Hi, my apology if this question is to the wrong group but I think this is teh group most qualified to explain the cause of the problem I saw. Basically our application uses about 5 GB heap space (JVM), which memory-maps about 1500 files, 100 MB, for total of 15 GB. The files are read-mostly and many pages can be inactive. The server runs RedHat 2.6.32-220.10.1.el6.x86_64, has 2 CPUs, 24 cores including hyper-threading. It has about 48 GB and thus plenty of free memory as shown meminfo output below. swappiness is set to 0. Other Linux kernel tunings are by default from RH. We ran into very intense pgscand/s even if free memory is 16 GB and we correlated it with counts of isolated files/anon. Needless to say, when this happens, our app slows down considerable. We would be grateful if you could explain what happened and what settings we should apply. Thank you in advance, Output of meminfo below: MemTotal: 49358752 kB MemFree: 16098532 kB Buffers: 335788 kB Cached: 26414224 kB SwapCached: 0 kB Active: 6444192 kB Inactive: 25673036 kB Active(anon): 5366776 kB Inactive(anon): 840 kB Active(file): 1077416 kB Inactive(file): 25672196 kB Unevictable: 0 kB Mlocked: 0 kB SwapTotal: 16776116 kB SwapFree: 16776116 kB Dirty: 260 kB Writeback: 0 kB AnonPages: 5366888 kB Mapped: 17348504 kB Shmem: 372 kB Slab: 530388 kB SReclaimable: 466752 kB SUnreclaim: 63636 kB KernelStack: 8864 kB PageTables: 55844 kB NFS_Unstable: 0 kB Bounce: 0 kB WritebackTmp: 0 kB CommitLimit: 61198992 kB Committed_AS: 8475504 kB VmallocTotal: 34359738367 kB VmallocUsed: 383776 kB VmallocChunk: 34332638712 kB HardwareCorrupted: 0 kB AnonHugePages: 4958208 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 2048 kB DirectMap4k: 4676 kB DirectMap2M: 2027520 kB DirectMap1G: 48234496 kB -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>