Exceptional software-defined block storage for unmatched application speed on your public or private cloud
StorPool is an ultra-high performance storage software, installed on a cluster of standard servers. It then creates a pool of shared storage from these servers (it can be thought of as being a virtual SAN). Because StorPool is flexible and efficient, it can be deployed on either separate storage servers, or on the compute servers themselves, besides the applications (in a converged / hyper converged set up). In each case StorPool delivers extremely fast, scalable and cost efficient high speed block storage. Featuring advanced fully distributed architecture, it is potentially the fastest and most efficient block storage software, available on the market today.
StorPool combines the storage IOPS performance of all of the drives in the cluster and optimizes drive access patterns, to provide both low latency and extremely efficient handling of storage traffic bursts. All storage operations are distributed equally between all of the servers through data striping and sharing, which allows StorPool to boost performance and remove I/O bottlenecks.
StorPool delivers outstanding block storage performance, low latency and seamless scalability. Below you can see performance test, where we measured 13.8M IOPS on a 12-node cluster built on standard off-the-shelf Intel servers. We configured and tested a 12-server KVM hyper-converged system on hardware graciously provided by the Intel®️ Datacenter Builders program. With these results, StorPool holds the new world record for the highest data storage performance reached by an HCI (Hyper-Converged Infrastructure) solution.
StorPool holds the new world record – 13.8 mln IOPS on an NVMe-storage system
Storage Performance Test Results Summary
StorPool is a high speed distributed data storage software running on standard server hardware. It uses minimal system resources, in order to achieve outstanding performance. The performance highlights below are measured on a a 12-node cluster built on standard servers, in a hyperconverged setup.
|All-NVMe storage system|
|Random reads||13,800,000 IOPS|
|Random writes||2,900,000 IOPS|
|Sequential reads||64,600 MB/s|
|Sequential writes||20,800 MB/s|
|Read latency||0.137 ms|
|Write latency||0.095 ms|
Summary: StorPool provides exceptional performance, whilst being unique at delivering end-to-end data integrity and shared storage capabilities. In addition, StorPool also maintains high performance (in terms of storage IOPS and bandwidth) and unrivaled efficiency (in terms of CPU and RAM usage), even whilst it is serving the demanding storage requests of many different and competing workloads.
7 Million IOPS and 0.15 ms Latency for an NVMe-powered VDI Cloud
Hyper-converged setup to achieve the best possible rack density and unit economics
These results delivered from a super-fast, hosted VDI solution, with high utilization. One of the hard requirements of the NVMe-powered VDI Cloud project was very high single thread performance which necessitates the use of CPUs with a relatively small number of high frequency CPU cores. There were 39 servers, each with 12 cores. StorPool utilized just 2 CPU cores and 8 GB RAM per node and delivered a blazing fast storage system, peaking at 6,800,000 IOPS and delivering latency of less than 0.15ms (!) under typical load. This allowed the customer to achieve unmatched performance and efficiency.
Leading Public Cloud Performance Measurement Report
Amazon AWS, Digital Ocean, DreamHost and OVH vs. a number of StorPool-based public cloud
In these public cloud performance tests, StorPool’s aim was to assess the block storage offerings of a number of public clouds, including Amazon AWS, Digital Ocean, DreamHost and OVH vs. a number of StorPool-based public cloud offerings. We’ve selected VM instance types and everything else in the configurations to be identical because we are comparing storage systems/offerings, not other aspects.
The tests were performed between November 2018 and February 2019 by StorPool. All systems under test are in production clusters.