What is a SANless Cluster?
Understanding Your Options: SAN vs SANless Clusters
SANless cluster software allows you to provide high availability protection for your important applications in the cloud. Simply add SIOS cluster software to a Windows Server Failover Clustering environment to enable it to work without the need for shared storage. In addition, you get incredible configuration flexibility for your cluster environment.
How does SANless Cluster Software Work?
Traditional cluster software that requires shared storage (SAN), In contrast, SIOS software lets you build clusters using local storage. More importantly, you can configure cluster nodes in any combination of cloud, on-premises data center, and/or remote disaster recovery site. SIOS cluster software synchronizes local storage using real-time (synchronous or asynchronous) block level replication. The synchronized storage appears to clustering software such as Windows Server Failover Clustering, as a traditional SAN-based storage.
More importantly, SIOS cluster software enables you to configure nodes within a data center, or across geographic areas and/or cloud availability zones. As a result, you get the flexibility to protect a single-site, multi-site, cloud, or a mixed environment. In the event of a failover, the standby server has immediate access to the most current data. No costly SAN hardware or complex administration. No costly vendor lock-in.
In the event of a failover, the standby server in a SANless cluster has immediate access to the most current data – just like a traditional cluster.
Learn more about SIOS Datakeeper cluster software solution.
Advantages of Using SANless Cluster Software
Cost-Effective High Performance and High Availability
SIOS cluster software lets you use high-performance flash/SSD storage for faster application response time without the cost of a SAN. SIOS software synchronizes high-performance local storage in your cluster nodes for cost-efficient high availability and disaster recovery.
SANless Clusters Save Money
SIOS cluster software saves you money in several ways. First, it lets you build a SANless cluster using your choice of cost-efficient industry-standard hardware and software on the primary and standby machines. You eliminate the cost and complexity of SAN storage.
Second, SIOS software also enables you to create a SANless cluster using native SQL Server clustering capabilities. For example, you can use SIOS DataKeeper with SQL Server Standard Edition AlwaysOn Failover Clustering. You get a high availability SQL Server cluster without expensive SQL Server Enterprise Edition.
Reduce Risk of Downtime
Traditional clusters require shared storage, which represents a potential single point of failure. SIOS cluster software ensures all storage is synchronized, eliminating this potential risk of downtime and data loss.
Unparalleled HA Flexibility for Cloud, Hybrid, and Virtual Environments
SIOS cluster software lets you protect Windows or Linux applications and data in any combination of physical, virtual, or cloud environments.
Cluster Software for Cost-Efficient Disaster Protection
You can use SIOS cluster software to replicate data to multiple geographically dispersed locations for protection from site-wide, local, and regional disasters. You can also locate a cluster node in the cloud for off-site disaster protection without the cost of building a remote DR data center. No costly SAN hardware or complex administration. No costly vendor lock-in.
Accurate Recovery from a Broad Variety of Failures
Failures can happen due to application, network, storage, or host issues. Simple replication and hypervisor availability solutions only protect the data and have little visibility into deeper causes of failures. These approaches miss problems with applications and resources sometimes leaving services down without enough knowledge to drive recovery.
SIOS cluster software has insight into application and system faults and takes only the most appropriate action that has least impact to operations. With insight into storage issues, SIOS software detects these and recovers promptly on a standby node. With insight into application operations, SIOS software chooses the recovery scenarios with minimal impact. In some cases, only services need to be restarted. In others, the application operations may need to move to the standby node. SIOS software makes the right decision to maintain optimal reliability and availability.
Higher Availability – Fast Recovery with SANless Cluster Software
Getting back up running fast and reliably are key high availability requirements for critical applications. VMware HA requires a full reboot and restart of the complete application and OS virtual machine environment on a failure. This is particularly problematic during ESX host failures because many virtual machines and applications will be starting all at the same time, contending heavily for resources – a “virtual recovery storm” so to speak. Depending on the application and environment, this can take a long time – 30 minutes or more in many cases. That’s not acceptable for critical applications.
SIOS software by design recovers quickly and accurately because the standby node is always ready and available with preconfigured and allocated resources. Since there is no need to reboot, only application services need to start. And since resources are already allocated, availability and performance are restored quickly. This can usually happen in minutes.
Availability During Upgrades and Maintenance
With SIOS software you get the benefit of application availability even during maintenance and software upgrades. Using rolling upgrades that update each node of the cluster sequentially, you can install new application software with minimal impact to an application’s availability. And if a software upgrade failure occurs, rolling back to the previous version of application or OS is just as simple.
Reliable Recovery From Failures
Unreliable available solutions simply make no sense. When there’s a failure, you expect your solution to recover. Reliable recovery from failures is integral to the design of SIOS cluster software. The standby node of a SIOS cluster is always operational and “ready” should the primary node fail. This means that appropriate resources needed to run the application during the critical recovery phases are assured. Memory, CPU, storage, and network resources are allocated and ready to run when needed – no failed recoveries! And simple recovery systems replicate the boot and application environment, not just the data. That means any errors, viruses, or corruption on the primary node boot environment are completely replicated to the standby; meaning the whole availability environment may be compromised or may never restart on a failure. Your only choice is to go to backups!
SIOS cluster software doesn’t replicate the application and OS boot environment. As a consequence any changes made to the operational environment or even viruses infecting the primary node aren’t replicated to the standby – that means the standby node can still run avoiding the downtime needed to manually recover from these threats.
Testing Without Disruption
SIOS software also lets you access a snapshot of the target side data for testing and validation without disruption to the production site. All of this occurs without impacting ongoing replication and HA/DR protection. When testing is complete, SIOS software drops the snapshot.