Cohesity Presents at Storage Field Day 8

Redefining Secondary Storage with Mohit Aron of Cohesity

Watch on YouTube
Watch on Vimeo

Mohit Aron, the founder and CEO of Cohesity, presented aims to redefine secondary storage. Founded in 2013, Cohesity focuses on addressing the challenges within secondary storage, specifically targeting issues like inefficiency and fragmentation. Secondary storage encompasses non-mission-critical data, including backups, file shares, archives, DevOps, analytics, and cloud data. Aron emphasized that secondary storage is often neglected and plagued by inefficiencies and the problem of fragmentation as organizations rely on multiple vendors to manage different aspects of this storage tier. This results in higher management complexities, costs, and the accumulation of redundant copies, which further complicates data utilization and insights extraction from secondary storage.

Cohesity aims to consolidate secondary storage into a single, infinitely scalable platform, eliminating the need to juggle between different systems for backups, archivals, DevOps, and analytics. By doing so, the platform aims to streamline the storage management, reduce costs associated with data fragmentation and redundancy, and make the process of managing and accessing data much more efficient. Aron’s vision for Cohesity is to offer a web-scale system that can handle petabytes of data without single points of bottleneck, capable of surviving component failures, and adaptable to handle hardware heterogeneity as organizations scale incrementally over time.

Aron’s philosophy is built around moving compute to the data rather than the other way around, thereby avoiding the inefficiencies of data migration. He emphasized the importance of running mixed workloads, including data protection, DevOps, and analytics on the same platform with inherent isolation and quality of service (QoS) mechanisms. Cohesity’s approach supports efficient data management and utilization through backup integration, random I/O support for DevOps, effective copy data management, and comprehensive analytics capabilities, ensuring that dark data can be illuminated and effectively used. The ultimate goal of Cohesity is to create a holistic secondary storage solution addressing all aspects of non-mission-critical data management efficiently and cohesively.

Personnel: Mohit Aron

Cohesity Data Platform Deep Dive

Watch on YouTube
Watch on Vimeo

In this presentation at Storage Field Day 8, Johnny Chen provides a detailed overview of the Cohesity Data Platform, focusing on its architecture and capabilities aimed at addressing issues related to secondary storage, which include fragmentation, silos, and challenges in copy data management. The talk is segmented into several parts, beginning with discussions on the scale-out distributed architecture of the Cohesity file system, which is designed to handle mixed workloads effectively and includes an adaptive self-healer for system maintenance and operations. The system’s hardware comprises a 2U chassis with four nodes, each equipped with dual CPUs, memory, SSDs, and hard drives, allowing substantial flexibility and scalability for various enterprise needs, including data protection, DevOps workflows, and analytics.

Chen delves into the specifics of the Cohesity OASIS architecture, highlighting elements such as the distributed lock manager, a strongly consistent NoSQL store, and the intelligent coordination required to ensure seamless integration and operation of multiple nodes within the cluster. Particularly noteworthy is their method of metadata management, including creating, managing, and ensuring the transactional integrity of file operations through a distributed NoSQL store and a two-phase commit process. The platform also employs innovative approaches to manage and optimize data storage through methods like global deduplication and adaptive tiering, which dynamically moves data between SSDs and HDDs based on access patterns, ensuring efficient utilization of storage resources and maintaining performance.

Additionally, Cohesity’s approach to mixed workloads and performance isolation is geared toward maintaining high efficiency and preventing heavy operations, such as large backup jobs, from affecting the performance of other tasks within the system. This is achieved through a user-defined quality of service (QoS) management system, which allocates resources proportionally based on predefined priorities. The self-healing capabilities of the system, running continuously at a low-priority backdrop, ensure that the system remains optimized and fault-tolerant, capable of handling tasks like garbage collection, disk rebalancing, and data replication seamlessly without disrupting primary operations. This continuous background process underscores the platform’s resilience and ability to operate efficiently even under diverse and demanding workload conditions.

Personnel: Johnny Chen

Cohesity User Interface Demonstration

Watch on YouTube
Watch on Vimeo

At Storage Field Day 8, Nick Howell from Cohesity provides a comprehensive demonstration of the Cohesity User Interface designed for modern data management platforms. The UI features a fully responsive design, displaying well on both mobile and desktop devices, and employs a tiled interface for usability. As users access the interface, they are greeted with essential metrics and alerts related to system health, job statuses, and storage utilization, aiming to deliver key insights immediately. The demo showcases not only the UI’s capability to provide a holistic overview of system performance but also its ability to handle infrastructure concerns effortlessly.

