StarWind Inc.: Supporting Customers' Sustainability and Business Continuity
StarWind HCA delivers a cost-effective solution for its clients as it merges compute, storage, and network into a single abstraction. Having only one on-site node and one in the cloud, it allows clients to achieve constant uptime and ensure flexible scalability routes. StarWind created HCA as a turnkey solution and further enhanced it with other StarWind soft products to create a fault-tolerant, virtual storage-based, software-defined solution—with none of the license-per-feature guessing games. “Customer effort in pre-configuration, deployment, and migration associated with integrating our HCA, as well as in troubleshooting or preventing an issue from happening in a StarWind-hyper-converged environment, is eliminated. All of that is done by our support engineers.”says Anton Kolomyeytsev, the CEO, CTO and Chief Architect of StarWind.
There are two possible failover strategies that clients can choose during the creation of a StarWind Highly Available (HA) device: HeartBeat and Node Majority. With the HeartBeat strategy, StarWind does not require additional hardware/software witnesses to create a two-node cluster, which is done through the creation of multiple HeartBeat connections between two or three nodes. StarWind requires at least two physical network connections between nodes, where one serves as HeartBeat and one as Synchronization plus HeartBeat.
Such baseline steps are key in achieving HA. If HeartBeat fails, both nodes can still communicate over the remaining Synchronization Channel (which also serves the HeartBeat), and both nodes remain active. This also applies ifeverything is healthysince both hosts are “masters.”However, each node has its own priority in StarWind devices, which is assigned during their creation.
Customer effort in pre-configuration, deployment, and migration associated with integrating our HCA, as well as in troubleshooting or preventing an issue from happening in a StarWind-hyper-converged environment, is eliminated. All of that is done by our support engineers
If Synchronization networking failure occurs on one of the nodes, the firstpriority node will continue working while the second one will be disabled for receiving any incoming requests in order to avoid split-brain.
What differentiates StarWind is how it develops and prices its solutions in a way that provides SMB, ROBO, and Edge businesses with enterprise-level features that help them increase their competitive advantage on the market. The company’s products can be integrated into any infrastructure, deployed in any environment, be relevant for any business area, and suit any desires for performance elevation, hardware footprint reduction, and IT resources cost management.
Our teambelieves that we live in a world that keeps spinning ever faster, forcing businesses to provide high-quality services at a rapid speed. To this end, StarWind has developed an all-flash version of its HCA and NVMe-over-Fabrics to help NVMe flash communicate without any old-fashioned protocol discrepancies. For the future, the company will keep pushing for market disruption and flexibility when it comes to businesses, institutions, and organizations requiring powerful IT environments to be cost-effective.