Revolutionize Aerospike data processing efficiency
Deploy next-generation SSDs with in-drive storage processing
Key customer benefits
Increase revenue
Handle a higher transaction load with lower latency to deliver better service.
Improve efficiency
Do more work per server and store more data per drive.
Simplify your cluster
Use fewer servers and fewer drives to satisfy your workload needs.
“By using the all-Flash deployment of Aerospike Enterprise Edition with ScaleFlux Computational Storage Drive, we can achieve a 110-224% improvement in our deployment’s performance while providing storage with high IOPS and continuous low latency high-quality performance comparable to memory.“
– Tongdun Technology
Challenge: Petabyte-scale struggles
Managing the trade-offs between consistency and performance
Customers choose Aerospike when they need low-latency access to relatively large datasets within a reasonable server footprint, making the predictable management of SSD latency a critical function.
But optimizing transaction processing per node amplifies the challenges as data grows to a petabyte scale. To maintain the best latency profile, Aerospike performs a process of defragmentation, requiring double the storage capacity on top of the built-in overprovisioning of the SSD. The process helps Aerospike scale, but it significantly reduces the usable capacity of SSDs, driving a more significant infrastructure footprint to support more storage and more nodes
Solution: Breakthrough scaling limits to unleash Aerospike’s full potential
Deploy a better NVMe SSD with on-board transparent compression
The CSD 3000 from ScaleFlux is a better SSD built with embedded Computational Storage Engines (CSEs) that offloads storage processing from the CPU. Aggressive latency targets can be achieved during high transaction loads using Transparent Compression. This unique Scaleflux feature manages and optimizes data directly on the drive with the added benefit of doubling the usable capacity and doubling the drive’s lifespan without the host being aware that compression is happening, which is why no special software or app configuration is necessary.
Results: Hit Aerospike SLA targets without adding complexity or cost
Offload storage processing to drive more transactions at lower latency while minimizing node count
Check it out for yourself, Aerospike published performance profiles for popular storage devices. Dominated by ScaleFlux, the Aerospike ACT 6.2 mixed workload test highlights the CSD 3000 as the drive with the best-performing latency profile compared to other industry-leading NVMe options. As workload intensity increases, competing drives fail the test, leaving the CSD 3000 as the sole device capable of fulfilling the most demanding workloads.
Want to make Aerospike sing?
Read about it on our solution brief.
