It is one thing to get OpenStack operational, but quite another to operate it scalably, efficiently and in a cost-optimized manner. FishOS’s key values lie in these crucial aspects of cloud operation.
Fundamentally, when managing data center resources, the challenge boils down to optimally resolving the live demand vs. resource availability problem, constantly knowing resource states, and swiftly executing preventive and corrective measures.
FishOS tracks and makes decisions based on a myriad of detailed resource utilization data relating to the entire cloud infrastructure, ensuring that your OpenStack data center is highly functional and highly available at optimal cost:
- Optimally and rapidly places VMs in a large, dynamically changing data center. VMs are instantly placed on the right host – first time and every time. No more waiting for ages for a VM to start;
- Automatically rebalances VMs’ workload across the facility while meeting workload requirements, ensuring optimal VM performance and maximizing data center capability at constant TCO;
- Right-sizes the number of servers to meet workload requirements, dramatically reducing energy OpEx and optimizing server and facility CapEx;
- Data center structure awareness allows consideration of a host server’s physical location when placing and rebalancing workload;
- Highly available. Every component of FishDirector has been architected to be operable in highly available mode, providing redundancy and ensuring service uptime;
- Health diagnosis framework tracks host servers’ health and can auto-migrate the workload to another host;
- Highly operable. FishDirector scalably handles different types of operational automation that might otherwise require operator intervention. This design allows a single operator to maintain a data center of thousands of host servers;
- Smooth upgrade. With FishOS, leave behind the worry of upgrading OpenStack from one release to the next. We provide an assured, smooth upgrade path.