Changes between Version 1 and Version 2 of jazz/12-06-02

Timestamp:

Jun 3, 2012, 11:27:27 AM (12 years ago)

Author:

jazz

Comment:

--

jazz/12-06-02

@@ -3 +3 @@
 == File System & Kernel Page Size ==
 
- * 前陣子大家在討論怎樣加速異地檔案傳輸的速度，Thomas 提到說 mount 指令不支援大於 4096 (4K) 的問>題，也提到這個現象受限於 CPU 架構，因為 Kernel Page Size 是相依於 CPU 架構。今天簡單地查了一下，>把這個問題紀錄一下，畢竟我們常要去探討 Disk I/O 速度，從硬體架構跟軟體架構的堆疊來看，最終還是得>設法找出最佳化的參數。唯有對問題更了解，才又辦法知道如何把速度榨出來～
- * [參考一] [http://mkl-note.blogspot.tw/2010/11/mount-xfs-64kb-block-size-cause.html mount XFS 64KB block size cause "Function not implemented"]
+ * 前陣子大家在討論怎樣加速異地檔案傳輸的速度，Thomas 提到說 mount 指令不支援大於 4096 (4K) 的問>題，也提到這個現象受限於 CPU 架構，因為 Kernel Page Size 是相依於 CPU 架構。今天簡單地查了一下，把這個問題紀錄一下，畢竟我們常要去探討 Disk I/O 速度，從硬體架構跟軟體架構的堆疊來看，最終還是得>設法找出最佳化的參數。唯有對問題更了解，才又辦法知道如何把速度榨出來～
+ * 結論(1) : '''檔案系統的 Block Size （檔案區塊大小）最大值等於 Linux Kernel 的 Page Size （虛擬記憶體分頁大小）'''
+  * [參考一] [http://mkl-note.blogspot.tw/2010/11/mount-xfs-64kb-block-size-cause.html mount XFS 64KB block size cause "Function not implemented"]
 {{{
 #!html
-<p><b>XFS: A high-performance journaling filesystem</b></p>
-<a href='http://oss.sgi.com/projects/xfs/'>http://oss.sgi.com/projects/xfs/</a><p/>
-<b>Filesystem Block Size</b><p/>
-          <u>The minimum filesystem block size is 512 bytes. <b>The maximum filesystem block size is the page size of the kernel</b>, which is 4K on x86 architecture and is set as a kernel compile option on the IA64 architecture (up to 64 kilobyte pages).</u> So, XFS supports filesystem block sizes up to 64 kilobytes (from 512 bytes, in powers of 2), when the kernel page size allows it.
+<p><b>XFS: A high-performance journaling filesystem</b><br/>
+<a href='http://oss.sgi.com/projects/xfs/'>http://oss.sgi.com/projects/xfs/</a><br/>
+<b>Filesystem Block Size</b><br/>
+          <u>The minimum filesystem block size is 512 bytes. <b><font color='red'>The maximum filesystem block size is the page size of the kernel</font></b>, which is 4K on x86 architecture and is set as a kernel compile option on the IA64 architecture (up to 64 kilobyte pages).</u> So, XFS supports filesystem block sizes up to 64 kilobytes (from 512 bytes, in powers of 2), when the kernel page size allows it.</p>
 }}}
- * [參考二] [http://superuser.com/questions/291228/mount-ext4-partition-with-4kib-block-size Mount ext4 partition with >4KiB block size]
+ * 結論(2) : '''支援大於 4K Page Size 的硬體架構為 ia64, mips, pa-risc, powerpc, sh, sparc64'''
+  * [參考二] [http://superuser.com/questions/291228/mount-ext4-partition-with-4kib-block-size Mount ext4 partition with >4KiB block size]
 {{{
 #!text
@@ -28 +30 @@
 it to find an old-PPC Mac.
 }}}
- * [參考三] [
+ * 那如何查出目前的 Kernel Page Size 呢？
+  * [參考三] [http://www.cyberciti.biz/faq/linux-check-the-size-of-pagesize/ Linux Find Out Virtual Memory PAGESIZE]
+{{{
+~$ getconf PAGESIZE
+OR
+~$ getconf PAGE_SIZE
+}}}
+ * 結論(3):'''除了加大 Block Size 之外，最常用的方式是啟動 read ahead （運用快取來提昇硬碟循序讀取的效率）'''
+  * [參考四] [http://forums.opensuse.org/forums/english/get-technical-help-here/install-boot-login/437078-changing-pagesize-kernel.html#post2152312 Changing PAGESIZE in kernel]
+{{{
+#!text
+AIX vio ioo is a read ahead cache system to improve speed of transfers when doing sequential reads.
+}}}
+  * [參考五] [http://www.linuxpinguin.de/2011/01/performance-increase-through-bigger-readahead-buffers/ Performance increase through bigger readahead buffers] - 以下示範是把預設的 RA 由 256 提升到 1024
+{{{
+~$ blockdev --report
+RO    RA   SSZ   BSZ   起始磁區            大小   裝置
+rw   256   512  4096          0    500107862016   /dev/sda
+rw   256   512  4096       2048    497999151104   /dev/sda1
+rw   256   512  4096  972656640      2106589184   /dev/sda2
+~$ sudo blockdev --setra 1024 /dev/sda1
+}}}