Load balancing NetApp StorageGRID
StorageGRID software from NetApp offers a secure, durable solution for storing and managing unstructured data at scale. Hundreds of petabytes can be stored in one cohesive system, which is flexible, cloud-ready and simple to configure – thereby optimizing workflows and reducing costs.
Loadbalancer.org complements intelligently designed storage systems, making sure that data isn’t just protected, but accessible at all times. Thanks to our object storage expertise, we can help businesses to meet growing data demands through scalability and interoperability.
Offering performance without limitations, the best-value hardware load balancer on the market supports any environment. Licensed for unlimited throughput, bandwidth and features, upgrading is seamless if your requirements change down the line.
NetApp StorageGRID architecture supports High Availability (HA) clustering by putting a load balancer in front of it.
Load balancers monitor and perform health checks on a node to ensure traffic is routed correctly to healthy nodes. Without the use of a load balancer, an offline or failed node would still receive traffic, causing failures.
For load balancing NetApp StorageGRID we recommend Layer 4 Direct Routing (DR) mode, sometimes referred to as Direct Server Return (DSR). However, Layer 4 NAT, Layer 4 SNAT and Layer 7 SNAT can also be used. The best method for your deployment depends on a variety of factors. Layer 4 DR mode is fastest, but requires the ARP problem to be solved, and the application running on the real servers to respond both to its own IP address and the VIP. Layer 4 NAT mode requires the default gateway on the real servers to be the load balancer. Layer 4 SNAT mode doesn’t require real server changes, but unlike the other Layer 4 methods, is non-transparent. Layer 7 SNAT mode is also non-transparent and requires no real server changes – but it doesn’t offer the raw throughput of Layer 4 methods.
With Layer 4 Direct Routing, a node does not rely on sending its reply traffic via the load balancer. Instead, the back-end nodes have the VIP or load balancer address attached to their local interface, and so can send a TCP reply directly back to the client with the source address the client is expecting.