真是天下最大的笑話,由國外,下載國內的 FTP 站的速度,比我們的種花電信,還要快
說什麼發展數位內容,雲端技術,光是網路建設不進步,只知道壟斷,吸老百姓的鈔票,真不知道要發展什麼
terry@www:/tmp$ wget http://ftp.tku.edu.tw/debian-cd/5.0.3/amd64/iso-cd/debian-503-amd64-CD-1.iso
–2010-07-08 16:33:44– http://ftp.tku.edu.tw/debian-cd/5.0.3/amd64/iso-cd/debian-503-amd64-CD-1.iso
Resolving ftp.tku.edu.tw… 163.13.240.201
Connecting to ftp.tku.edu.tw|163.13.240.201|:80… connected.
HTTP request sent, awaiting response… 200 OK
Length: 678301696 (647M) [application/x-iso9660-image]
Saving to: debian-503-amd64-CD-1.iso'<br /><br />100%[===================================================================================>] 678,301,696 10.3M/s in 69s <br /><br />2010-07-08 16:34:54 (9.31 MB/s) -
debian-503-amd64-CD-1.iso’ saved [678301696/678301696]
Posts for: #VPS
NEW VPS HazeNet
買了新的 VPS
RAM 1024 MB,OpenVN,現在 40% OFF,花了 8.4 USD,真是超值,網路實測,封包不掉,到台灣學術往錄下載 ISO 檔,2host 因為是沒有限速,可以跑到1~3M之間網路速度,2host 算是大勝,因為 10TB ,幾乎算是不限頻寬, HazeNet 的VPS 大約在 600 KB/s 左右,跟我選的規格上面限速的差不多,他可以選,UNMETERED 5mbit,或是 500GB @ 50mbit,我暫時還是跑不限流量的,在 SolusVM 的管理介面裡,所有的資訊都會跟 VPS 的狀態同步,資訊比 2host 的圖表資料多,不過接下來,就看穩定性,還有看他們會不會跑路了
http://hazenet.co.uk/vps.html#
==============================================================
BYTE UNIX Benchmarks (Version 4.1-wht.2)
System – Linux www 2.6.18-194.3.1.el5.028stab069.6 #1 SMP Wed May 26 18:31:05 MSD 2010 i686 i686 i386 GNU/Linux
/dev/simfs 41943040 494132 41448908 2% /
Start Benchmark Run: Tue Jul 6 20:18:41 MSD 2010
20:18:41 up 6:11, 1 user, load average: 0.03, 0.03, 0.00
End Benchmark Run: Tue Jul 6 20:29:08 MSD 2010
20:29:08 up 6:21, 1 user, load average: 11.50, 5.26, 2.36
INDEX VALUES
TEST BASELINE RESULT INDEX
Dhrystone 2 using register variables 376783.7 18596539.0 493.6
Double-Precision Whetstone 83.1 1446.7 174.1
Execl Throughput 188.3 16969.6 901.2
File Copy 1024 bufsize 2000 maxblocks 2672.0 345072.0 1291.4
File Copy 256 bufsize 500 maxblocks 1077.0 98371.0 913.4
File Read 4096 bufsize 8000 maxblocks 15382.0 2285021.0 1485.5
Pipe-based Context Switching 15448.6 891549.4 577.1
Pipe Throughput 111814.6 3107472.1 277.9
Process Creation 569.3 41569.8 730.2
Shell Scripts (8 concurrent) 44.8 2667.6 595.4
System Call Overhead 114433.5 2556660.4 223.4
=========
FINAL SCORE 571.3
順便在貼一下前一篇 2host Xen 512 MB 的 VPS 分數
BYTE UNIX Benchmarks (Version 4.1-wht.2)
System – Linux www.?????.com 2.6.18-164.15.1.el5xen #1 SMP Wed Mar 17 12:04:23 EDT 2010 x86_64 GNU/Linux
/dev/sda1 10321208 2974096 6822824 31% /
Start Benchmark Run: Tue Jun 29 17:17:07 CST 2010
17:17:07 up 5:08, 3 users, load average: 0.19, 0.17, 0.21
End Benchmark Run: Tue Jun 29 17:27:23 CST 2010
17:27:23 up 5:18, 3 users, load average: 16.73, 7.01, 3.15
INDEX VALUES
TEST BASELINE RESULT INDEX
Dhrystone 2 using register variables 376783.7 10227540.3 271.4
Double-Precision Whetstone 83.1 1484.0 178.6
Execl Throughput 188.3 1972.1 104.7
