An Introduction to Ceph Storage: A Free and Open Source Distributed Storage Solution

In 2004, the first lines of code that wound up becoming the starting point for Ceph were written by Sage Weil as he attended a summer internship at the Livermore National Laboratory (LLNL). Initially, Weil created Ceph as a part of his doctoral dissertation. He began working on what is known today as Ceph’s MDS, a scalable file system metadata management system.

Ceph was assisted by the University of California, the University of California Santa Cruz (UCSC), and Doctor Scott A. Brandt of Computer Science at the University of California, Santa Cruz. Additionally, Ceph was sponsored by Lawrence Livermore National Laboratory (LLNL), Sandia National Laboratories (SNL), Los Alamos National Laboratory (LANL), and the Advanced Simulation and Computing Program (ASC).

In 2005, Weil and Carlos Maltzahn worked together on a summer project to create a fully functional file system prototype—that prototype eventually was called Ceph. It wouldn’t be until 2006 that Ceph would make its debut, with Weil giving two presentations about Ceph at USENIX 2006 and SC’06.

Ceph Offers a Single, Open-Source, Consolidated Storage Platform

Ceph is a completely free, open-source storage platform that implements object-based storage on a single distributed computer cluster. Ceph also provides interfaces for file-, block-, and object-based storage. It features several advanced features such as Ceph File System (CephFS), RADOS Gateway (RGW), and RADOS Block Device (RBD). Client applications are afforded direct access to the Reliable Autonomic Distributed Object Store (RADOS) object-based storage system via Ceph’s software libraries as well.

With Ceph powering your industry-standard hardware, you’re not merely limited to Ceph’s native binding or RESTful APIs; Ceph can also be mounted as a thinly provisional block device—this means Ceph will automatically stripe and replicate data across clusters. Ceph’s RADOS Block Device (RBD) blends well with Kernel Virtual Machines (KVMs), delivering practically limitless storage to KVMs operating on your Ceph clients.

According to ceph.io, “Ceph RBD interfaces with the same Ceph object storage system that provides the librados interface and the Ceph FS file system, and it stores block device images as objects. Since RBD is built on top of librados, RBD inherits librados capabilities, including read-only snapshots and revert to snapshot. By striping images across the cluster, Ceph improves read access performance for large block device images.”

Furthermore, unlike most object-based storage systems on the market, Ceph offers a file system interface with POSIX. Object-based storage systems are an important innovation, however, they complement conventional file-based systems rather than replace them. As storage requirements advance for legacy applications, “organizations can configure their legacy applications to use the Ceph file system” as well. This means one can run storage clusters for file-, block-, and object-based data storage.

Ceph Offers High-Performance Data Storage That Scales

Emerging applications such as network function virtualization (NFV), cloud-based DVR, and video delivery networks can greatly benefit from Ceph’s high-performance object-based storage. You can deploy RAID.Inc’s Ceph storage clusters and NVMe SSD in high-performance tiers that are optimized to support the IOPS requirements of high-performance workloads, as well as supporting bandwidth and latency.

Other Perks Related to Using Ceph Storage

When you deploy Ceph open-source distributed storage, you will benefit in many ways, including being able to rely on the expertise of Ceph’s creators and main open-source sponsors by way of best-in-class consulting. Ceph offers an award-winning global support team that offers support in numerous languages—expert training is included! You will gain access to over 400 hours of hands-on labs, 1,000 instructor-led recordings, and much more. Additionally, one can complete audio classes that teach you the link between storage cloud computing, virtualization, and server administration.

On top of all that, you will be able to significantly reduce the cost of your data storage by constructing a storage cluster using only conventional, inexpensive hard disks and servers. RAID.Inc Ceph Storage is compatible with most hardware, allowing you to choose the servers you feel meet your needs the best, based on their performance specifications, not the other way around.

For more information on Ceph Storage and whether it is right for you, please contact one of our experts here at RAID.Inc today to learn more.