and integrity across external organizations. Experts emphasized the lack of comprehensive definitions and standards in cloud and virtualization security, and although Cisco shows promise by being involved in efforts like CloudAudit and the Cloud Security Alliance, ambiguity remains. Ultimately, the consensus suggested that while UCS is a forward-thinking platform with strong potential, particularly in terms of orchestration and manageability, its security model is still evolving and may not meet every organization’s rigorous requirements just yet.

Lastly, the conversation also touched on UCS’s approach to integrating Fibre Channel over Ethernet (FCoE), offering compatibility with existing storage networks while consolidating bandwidth. This convergence potentially reduces infrastructure and overhead but still requires precise zoning and provisioning on both Fibre Channel and iSCSI. Despite its powerful features, UCS’s appeal remains dependent on customer understanding and awareness; some IT professionals noted confusion about its value proposition beyond hardware convergence, particularly in smaller enterprises. However, the collective agreement was that UCS serves as a strong foundation for modern data centers, with potential cost advantages and streamlined operations—especially when the full scope of its features and integration models are thoroughly understood and leveraged.


Carter George presents Ocarina Networks capacity optimization in 2009 at TFD1

Event: Tech Field Day 1

Appearance: Ocarina Networks Presents at Tech Field Day 1

Company: Ocarina Networks

Video Links:

Personnel: Carter George

Recorded at Tech Field Day 1 on November 13, 2009, Carter George presented Ocarina Networks’ capacity optimization technology. He discussed various data deduplication and compression approaches, contrasting them with Ocarina’s “application-aware” technology. Ocarina Networks, which was later acquired by Dell, focused on creating free space on existing storage rather than selling new storage solutions. George emphasized the importance of integrating their technology with existing storage vendors to ensure seamless operation and user transparency. This presentation is part of the “Best of Tech Field Day” series.

Carter George explained that Ocarina’s solution, named the “ecosystem,” stands for extract, correlate, and optimize. This process involves decompressing files to their raw data, deduplicating them, and then applying content-aware compression. Unlike traditional deduplication methods that operate on the zeros and ones stored on disk, Ocarina’s approach extracts and decompresses data to find duplicates that standard methods might miss. This is particularly effective for files that are already compressed, such as Office documents, videos, and images. By identifying and removing duplicates at a more granular level, Ocarina can significantly reduce storage requirements.

George also highlighted the flexibility and efficiency of Ocarina’s technology. The solution can be deployed as an appliance or software, depending on the storage environment. It reads files from existing storage, processes them to reduce size, and writes them back, ensuring data integrity through checksum comparisons. The technology supports various file types and can be tuned for specific use cases, such as photo sites or virtual machine environments. Additionally, Ocarina’s algorithms include specialized methods for images and videos, offering both lossless and visually lossless compression options. This comprehensive approach allows Ocarina to provide substantial storage savings and improved data management for large-scale data centers and diverse storage environments.


Virsto Technology Preview – VF Cache on VNX

Event: Storage Field Day 2

Appearance: Virsto Presents at Storage Field Day 2

Company: Virsto

Video Links:

Personnel: Brian Martin


Virsto for vSphere 2.0

Event: Storage Field Day 2

Appearance: Virsto Presents at Storage Field Day 2

Company: Virsto

Video Links:

Personnel: Mike Gigante


Introduction to Virsto Software

Event: Storage Field Day 2

Appearance: Virsto Presents at Storage Field Day 2

Company: Virsto

Video Links:

Personnel: Mark Davis


Virsto Technical Architecture

Event: Storage Field Day 2

Appearance: Virsto Presents at Storage Field Day 2

Company: Virsto

Video Links:

Personnel: Brian Martin


Solid-state considerations for scale-out architectures

Event: Storage Field Day 2

Appearance: NexGen Presents at Storage Field Day 2

Company: NexGen

Video Links:

Personnel: John Spiers, Kelly Long


NexGen chose a hybrid design for long term architectural viability

Event: Storage Field Day 2

Appearance: NexGen Presents at Storage Field Day 2

Company: NexGen

Video Links:

Personnel: John Spiers, Kelly Long

Storage QoS is more than just a way to isolate workloads on a shared storage system to guarantee performance levels, it is the key enabler for long-term hybrid architecture viability, John Spiers, Co-Founder and CEO and Kelly Long, Co-Founder and CTO describe why. Topics include hybrid, benefits of storage QoS, storage media proliferation, hybrid storage system challenges, QoS controlled tiering and caching, storage efficiency and affordability.


Managing Shared Storage Performance with NexGen

Event: Storage Field Day 2

Appearance: NexGen Presents at Storage Field Day 2

Company: NexGen

Video Links:

Personnel: John Spiers, Kelly Long

Chris McCall, Vice President of Marketing discusses how NexGen provides new capabilities that allow customers to isolate, control, automate, and guarantee performance in a shared storage environment. Topics include storage QoS, and performance service levels.


« Previous Page
Next Page »