File Copy 1024 bufsize 2000 maxblocks 2672.0 108726.0 406.9
File Copy 256 bufsize 500 maxblocks 1077.0 35313.0 327.9
File Read 4096 bufsize 8000 maxblocks 15382.0 1262429.0 820.7
Pipe-based Context Switching 15448.6 230394.6 149.1
Pipe Throughput 111814.6 1214489.5 108.6
Process Creation 569.3 3573.8 62.8
Shell Scripts (8 concurrent) 44.8 591.7 132.1
System Call Overhead 114433.5 1044142.2 91.2
=========
FINAL SCORE 181.0
PS: 分數不代表一切,也不代表,效能你可以全用,不過爽度也蠻重要的,HazeNet 算是非常新的 VPS,可能等他一切都穩定了才知道,到底行不行,我三天,被換了,兩個 IP,老闆說,他放在 isp 的 server 出了點問題,所以現在是 offline 的
2host 心得
一開始使用時候,做一些壓力測試分數不算是突出的 VPS,可是實際使用起來,穩定性還不錯,網路連美國本土,還有台灣速度都還不錯,頻寬一個月10T 幾乎是無限使用,我根本用不完,網站跑起來,速度一點也不輸給,原本放在種花電信的速度,所以開始不用把 server 再放在自己家中,雖然管起來也很輕鬆,不過,要響應節能減碳,既然,自己的 server 用電一個月都花得比買 VPS 還貴,乾脆就用 VPS 吧,目前用一個星期來,很滿意,除了,有一次,我跑壓力測試,測的太兇,把 512 MB 的記憶體,還有 1G 的 SWAP 吃光光,機器自己 reboot 外,目前用起來,還蠻穩定的,跑兩個站,用 Django跑 fastcgi 再 nginx 後面,吃不到 200 MB (32 bit OS 可以吃更少),不過提供的 kernel 有點舊了,我用的是 Debian 2.6.18-164.15.1.el5xen,server 穩定就好了,也不必一昧的追新,這是使用一個星期的感想
這一星期,一直在比其他 VPS 的穩定性, unixbench 分數,還有就是網友的口碑,最好的是 Linode 不過價格就很硬了,現在記憶 512 MB的方案 19.95 元美金,和 2host 7.99 元美金貴了很多,雖然硬碟空間,還有 unixbench 及 redis-benchmark 的分數比較高,不過便宜是王,只要穩定性不差,就好了,另一家是 ramhost 口碑不錯,資訊很公開,有實際每一台主機的狀態,算是沒有超賣有口碑的 VPS 供應商,不過他是 OpenVZ 的我比較不喜歡,還是喜歡 Xen 的 VPS,資源分的比較乾淨,所以這完全是對個人口味的選擇。
另一點重要的是,他提供的價錢很迷人,頻寬很大,所以定期的備份是一定的,怕哪一天他跑路,這是不管用哪一家都必須注意的問題
再穩穩跑一陣子,以他連美國本土的速度優勢,程式交易,下美國的商品,應該延遲就可以贏很大了
最後說說缺點
管理介面的數值報表幾乎沒有,
硬碟空間,我選的 512 MB 的這一個方案只有 10G 的硬碟空間,Debian 只有 64 bit 可以選
還有更便宜的,像是新公司 http://hazenet.co.uk/
買了 2host 512 RAM XEN VPS
2010-06-29 更正,實際分數跑起來,不算高,我問問看客服,有沒有官方的數據可以比較,還是要多比較一下,我會再看看 Linode 的
便宜又大碗,先用用看,目前還蠻穩定的,連台灣,網路還算 OK
我是選 512M的 VPS 一個月 8 塊美金左右,大概比我自己擺 server 的電費還省,下面是 CPU info 可以參考一下,不過,我自己跑起來實測,有比 Amazon em2 的最便宜的 small instance 大約快兩倍,不過就比不過 middle 的了,如果有問題就發問吧
跑了 nginx,mysql ,兩個 Django site 大概用了 150M RAM
cat /proc/cpuinfo
processor : 3
vendor_id : GenuineIntel
cpu family : 6
model : 26
model name : Intel(R) Xeon(R) CPU E5520 @ 2.27GHz
stepping : 5
cpu MHz : 2260.998
cache size : 8192 KB
physical id : 3
siblings : 1
core id : 0
cpu cores : 1
fpu : yes
fpu_exception : yes
cpuid level : 11
wp : yes
flags : fpu tsc msr pae mce cx8 apic mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc pni monitor ds_cpl vmx est tm2 cx16 xtpr lahf_lm
bogomips : 5655.56
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management:
========================================================================
BYTE UNIX Benchmarks (Version 5.1.2)
System: www.digez.com: GNU/Linux
OS: GNU/Linux – 2.6.18-164.15.1.el5xen – #1 SMP Wed Mar 17 12:04:23 EDT 2010
Machine: x86_64 (unknown)
Language: en_US.utf8 (charmap=“UTF-8”, collate=“UTF-8”)
CPU 0: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (5662.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (5662.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET, Intel virtualization
