|
Qumulo presented at Storage Field Day 8, highlighting their innovative approach to data storage. CEO Peter Godman discussed their data-aware primary storage system, emphasizing scalability, reliability, and a SaaS model with frequent updates. CTO Aaron Passey focused on the Qumulo Core Block System, optimizing data protection and reliability through rapid restriping and efficient component use.
Qumulo’s shared-nothing clustered architecture, utilizing 5GB data chunks and a hybrid SSD/HDD model, enables fast rebuilds and optimized I/O performance. Their metadata analytics, integrated directly into the file system, provide real-time insights and efficient data management. Qumulo’s scalable solutions, showcased at Storage Field Day 8, offer user-friendly interfaces and advanced analytics for large-scale data environments.
Qumulo Overview with Peter Godman
Watch on YouTube
Watch on Vimeo
In this presentation by Qumulo at Storage Field Day 8, CEO Peter Godman discussed their innovative approach to data storage, emphasizing their company’s evolution and vision. Founded in 2015, Qumulo aims to address the growing complexities associated with managing large-scale data environments. Godman highlighted that the traditional means of handling data have become increasingly cumbersome, with businesses losing sight of what they are storing and struggling to manage the sheer volume of information. By engaging with approximately 600 end users, Qumulo learned that while storage administration is becoming more manageable, understanding and maintaining large data sets is becoming more difficult, presenting an opportunity for growth and innovation in the storage sector.
Godman shared insights into Qumulo’s business model and the technology lies at the heart of their data solutions. Central to their approach is a data-aware primary storage system that integrates real-time analytics, allowing organizations to grasp the intricacies of their data in a transparent manner. This has been designed so that users can acquire scalable and reliable storage without being locked into proprietary hardware, making their solutions adaptable across various environments. Moreover, Qumulo operates on a SaaS model, pushing for continuous development by releasing updates every two weeks, which not only enhances functionality but also aligns with the modern expectations of enterprise software.
The session concluded with a clear message on Qumulo’s commitment to providing user-friendly, high-performance storage solutions tailored for sectors such as media and entertainment and oil and gas, where large quantities of unstructured data are prevalent. Godman reasserted the importance of making data manageability less of a chore for users by focusing on invisible storage that simultaneously elevates data visibility. This innovative philosophy seeks to address the prevalent challenge of understanding data growth and utilization, thus enabling users to derive greater value from their information assets while simplifying the operational aspects of storage management.
Personnel: Peter Godman
Qumulo Storage Physics
Watch on YouTube
Watch on Vimeo
The presentation by Qumulo, titled “Qumulo Storage Physics” and led by Peter Godman, delves into the evolving landscape of storage technologies, particularly focusing on the issues associated with managing unstructured data. Godman emphasizes the significant changes that different storage media have undergone over the years, with attention given to the limitations of hard disk drives (HDDs) versus solid-state drives (SSDs). By tracing the historical development of these technologies, he illustrates how throughput, IOPS (Input/Output Operations Per Second), and storage capacities have evolved, highlighting the challenges faced by businesses as data volumes continue to multiply.
Throughout his talk, Godman addresses the fundamental physics of how storage devices operate, illustrating that while throughput may improve, the IOPS performance of HDDs does not scale proportionately with increased storage capacity. He shares specific statistics over time, showing how the performance of HDDs has stagnated compared to the rapid advancement of SSD technology. While SSDs provide faster performance, they also present cost challenges, as the dollar per capacity ratio remains significantly higher than that of HDDs. This disparity creates a pressing need for innovation in storage solutions to bridge the gap between speed and affordability as organizations increasingly demand faster data access amidst growing datasets.
To combat these challenges, Godman outlines the factors that contribute to reduced performance in storage systems, such as fragmentation, the prevalence of small files, random I/O patterns, and inefficient data protection mechanisms. He emphasizes the importance of understanding and addressing these factors to enhance the performance of storage systems. The presentation concludes with a preview of Qumulo’s upcoming innovations aimed at addressing these inefficiencies in network-attached storage and optimizing system performance. Overall, it advocates for a balanced approach to managing storage that both maximizes speed and affordability, critical for the evolving needs of data-intensive environments.
Personnel: Peter Godman
Qumulo Core Block System
Watch on YouTube
Watch on Vimeo
The presentation by Qumulo at Storage Field Day 8, led by CTO Aaron Passey, focused on the Qumulo Core Block System, emphasizing its capacity for data protection and enhanced reliability. Passey detailed how failure rates and the chances of a storage system failing correlate directly with factors such as restripe time, component reliability, and cluster size. By optimizing these elements, Qumulo aims to achieve high reliability while lowering protection levels, enabling organizations to minimize costs related to data storage while still safeguarding against data loss. This is achieved through rapid restriping and effective use of storage components.
