The future is hybrid.
The IT Press Tour #37 had the opportunity to meet with Boyan Ivanov, CEO and Co-founder, and Boyan Krosnov, Chief of Product and Co-founder at StorPool.
StorPool is a software company offering the latest-generation block-storage software that turns standard servers and networks into high-end storage systems and shared storage that's faster than local SSD. They focus on Tier 0/Tier 1 (primary, NVMe), at scale 100 TB+.
They own the storage stack with on-disk format, protocols, quorum, and client. They've been in production for more than 8 years with 20 major releases. Their solution is robust, fast, and scalable. The latest generation is cloud-native, API-driven, and feature-rich.
StorPool is building for a hybrid multi-cloud future given that IDC projects the cloud infrastructure market to bd 44.1% public cloud, 19.5% private cloud, and 36.4% on-prem in 2024.
Target customers are new age/modern IT companies with cloud-native automation, APIs, CMPs, DevOps, and site-reliability engineering, public cloud service providers (IaaS, PaaS, SaaS, MSP, cloud, and hosting) and private cloud enterprises (SME and larger enterprises).
Typical use cases are flash-shared storage for virtualized, container, and bare-metal environments. Databases, VM disks, VDI large applications or SaaS, and persistent Kubernetes volumes. Handle data at scale from 1000TB+ to 10's of PB of data across multiple datacenters. Able to convert legacy IT stacks to SDDC designs and consolidate multiple IT systems.
Architecture
SDS runs on Linux. There are no special hardware requirements -- just NICs and NVMes. They also support SAS and SATA drives and controllers. The system scales out with an active multi-controller, pooling of capacity, and polling of performance. Copy-on-write on-disk format. End-to-end data integrity and three-way replication. Low latency, high throughput, and highly efficient storage protocol and implementation with Kernel bypass. API controls. Common deployments with 10,000+ volumes and snapshots. Integrations with OpenStack, CloudStack, OpenNebula, Kubernetes, et al. Snapshots and multi-site capabilities. Monitoring and fine-grained metrics and analytics.
Performance
Storage performance drives application performance. Applications running on seemingly the same infrastructure can have orders of magnitude difference in performance. Applications running on StorPool-powered clouds perform up to 2.5X better than the next best public cloud and more than 10X better on other leading clouds.
Features and Capabilities
Performance -- low latency and extreme efficiency. NVMe-class performance where needed -- transaction databases. NVMe-class latency -- sub-100 microsecond under moderate load. Faster than local SSD by aggregating the performance of all drives ina cluster with very low latency overhead. Good performance fit for OLTP and OLAP (transactional databases and business intelligence). Good fit for hyper-converged architectures due to extreme CPU efficiency -- 250k IOPs per CPU core, 1M IOPS+ even from a small cluster.
Scale-out architecture. Datacenter scale, multi-PB deployments. Support for simple flat ethernet networks and routed datacenter-scale networks. The pooling of capacity and performance, High availability by design. In-service hardware and software upgrades.
Reliable data storage platform. End-to-end data integrity and protection for high data and transaction-intensive environments. Data redundancy across storage nodes. Fine-grained copy-on-write with 4k granularity, efficient snapshots, clones, thin provisioning, and changed block tracking. Fast mesh recovery.
Data management. Online data mobility between performance tiers, availability zones, and between locations. LUN per virtual disk. Snapshot-based backups. Good fit for disaster recovery use cases and business continuity.
API Control, DevOps. StorPool is well suited for "new IT" workflows characterized by scalability and automation. REST-full API. Integration with common container/VM orchestration systems -- Kubernetes CSI, OpenStack Cinder, and more. Deploy storage nodes and hosts with Ansible roles and playbooks. Easy to build custom automation and integrations. Monitoring is hosted by StorPool or on-prem. Fine-grained metrics collection and analytics provide deep visibility into the storage behavior of each container or VM hosted by StorPool or on-prem.
Commenti