Nick further elaborates on the infrastructure management through the cohesive design that facilitates the addition and monitoring of cluster nodes. By double-clicking on specific nodes, users can view detailed information, including software versions and node activities. The process of adding a new node to an existing cluster is demonstrated, emphasizing the simplicity and automation inherent in the UI. This includes automatic discovery and software version matching, significantly reducing the time and effort required to expand or update the cluster. The ease of managing up to 32-node clusters is highlighted, underscoring the design’s scalability.

The session also delves into the handling of storage partitions and view boxes, which introduce a hierarchical structure aiding in the physical and logical segregation of data. Partitions represent a physical separation, suitable for departmental needs, while view boxes facilitate logical data separation and target deduplication policies. The demo continues with a focus on performance monitoring tools within the UI, offering real-time graphs that allow users to zoom in for detailed analysis. This robust integration proves beneficial for various administrative tasks, such as defining storage tiers and policies that accommodate diverse drive types and performance needs.

Personnel: Nick Howell

Cohesity Converged Data Protection

Watch on YouTube
Watch on Vimeo

The Cohesity Converged Data Protection presentation at Storage Field Day 8, delivered by Mark Thomas, focuses on the company’s advanced data protection architecture geared towards simplifying enterprise IT infrastructures. Thomas aims to explain the underlying technology of Cohesity’s solutions, particularly highlighting SnapTree—a data structure that supports efficient data management and protection workflows. The primary purpose of Cohesity’s architecture is to consolidate various operational data workloads onto a unified platform, which significantly streamlines data backup and recovery processes.

The conventional enterprise backup infrastructure, according to Thomas, is often burdened by a complex architecture that includes master servers, media servers, and multiple backup targets, which grow increasingly complicated as the enterprise scales its IT landscape. This complexity results in various inefficiencies such as multiple data silos, ineffective deduplication, and increased operational overhead. Cohesity’s solution replaces these intricate setups by offering a scalable platform that directly integrates with existing virtualized environments and applications, eliminating the need for separate media and master servers. Cohesity also features a distributed file system with SnapTree technology that allows for rapid and frequent data cloning without performance degradation.

The discussion also touched on how Cohesity supports additional workflows such as DevOps and instant VM recovery. Through its ability to create writable and modifiable clones instantly, use of integrated backup software, and a user-friendly interface, Cohesity provides robust solutions for both data protection and development environments. By indexing all backed-up data and enabling API integrations, Cohesity ensures ease of data management and swift recovery operations, making it a versatile tool for contemporary enterprise IT landscapes. The simplified architecture not only reduces the number of moving parts but also consolidates data management tasks onto a single platform, thus fostering a more efficient and less error-prone system.

Personnel: Mark Thomas

A More Efficient Approach to Analytics with Cohesity

Watch on YouTube
Watch on Vimeo

Abhijit Chakankar from Cohesity discusses how their platform addresses the challenges enterprises face with analytics, particularly concerning secondary data. He outlines the inefficiencies associated with copying data to separate analytics stacks and how Cohesity’s architecture offers a streamlined, integrated solution for in-place analytics.

Chakankar delves into the architecture of Cohesity’s solution, highlighting how production systems can write directly onto Cohesity, where snapshots are taken and potentially cloned for test and development environments. This setup consolidates data into a single, uniform view that simplifies and enhances analytic operations. With built-in analytics features, quality of service (QoS) capabilities, and ample compute resources, Cohesity supports powerful in-place analytics without the need for data redundancies or additional infrastructure.

He further explains key components of Cohesity’s analytics capabilities, including built-in analytics for utilization and capacity, a real-time indexing engine, and the Analytics Workbench (AWB), which uses a MapReduce-based framework. AWB allows for deep, customizable analysis by accepting user-defined code, enabling extensive use cases such as e-discovery, threat analysis, and data anonymization. Chakankar provides a detailed demonstration of creating an AWB app, showcasing its ability to scan for specific patterns, such as social security numbers, and underscores the platform’s flexibility and efficiency in managing and analyzing data directly within the Cohesity ecosystem.

Personnel: Abhijit Chakankar

Cohesity SFD8 Closing Remarks with Mohit Aron

Watch on YouTube
Watch on Vimeo

In the closing remarks of Cohesity’s presentation at Storage Field Day 8, Mohit Aron discussed the benefits of Cohesity’s platform, emphasizing its capability to simplify complex legacy storage architectures. By consolidating multiple secondary storage workflows into a unified web-scale storage platform, Cohesity aims to integrate key functions like data protection, DevOps, and analytics. Aron justified the company’s focus on secondary storage over primary storage, explaining that the more significant challenges and opportunities for innovation lie in that space.

Personnel: Mohit Aron