At the heart of Qumulo’s design is a shared nothing clustered architecture that operates efficiently even during drive failures. Passey illustrated how data is organized into 5 GB chunks, which can be mirrored or erasure-coded for protection. This approach allows for a widespread rebuild process across the cluster, utilizing the performance of multiple drives to expedite data recovery effectively. As a result, Qumulo claims rebuild times significantly lower than those of traditional systems—often completing in minutes rather than days, which is a critical factor in maintaining operational efficiency and data availability.
In addition to the technical aspects, Passey addressed the architecture’s advantages in maintaining optimal I/O performance. By leveraging a hybrid model that combines SSDs with traditional spinning disks, Qumulo optimizes read and write operations for both speed and resource efficiency. The system’s ability to handle large block sizes ensures greater utilization of storage resources while minimizing fragmentation. Moreover, Qumulo is actively working towards features such as data-at-rest encryption and enhanced management capabilities, affirming its commitment to evolving its offerings in response to customer needs. Overall, the presentation highlighted Qumulo’s innovative approach to storage solutions that prioritize performance and reliability.
Personnel: Aaron Passey
Qumulo Core Metadata Analysis
Watch on YouTube
Watch on Vimeo
In his presentation on Qumulo Core Metadata Analysis at Storage Field Day 8, Aaron Passey focused on the importance of analytics, particularly metadata analytics, in managing large datasets. He emphasized the need for rapid answers to data-related queries in environments with billions of files, highlighting the limitations of traditional methods such as running disk usage commands or using find commands. As data scales up, these conventional approaches become less effective, leading to significant latencies and outdated information when conducting routine operations like data scans for backups.
To tackle the challenges presented by large-scale data, Passey described Qumulo’s innovative approach, which involves integrating metadata analytics directly into the file system itself. By utilizing aggregates—functions that summarize key attributes of files and directories—Qumulo enhances the efficiency of metadata queries. This method allows for instantaneous access to aggregated data such as total blocks used or the last changed times, drastically reducing the need for lengthy tree scans that plague traditional systems. The ability to query the metadata without exhaustive searches not only saves time but also minimizes I/O overhead on storage systems.
Additionally, Passey addressed how aggregates can enhance the search for files modified within specific timeframes, thereby enabling much more efficient incremental backups. He pointed out that unlike conventional methods that can leave data outdated and irrelevant, Qumulo’s system ensures that metadata remains relatively fresh, generally updated within a minute or less. By avoiding the bottlenecks associated with extensive scans and establishing a system that allows for quick access to detailed file metrics, Qumulo positions itself as a leader in storage solutions, particularly suitable for enterprises managing vast amounts of unstructured data.
Personnel: Aaron Passey
Qumulo Summary and Demo
Watch on YouTube
Watch on Vimeo
Qumulo’s presentation at Storage Field Day 8, led by Peter Godman, highlighted the company’s scalable storage solutions and analytics capabilities. Godman began by demonstrating the ease of building a cluster, showcasing how users can rapidly establish a scalable storage system using Qumulo’s interface. He illustrated this setup process along with the deployment of sample data, emphasizing the user-friendly nature of Qumulo’s configuration and management features, all conducted through a straightforward HTML5-based web interface.
A significant portion of the presentation focused on Qumulo’s advanced analytics features, which are designed to tackle common storage administration challenges. Godman addressed issues that arise during high Input/Output Operations (IOPS), particularly the difficulty in identifying the root causes of spikes in workload without adequate visibility. He explained how Qumulo’s analytics provide users with real-time insights into IOPS hotspots and capacity usage across large-scale file systems, using visually intuitive tools to track where data is being accessed and what drives operational load. This level of detailed visibility is vital for administrators managing vast amounts of data, as it simplifies diagnosing performance issues and optimizing system resources.
In closing, Godman shared examples of Qumulo’s scalability, including a system with over 9.2 billion files and discussions about how their infrastructure is suited to both large and small file environments. He acknowledged the challenges posed by scaling storage systems effectively, especially in handling numerous small files, which is often difficult for traditional systems. By illustrating the architectural strengths and the ongoing innovation at Qumulo, the presentation positioned the company’s storage solutions as robust options for enterprises with demanding data requirements.
Personnel: Peter Godman