Show Table of Contents
2.3. Block Allocation Issues
This section provides a summary of issues related to block allocation in GFS2 file systems. Even though applications that only write data typically do not care how or where a block is allocated, a little knowledge about how block allocation works can help you optimize performance.
2.3.1. Leave Free Space in the File System
When a GFS2 file system is nearly full, the block allocator starts to have a difficult time finding space for new blocks to be allocated. As a result, blocks given out by the allocator tend to be squeezed into the end of a resource group or in tiny slices where file fragmentation is much more likely. This file fragmentation can cause performance problems. In addition, when a GFS2 file system is nearly full, the GFS2 block allocator spends more time searching through multiple resource groups, and that adds lock contention that would not necessarily be there on a file system that has ample free space. This also can cause performance problems.
For these reasons, it is recommended that you not run a file system that is more than 85 percent full, although this figure may vary depending on workload.
2.3.2. Have Each Node Allocate Its Own Files, If Possible
Due to the way the distributed lock manager (DLM) works, there will be more lock contention if all files are allocated by one node and other nodes need to add blocks to those files.
In GFS (version 1), all locks were managed by a central lock manager whose job was to control locking throughout the cluster. This grand unified lock manager (GULM) was problematic because it was a single point of failure. GFS2’s replacement locking scheme, DLM, spreads the locks throughout the cluster. If any node in the cluster goes down, its locks are recovered by the other nodes.
With DLM, the first node to lock a resource (like a file) becomes the “lock master” for that lock. Other nodes may lock that resource, but they have to ask permission from the lock master first. Each node knows which locks for which it is the lock master, and each node knows which node it has lent a lock to. Locking a lock on the master node is much faster than locking one on another node that has to stop and ask permission from the lock’s master.
As in many file systems, the GFS2 allocator tries to keep blocks in the same file close to one another to reduce the movement of disk heads and boost performance. A node that allocates blocks to a file will likely need to use and lock the same resource groups for the new blocks (unless all the blocks in that resource group are in use). The file system will run faster if the lock master for the resource group containing the file allocates its data blocks (it is faster to have the node that first opened the file do all the writing of new blocks).
2.3.3. Preallocate, If Possible
If files are preallocated, block allocations can be avoided altogether and the file system can run more efficiently. Newer versions of GFS2 include the
fallocate(1) system call, which you can use to preallocate blocks of data.