[HBASE-13259] mmap() based BucketCache IOEngine

• Type: New Feature
• Status: Closed
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: 0.98.10
• Fix Version/s: 2.0.0
• Component/s: BlockCache
• Hadoop Flags: Reviewed

Release Note

mmap() based bucket cache can be configured by specifying the property
{code}
<property>
  <name>hbase.bucketcache.ioengine</name>
  <value> mmap://filepath </value>
</property>
{code}
This mode of bucket cache is ideal when your file based bucket cache size is lesser than then available RAM. When the cache is bigger than the available RAM then the kernel page faults will make this cache perform lesser particularly in case of scans.

• Assignee: Zee Chen
• Reporter: Zee Chen
• Created: 17/Mar/15 01:20
• Updated: 23/Jun/22 20:40
• Resolved: 23/Feb/16 11:39

Description

Of the existing BucketCache IOEngines, FileIOEngine uses pread() to copy data from kernel space to user space. This is a good choice when the total working set size is much bigger than the available RAM and the latency is dominated by IO access. However, when the entire working set is small enough to fit in the RAM, using mmap() (and subsequent memcpy()) to move data from kernel space to user space is faster. I have run some short keyval gets tests and the results indicate a reduction of 2%-7% of kernel CPU on my system, depending on the load. On the gets, the latency histograms from mmap() are identical to those from pread(), but peak throughput is close to 40% higher.

This patch modifies ByteByfferArray to allow it to specify a backing file.

Example for using this feature: set hbase.bucketcache.ioengine to mmap:/dev/shm/bucketcache.0 in hbase-site.xml.

Attached perf measured CPU usage breakdown in flames graph.