ÿØÿà JFIF    ÿÛ „  ( %!1!%)+//.383,7(-.+  -%%-////---/-.+/--+------/------/--0+--/-/-----.-----ÿÀ  ¥2" ÿÄ     ÿÄ J    ! 1AQ"aq2‘#BR‚¡ÁÑ3br’¢±Âð$CSƒ²á4c“%DsÓñÿÄ   ÿÄ *  !1AQa‘"2q3±ð#b¡ÿÚ   ? ¼QxJQaÍuò¸Zö Úü8,ÐÚú "SSn<rçù–´âE—^ªBÖ9À\†¸ÔÁT­ÃÛ5 ëd´³Í#Ý;Þ38œî ¶H£M:wÎ3…³…âpÔF&‚FK¸9„â4àGEõªfÿ ‘ñ(ßw­pŽF|È¥ù®häðÍѶ¹‘[ÒinÙW¶ùñY˜Q{›K"išÒ[Ú8žë\F¹@-?v"ÔU”,ìöžkÿ {I‡£šÍ?e ríV .............................................................................................................................................................................. ............................................................................. ÿØÿà JFIF    ÿÛ „  ( %!1!%)+//.383,7(-.+  -%%-////---/-.+/--+------/------/--0+--/-/-----.-----ÿÀ  ¥2" ÿÄ     ÿÄ J    ! 1AQ"aq2‘#BR‚¡ÁÑ3br’¢±Âð$CSƒ²á4c“%DsÓñÿÄ   ÿÄ *  !1AQa‘"2q3±ð#b¡ÿÚ   ? ¼QxJQaÍuò¸Zö Úü8,ÐÚú "SSn<rçù–´âE—^ªBÖ9À\†¸ÔÁT­ÃÛ5 ëd´³Í#Ý;Þ38œî ¶H£M:wÎ3…³…âpÔF&‚FK¸9„â4àGEõªfÿ ‘ñ(ßw­pŽF|È¥ù®häðÍѶ¹‘[ÒinÙW¶ùñY˜Q{›K"išÒ[Ú8žë\F¹@-?v"ÔU”,ìöžkÿ {I‡£šÍ?e ríV .............................................................................................................................................................................. ............................................................................. tuned.conf000064400000002560151557615600006550 0ustar00# # tuned configuration # [main] summary=Optimize for HPC compute workloads description=Configures virtual memory, CPU governors, and network settings for HPC compute workloads. include=latency-performance [vm] # Most HPC application can take advantage of hugepages. Force them to on. transparent_hugepages=always [disk] # Increase the readahead value to support large, contiguous, files. readahead=>4096 [sysctl] # Keep a reasonable amount of memory free to support large mem requests vm.min_free_kbytes=135168 # Most HPC applications are NUMA aware. Enabling zone reclaim ensures # memory is reclaimed and reallocated from local pages. Disabling # automatic NUMA balancing prevents unwanted memory unmapping. vm.zone_reclaim_mode=1 kernel.numa_balancing=0 # Busy polling helps reduce latency in the network receive path # by allowing socket layer code to poll the receive queue of a # network device, and disabling network interrupts. # busy_read value greater than 0 enables busy polling. Recommended # net.core.busy_read value is 50. # busy_poll value greater than 0 enables polling globally. # Recommended net.core.busy_poll value is 50 net.core.busy_read=50 net.core.busy_poll=50 # TCP fast open reduces network latency by enabling data exchange # during the sender's initial TCP SYN. The value 3 enables fast open # on client and server connections. net.ipv4.tcp_fastopen=3