CPU 2: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (5662.0 bogomips)
CPU 3: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (5662.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSCALL/SYSRET, Intel virtualization
14:53:21 up 2:44, 3 users, load average: 0.11, 0.09, 0.07; runlevel 3
————————————————————————
Benchmark Run: Tue Jun 29 2010 14:53:21 - 15:21:28
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 9705725.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2545.7 MWIPS (9.2 s, 7 samples)
Execl Throughput 1004.9 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 194004.6 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 50378.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 530237.1 KBps (30.0 s, 2 samples)
Pipe Throughput 340164.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 68815.1 lps (10.0 s, 7 samples)
Process Creation 2548.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2485.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 582.7 lpm (60.0 s, 2 samples)
System Call Overhead 366327.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 9705725.4 831.7
Double-Precision Whetstone 55.0 2545.7 462.9
Execl Throughput 43.0 1004.9 233.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 194004.6 489.9
File Copy 256 bufsize 500 maxblocks 1655.0 50378.3 304.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 530237.1 914.2
Pipe Throughput 12440.0 340164.4 273.4
Pipe-based Context Switching 4000.0 68815.1 172.0
Process Creation 126.0 2548.5 202.3
Shell Scripts (1 concurrent) 42.4 2485.8 586.3
Shell Scripts (8 concurrent) 6.0 582.7 971.1
System Call Overhead 15000.0 366327.5 244.2
========
System Benchmarks Index Score 398.9
————————————————————————
Benchmark Run: Tue Jun 29 2010 15:21:28 - 15:50:39
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 36206722.0 lps (10.0 s, 7 samples)
Double-Precision Whetstone 10039.4 MWIPS (9.0 s, 7 samples)
Execl Throughput 2053.0 lps (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 139297.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 39443.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 413460.4 KBps (30.0 s, 2 samples)
Pipe Throughput 794207.7 lps (10.1 s, 7 samples)
Pipe-based Context Switching 197504.1 lps (10.0 s, 7 samples)
Process Creation 3082.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 3750.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 498.9 lpm (60.2 s, 2 samples)
System Call Overhead 848080.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 36206722.0 3102.5
Double-Precision Whetstone 55.0 10039.4 1825.4
Execl Throughput 43.0 2053.0 477.4
File Copy 1024 bufsize 2000 maxblocks 3960.0 139297.5 351.8
File Copy 256 bufsize 500 maxblocks 1655.0 39443.2 238.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 413460.4 712.9
