Shulou(Shulou.com)05/31 Report--

This article mainly introduces what CRUSH is in Ceph. It is very detailed and has certain reference value. Friends who are interested must finish reading it.

The CRUSH algorithm determines how to store and retrieve data by calculating the location of data storage. CRUSH authorizes Ceph clients to connect directly to OSD rather than through a central server or broker. With the use of data storage and retrieval algorithms, Ceph avoids the physical limitations of single point of failure, performance bottleneck, and scalability.

CRUSH needs a map of the cluster and uses CRUSH to pseudorandom store and retrieve the data in the OSD of the entire cluster. For more information on the discussion of CRUSH, see CRUSH-controllable, scalable, distributed location of multiple copies of data.

The CRUSH diagram contains a list of OSD, a list of "buckets" that aggregate devices into physical locations, and a list of rules that instruct CRUSH how to replicate data in a storage pool. Due to the expression of the underlying physical organization installed, CRUSH can model and therefore locate the source of potential related failed devices, typical sources are physical distance, shared power, and shared network, and encode this information into the cluster operation diagram. CRUSH placement strategy can separate the copies of objects into different failure domains, but still maintain the desired distribution. For example, to locate the possibility of simultaneous failure, you may want to ensure that the devices to which the data is copied are located in different racks, different pallets, different power supplies, different controllers, or even different physical locations.

When you write the configuration file and deploy Ceph with ceph-deploy, it generates a default CRUSH diagram, which is good for your sandboxie environment. However, when deploying a large data cluster, you should design your own CRUSH diagram because it helps you manage Ceph clusters, improve performance, and ensure data security.

For example, if an OSD is down, the CRUSH graph can help you locate the physical data center, room, row, and rack of the host where the OSD is located in this event, based on which you can request online support or replace hardware.

Similarly, CRUSH can help you identify problems more quickly. For example, if all the OSD on a rack is down at the same time, the problem may be the rack switch or power supply, not the OSD itself.

Customized CRUSH diagrams can also help you find out the physical location of the host where the redundant copy is located when the configuration group is degraded.

The above is all the content of this article "what is CRUSH in Ceph?" Thank you for reading! Hope to share the content to help you, more related knowledge, welcome to follow the industry information channel!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.