ROBIN Scales Up to 500 Nodes of Cloudera cluster in a Record 54 minutes
SEATTLE, WA (KUBECON) — December 11, 2018 — ROBIN, the provider of industry’s leading Hyper-converged Kubernetes platform, today announced results from a new benchmark study, underscoring the enterprise-class capabilities of Robin Hyper-converged Kubernetes Platform when delivering Big Data, Database and AI/ML as a Service.
Results from the study validate ROBIN’s ability to deliver the highest level of performance and offer quick and seamless scalability while containerizing data-intensive applications. Key findings include:
- ROBIN demonstrates near bare-metal storage and workload performance, based on Oracle Slob, Cassandra YCSB and Synthetic (Vdbench) benchmarks that stress storage performance across different IO patterns.
- ROBIN scales seamlessly to up to 500 nodes of Cloudera in a record 54 minutes.
ROBIN Hyper-Converged Kubernetes Platform is engineered to address the unique requirements of containerizing Stateful workloads such as Databases, Big Data and AI/ML by extending Kubernetes with built-in storage, networking, and application management to deliver a production-ready solution. Be it relational databases such as Oracle RAC, NoSQL databases such as MongoDB, Big Data Platforms such as Cloudera, or Search/Index solutions such as the Elastic Stack, ROBIN automates the deployment and management so that you can deliver an “as-a-Service” experience with 1-Click simplicity to the users.
Benchmark Study Methodology
The benchmark study was executed using standard benchmark tools such as vdbench for Storage performance comparison and Oracle Slob and Cassandra YCSB for workload-specific benchmarking. In addition, scaling tests were done using AWS cloud resources.
“When we looked at the available storage solutions, such as Ceph, Gluster, BTRFS etc., we quickly realized that they are not designed to run mission-critical workloads such as Big Data and Databases,” said Partha Seetala, CTO of ROBIN. “These workloads require high performance that is stable when multiple-workloads share the same physical storage and compute resources. Also, none of the existing storage solutions, open source or commercial, are application-aware. We built the ROBIN storage-stack to be application-aware, which means it understands application typologies, their performance needs and the isolation constraints for storage volumes across different partitions of the application. Using this, ROBIN enables industry-first features such as application-aware QoS, application-consistent snapshots and strong data-locality, affinity and anti-affinity policies. We are glad to see our customers benefit from the architectural strengths of the ROBIN solution to run production workloads with billions of transactions per day and petabytes of storage under management.”
Robin Systems is the first hyper-converged Kubernetes solution for big data, databases and AI/ML. Robin is the only solution that embeds application workflows natively into the storage, network, and cloud infrastructure stack. As a result, every application benefits from a simple managed the service-like experience, along with automated deployment and lifecycle management. Traditional hypervisor-based virtualization typically suffers from poor performance and complex and time-consuming application lifecycle management. With Robin, enterprises can bring their applications online much faster, at lower cost, and in any location (on-premise, or in the cloud). With a team that includes industry veterans from leading enterprise technology companies such as NetApp, Oracle, and Veritas, Robin Systems seeks to disrupt the $20 billion-plus virtualization market with its hyper-converged Kubernetes platform. The San Jose California-based company is backed by leading investors such as Clear Ventures, DN Capital, USAA, Hasso Plattner Ventures, and CloudScale Capital Partners.
Robin Systems, the Robin Systems logo and ROBIN Hyper-Converged Kubernetes Platform for Enterprise Applications and Application-to-Spindle Quality of Service Guarantee are trademarks or registered trademarks of Robin Systems, Inc., and are protected by trademark laws of the United States and other jurisdictions. All other product and company names are trademarks or registered trademarks of their respective companies.
Media Contact for Robin Systems