Ceph Object-Based Storage Cluster Next Generation Object, Block, S3 and OpenStack Storage Platform for Dynamic Enterprise Environments
Designed for Reliable Large Scale Storage Deployments:
With the advancement of large scale cloud computing platforms like OpenStack, the business of data storage has been forever changed. Often called “the serverization of storage”; the replacement of purpose-built/proprietary data silos with economical server hardware, has proven itself to be both more reliable and more resilient. Supermicro has embraced this change, offering the widest selection of server hardware in the industry.
Ready to Deploy Configurations
• Role specific server models – base server configurations offer performance, capacity and density to fit popular storage applications. Components are easily customized to meet specific requirements
• Optimized network configurations – rack level integration offers streamlined deployment of storage and infrastructure with consistency not attainable using improvised expansion methods
• Storage/Media Ratios to fit user applications – deployment of SSD and rotating media allows the solution to meet demanding performance and density targets
Object-based Storage
Organizations prefer object-based storage when deploying large-scale storage systems because it stores data more efficiently. Object-based storage systems separate the object namespace from the underlying storage hardware—This simplifies data migration.
The Inktank Ceph Enterprise Difference
Ceph’s CRUSH algorithm liberates storage clusters from the scalability and performance limitations imposed by centralized data table mapping. It replicates and re-balances data within the cluster dynamically— eliminating this tedious task for administrators while delivering high performance and infinite scalability.
The SUPERMICRO / CEPH Solution at-a-glance:
- Ceph Optimized Server Configurations
- Object and Block Level Storage with S3 and OpenStack Integration
- Hybrid Disk configurations deliver low-latency performance
- 10G Frontend Networking
- Bonded 10G Backend Networking (20Gb/s OSD-toOSD)
- Out-of-Band Server Management Software (OOB)
- Full Rack Integration with Onsite Service Available!
Typical Deployment Model:
The following diagram highlights optimized data path for clients to read and write shared objects and blocks directly from the storage nodes.
Out-of-Band Server Management:
Our solutions are designed for easy automation with existing management infrastructure. In data centers, Supermicro Server Management
Utilities provides you all the necessary functions
- Remotely managing the health of hardware and operating system services
- Managing power consumption of nodes in cluster
- Managing BIOS provisioning through BMC/IPMI Execute commands on multiple target systems in parallel
Supermicro Server Management Details are here: http://www.supermicro.com/products/nfo/SMS_SUM.cfm
SMC Scale-Out Storage C Series Key Features Based on Ceph:
Features | Benefit |
Unlimited storage | Support Object, Block and File based storage |
Massively Scalable (scale out architecture) | Global namespace, Highly scalable read/write access, Backup and archive large amounts of data with linear performance |
Software Defined, Intelligent Devices | Self Healing and self managing, No single point of failure |
Intelligent Placement With CRUSH | Automatically distribute data and balance load across the cluster |
Non Centralized Metadata Server | Higher performance, no bottlenecks |
Built-in Replication | 3x data redundancy, Rack awareness, Data security |
Intelligent Objects | Easy to run storage-aware applications and scalable programs |
Multi API Support | Utilize tools that were designed for the popular S3 and Swift API |
Full and Incremental Snapshots | Data protection and recovery for VM data |
Open Source Software+ World Class Servers | No lock-in, No License fees, Highly reliable, Best price/performance, best price/capacity |
Multi Cloud Platform Support | Supports Openstack, Cloudstack, KVM, XEN, and RESTful API |
Fully Integrated | Fully integrated, Fully optimized and completely tested turnkey cluster |
Easy To Run | No silos, Only one system to manage, Low Op-Ex |