VolanoMark 2.1.2. Loopback benchmark was run to measure the impact of the rmap roll up patch developed by Dave McCracken. The rmap roll up patch did not improve or degrade VolanoMark Loopback performance. Hardware Configuration: ----------------------------------- 8 -way Intel Pentium III Xeon 700 MHZ Software Configuration: ---------------------------------- Linux 2.5.26 (from kernel.org) and rmap rollup patch IBM JVM J2RE 1.3.1. IBM build cxia32131-20020622 (JIT enabled; jitc) JVM Heap 256 MB Loopback MTU 512 bytes 1000 msgs were sent by each client and 10 chat rooms were used. For more information about VolanoMark, please visit http://www.volano.com/benchmarks.html |--------------------------+------------+--------------------------| | | 2.5.26 base| 2.5.26 base + rmap | | | | roll_up | |--------------------------+------------+--------------------------| | Avg Throughput (msg/sec) | 50984 | 50775 | |--------------------------+------------+--------------------------| | Convergence | 1.180 | 0.551 | |--------------------------+------------+--------------------------| | user | 39.56% | 40.61% | |--------------------------+------------+--------------------------| | System | 54.17% | 52.57% | |--------------------------+------------+--------------------------| | Idle | 2% - 3% | 2% - 3% | |--------------------------+------------+--------------------------| Partha ------------------------------------------------------------------------------- Partha Narayanan Linux Technology & Solutions Centre - Performance (Tel) 512-838-8893 T/L 678-8893 email: partha@us.ibm.com dmccr@us.ltcfwd.linux.ibm .com To: Linux Scalability Effort List , Linux Sent by: Memory Management lse-tech-admin@lists.sour cc: ceforge.net Subject: [Lse-tech] Rollup patch of basic rmap against 2.5.26 09/17/2002 01:21 PM Over the past couple of weeks we've been doing some basic performance testing of the rmap overhead. For this I put together a rollup patch against 2.5.26 that includes what I'd consider basic rmap. As a reference, I'm also posting the patch here, so people can see what it consists of. The list of patches included are: minrmap The original minimal rmap patch truncate_leak A bug fix dmc_optimize Don't allocate pte_chain for one mapping vmstat Add rmap statistics for vmstat ptechain slab Allocate pte_chains from a slab daniel_rmap_speedup Use hashed pte_chain locks akpm_rmap_speedup Make pte_chain hold multiple pte ptrs Again, this patch applies against 2.5.26, and clearly does not include many of the recent rmap optimizations. Dave McCracken ====================================================================== Dave McCracken IBM Linux Base Kernel Team 1-512-838-3059 dmccr@us.ibm.com T/L 678-3059 (See attached file: rmap-rollup-2.5.26.diff)