Both are open-source programs, which are designed to run on commercial off-the-shelf (COTS) hardware. While each Red Hat storage program can work on enterprise data loads on data-farms they're most likely to be deployed on clouds.
Ceph is a distributed object store and file system. It's designed to provide excellent big data performance, reliability, and scalability. What RHCS 1.3 brings to the table are the following new features.
- Improved automatic re-balancing logic that prioritizes degraded, rather than misplaced, objects
- Re-balancing operations that can be temporarily disabled so those operations don't impede performance
- Scrubbing that can be scheduled to avoid disruption at peak times
- Object buckets that can be sharded to avoid hot-spots
Red Hat also claims that "RHCS has been tweaked and tuned to improve speed and increase I/O consistency." This includes flash storage optimization; read-ahead caching, which accelerates OpenStack virtual machine booting; allocation hinting, which reduces XFS fragmentation to minimize performance degradation over time; and cache hinting, which improves performance.
RHCS management has also been improved. Calamari, Ceph's management platform, now supports multiple users and clusters. Want to kick RHCS's tires? You can take RHCS for a free test drive on Amazon Web Services (AWS).
RHGS is a scale-out file store with a straightforward architecture suitable for public, private and hybrid cloud environments. The program combines GlusterFS 3.7 and Red Hat Enterprise Linux. It supports network file system (NFS), Server Message Block (SMB), and Hadoop Distributed File System (HDFS) file interfaces.
The new top-of-mind features in RHGS are:
- Support for erasure-coded dispersed storage volumes. With this, sysadmins can reconstruct corrupted or lost data by using information about the data stored elsewhere in the system. This can reduce the need for RAID and replication and total cost of ownership by up to 75 percent.
- Tiering provides for automated movement of data between "hot" and "cold" tiers in a volume-based on-access frequency. This gives operators fine-grain control to deploy storage clusters that are both cost-effective and high performance for a wide variety of workloads.
- Bit-rot detection scans data periodically to detect data corruption resulting from silent failures in underlying storage media, enhancing end-to-end data integrity.
- Enhanced security via support for SELinux in enforcing mode and SSL-based network encryption to increase security across the deployment. RHGS also now supports active NFSv4, based on the NFS-Ganesha project. With this you can secure data access through clustered NFSv4 endpoints; and SMB 3 capabilities to allow enable efficient file transfer and secure access in Linux and Microsoft Windows environments.
RHGS will be available this summer. You can get an early look at RGHS for free on AWS.