《Solaris 性能参数监控及调试》由会员分享，可在线阅读，更多相关《Solaris 性能参数监控及调试（75页珍藏版）》请在金锄头文库上搜索。
1、Solaris 性能参数监控及调试Jian Wang Solution Architect Greater China Client Services J我们首先要做的问客户正确的问题打没打补丁？在哪个环节出现问题？ 吞吐量, 响应时间, 网络 , I/O, 应用,明白客户怎么和怎样提出要求测试和想法是否合理 ?它应该跑得更快 ., 很多的CPU空闲 . 开始的支持和建议收集基本的性能参数使用工具进行分析 , 将图形化结果展现给客户 , 观察所谓的 峰值识别资源的限制 用自己的发现提出自己的建议Solaris 性能分析的目的识别Solaris操作系统的瓶颈做出增加性能的适当建议并不作为应用的调
2、试和Kernel调试从基本的统计数据中发现有用的东西4首先注意不要着急!有果必有因!Panic性能管理Section 1 性能管理的定义性能管理就是：测量, 分析和优化计算 资源，以提供给最终用户可认可的服务一般术语Throughput Count of number of transactions Latency Time it takes to do something Utilization Amount of resources consumed during action性能的层面应用 操作系统 硬件 网络 商务Solaris 性能分析Section 3 10基本的原则系统是一系列的
3、资源CPU(s)memorybussesdisks, disk controllersnetworksoperating systemsDBMS systems (especially locks and internal latches)性能的降低是以上一个或几个资源枯竭所 造成的。11从哪做起一旦瓶颈被发现，按下列顺序调试：应用数据库硬件Solaris 内核参数潜在的瓶颈问题Disk Network Memory CPU磁盘 性能分析Section 3 硬盘瓶颈无足够空间 响应时间长 不良的规划 RAID 设定文件系统问题 数据库问题确认硬盘瓶颈使用 sar, iostat 察看 响应时间
4、, 磁盘使用率, 队列长 度, 请求分配了解磁盘缓存是否工作相关的访问时间DeviceRealTime Second Rel. timeCPU Register2 nsec2 x 10-92 sec. CPU cache20 nsec 2x10-920 sec. Main Memory2000 nsec.2x10-82-3 min. Disk20msec 2x10-37 monthiostat -x 30extended device statistics device r/s w/s kr/s kw/s wait actv svc_t %w %b fd0 0.0 0.0 0.0 0.0 0.
5、0 0.0 279.0 0 0 sd0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 sd1 0.1 0.9 0.7 6.3 0.0 0.1 72.1 0 1 sd4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 sd6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 nfs1 0.0 0.0 0.0 0.0 0.0 0.0 0.7 0 0 nfs2 0.0 0.0 0.1 0.4 0.0 0.0 298.1 0 0 nfs3 0.0 0.0 0.2 0.0 0.0 0.0 35.4 0 0 Wait 队列长度, 等待磁盘处理的请求个数 sv
6、c_t 平均服务时间（毫秒） %w 等待处理请求次数的百分比 %b 忙时百分比 60%有问题iostatI/O size = (kr/s) / (r/s) srv_t 接近磁盘寻址时间 Wait : # OS对磁盘请求等待, 0 warning Actv : # 在磁盘处理中而未完成的请求个 数iostatr/s, w/s: Average reads/writes per second. Kr/s, Kw/s: Average Kb read/written per second. wait: Time spent by a process while waiting for block (
7、eg disk) I/O to complete. actv: Number of active requests in the hardware queue. %w: Occupancy of the wait queue. %b: Occupancy of the active queue with the device busy. svc_t: Service time (ms). Includes everything: wait time, active queue time, seek rotation, transfer time. us/sy: User/system CPU
8、time (%). wt: Wait for I/O (%). id: Idle time (%). 使用 iostat查问题The %busy column shows disk utilization. Disks utilized over 65% are a problem; disks over 30%- 35% can be improved by RAID mehtods. Any disk utilized more than 95% is a SERIOUS problem.The service time shows how long an individual disk
9、request waits. It should be close to the average seek time of the disk. Spikes of high svc_t are not a problem.This disk is a typical problem disk. The actv column shows that there are 20 outstanding I/O requests against this drive, in addition to being fully busy and very slow.This disk is not a pr
10、oblem, despite its high service time, because it is not very busy. It is often useful to compute average I/O size from r/s and Kr/s.# iostat -xc 2extended device statistics cpu device r/s w/s kr/s kw/s wait actv svc_t %w %b us sy wt id md0 41.0 0.0 382.7 0.0 0.0 0.6 14.9 0 41 41 59 0 0 c0t3d0 43.9 2
11、.8 405.4 4.1 0.0 0.7 14.7 1 45 c1t0d0 21.9 2.8 203.8 4.1 0.0 0.3 13.9 0 31 c2t0d0 21.9 2.8 201.6 4.1 0.0 0.3 13.7 0 31 c7t2d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c7t3d0 0.0 0.1 0.0 2.0 0.0 0.0 159.0 0 1 c7t4d0 115.9 2.8 228.4 4.1 0.0 19.7 311.7 1 99Typical I/O size on this disk is Kr / r; in this case t
12、ypical I/O size is 224.8 / 115 = 2K.svc_t includes time waiting in the queue. When svc_t is high, the service time is approx svc_t / actv, about 15ms in this case.actv and wait are respectively the number of requests pending in the disk itself and in the OS waiting to get to the disk. wait 0 is a wa
13、rning sign, especially if the disk is in a disk array.iostat磁盘名称 (-n)I/O 每个分区 (-p)可显示磁带的 I/O shownCPU占用资源（-c)显示每个磁盘的信息 (-x)# iostat -xcpn 60extended device statistics cpu device r/s w/s kr/s kw/s wait actv svc_t %w %b us sy wt id rmt0 48.3 0.0 14382.7 0.0 0.0 0.6 0.0 0 91 41 59 0 0 c0t6d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c11t14d0 0.2 0.1 1.4 1.2 0.0 0.0 18.6 0 0 c11t14d0s0 0.0 0.1 0.4 1.2 0.0 0.0 24.9 0 0 c11t14d0s1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c11t14d0s2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0