Pipe Throughput 12440.0 794207.7 638.4
Pipe-based Context Switching 4000.0 197504.1 493.8
Process Creation 126.0 3082.2 244.6
Shell Scripts (1 concurrent) 42.4 3750.5 884.5
Shell Scripts (8 concurrent) 6.0 498.9 831.6
System Call Overhead 15000.0 848080.4 565.4
========
System Benchmarks Index Score 645.0
===================================================================================
BYTE UNIX Benchmarks (Version 4.1-wht.2)
System – Linux www.?????.com 2.6.18-164.15.1.el5xen #1 SMP Wed Mar 17 12:04:23 EDT 2010 x86_64 GNU/Linux
/dev/sda1 10321208 2974096 6822824 31% /
Start Benchmark Run: Tue Jun 29 17:17:07 CST 2010
17:17:07 up 5:08, 3 users, load average: 0.19, 0.17, 0.21
End Benchmark Run: Tue Jun 29 17:27:23 CST 2010
17:27:23 up 5:18, 3 users, load average: 16.73, 7.01, 3.15
INDEX VALUES
TEST BASELINE RESULT INDEX
Dhrystone 2 using register variables 376783.7 10227540.3 271.4
Double-Precision Whetstone 83.1 1484.0 178.6
Execl Throughput 188.3 1972.1 104.7
File Copy 1024 bufsize 2000 maxblocks 2672.0 108726.0 406.9
File Copy 256 bufsize 500 maxblocks 1077.0 35313.0 327.9
File Read 4096 bufsize 8000 maxblocks 15382.0 1262429.0 820.7
Pipe-based Context Switching 15448.6 230394.6 149.1
Pipe Throughput 111814.6 1214489.5 108.6
Process Creation 569.3 3573.8 62.8
Shell Scripts (8 concurrent) 44.8 591.7 132.1
System Call Overhead 114433.5 1044142.2 91.2
=========
FINAL SCORE 181.0
redis-benchmark
====== PING ======
10024 requests completed in 0.29 seconds
50 parallel clients
3 bytes payload
keep alive: 1
8.30% <= 0 milliseconds
64.88% <= 1 milliseconds
85.17% <= 2 milliseconds
99.87% <= 3 milliseconds
100.00% <= 4 milliseconds
34328.77 requests per second
====== PING (multi bulk) ======
10011 requests completed in 0.22 seconds
50 parallel clients
3 bytes payload
keep alive: 1
10.27% <= 0 milliseconds
82.24% <= 1 milliseconds
99.25% <= 2 milliseconds
99.90% <= 3 milliseconds
100.00% <= 4 milliseconds
45094.59 requests per second
====== SET ======
10001 requests completed in 0.23 seconds
50 parallel clients
3 bytes payload
keep alive: 1
8.30% <= 0 milliseconds
83.05% <= 1 milliseconds
99.10% <= 2 milliseconds
99.33% <= 3 milliseconds
99.36% <= 4 milliseconds
99.39% <= 5 milliseconds
99.42% <= 6 milliseconds
99.44% <= 7 milliseconds
99.47% <= 8 milliseconds
99.49% <= 9 milliseconds
99.80% <= 10 milliseconds
100.00% <= 11 milliseconds
43294.37 requests per second
====== GET ======
10001 requests completed in 0.23 seconds
50 parallel clients
3 bytes payload
keep alive: 1
8.78% <= 0 milliseconds
83.75% <= 1 milliseconds
99.21% <= 2 milliseconds
99.51% <= 3 milliseconds
99.64% <= 7 milliseconds
99.83% <= 8 milliseconds
100.00% <= 9 milliseconds
43672.49 requests per second
VPS 找便宜
找便宜的 VPS 的好站,以前,都只看 Linode,價格很久沒動了,所以想找便宜的可以試試這一個站,我自己也記一下,免得忘了
http://vpslook.com/