ScaleFlux’s flagship CSD 2000 solution, when combined with Supermicro’s FatTwin, a high-density 4U multi-node system, would provide a fully integrated high-performance, cost-effective server storage platform that is uniquely suited for Redis on Flash (RoF) deployments.
RoF would expand a Redis cluster’s capacity by using high-performance Flash memory as a storage tier beneath main memory (RAM), and its performance is heavily impacted by data access patterns. The greater the Flash-to-RAM ratio, the more dominant the Flash storage layer becomes for database performance, and in many circumstances, the severe demand can wear out regular solid-state drives (SSDs). According to ScaleFlux and Supermicro, users must either buy and install 2-3x the capacity of SSDs that their database truly requires, or replace worn-out SSDs after as little as 18 months to address this issue.
The CSD 2000 in conjunction with Supermicro’s FatTwin offering would provide a convincing answer to this issue. CSD 2000 matches the endurance and performance requirements of RoF workloads, while FatTwin’s half-width architecture provides high-density computation. With just two CSD 2000 devices per node, benchmarking data would reveal consistent, low-latency performance with RAM hit ratios as low as 60%, according to ScaleFlux and Supermicro.
The FatTwin platform’s half-width design would reduce physical space needs, while the CSD 2000’s transparent datapath compression would reduce storage space requirements – all while solving the main workload issues of a RoF deployment: write endurance and read performance.
“The small number of SSDs required by RoF deployments leads to some critical design considerations, as users require SSDs to individually offer high endurance mixed with excellent workload performance,” said Tong Zhang, co-founder and chief scientist of ScaleFlux. “Our collaboration with Supermicro addresses these needs and appeals to a variety of use cases. We look forward to astounding customers with the performance and cost efficiency.”
Data Center Optimization
“We’re excited to collaborate with ScaleFlux and other ecosystem players to enable solutions for improving data center density and application performance,” said Vik Malyala, senior vice president, Supermicro. “As a global leader in high performance, high-efficiency server technology, and innovation, our goal is to work with industry leaders offering leading computational storage solutions.”
To sum up, key benefits of using CSD 2000 with FatTwin for RoF deployments would include:
- Data center optimization – FatTwin is meant to enable two nodes per rack unit, twice the compute density compared to full-width systems, while RoF only requires a modest number of SSDs to meet the required capacity
- Aligning service intervals and reducing total cost of ownership (TCO) – In the write-intensive RoF scenario, the CSD 2000’s compression feature allows SSDs to endure three years or more
“This high performance, innovative combination of ScaleFlux computational storage and high-density next generation systems based on Supermicro FatTwin server platforms clearly delivers enhanced scalability and performance for Redis,” said Joel Wineland, Chief Technology Officer (CTO) at Racklive, data center division of ASA Computers, who provided resources for this testing. “We expect to see many customers in a variety of verticals including Ad tech benefit from such implementation. Looking forward to incorporating these elements into our portfolio of rack-level roll in solutions to maximize value and enhance performance.”
Preconfigured Redis on Flash (RoF) nodes are now available at Racklive. Read the benchmarking white paper here (PDF opens in new window) to learn more.
