MemFS 2.0 - A Memory-based File System on HP-UX 11iv3

6
Performance
A MemFS is expected to perform much better than normal disk file system for smaller file and file
system sizes. As the memory occupied by MemFS increases to a level where swapping starts
happening, the performance may begin to taper off. If the swapping becomes excessive,
performance will decrease. MemFS performance varies depending on usage and machine
configurations. The benchmark tests chosen to characterize the performance of MemFS are:
Postmark, Connectathon basic tests and SDET.
Postmark Benchmark
Network Appliance‟s Postmark is an appropriate benchmark to use for MemFS as it is an industry-
standard benchmark for small file and metadata-intensive workloads. It was designed to emulate
Internet applications such as e-mail, netnews, and e-commerce.
Postmark works by creating a configurable sized pool of random text files ranging in size
between configurable high and low bounds. These files are continually modified. The building of
the file pool allows for the production of statistics on small file creation performance. Transactions
are then performed on the pool. Each transaction consists of a pair of create/delete or
read/append operations. The use of randomization and the ability to parameterize the proportion
of create/delete to read/append operations are used to overcome the effects of caching in
various parts of the system.
Test Configuration: The system under test was on HP rx7620 comprising of 2 x 1600MHz Itanium
CPUs, 16GB RAM. Postmark v1.5 was used. The following graphs show the effects of applying
increasing transaction loads to the various file systems. . Figure 2 is an illustration of the
performance of various file systems on various workloads.