EExcel 丞燕快速查詢2
EExcel 丞燕快速查詢2 https://sandk.ffbizs.com/
一周一次的集保股權階怖出爐了
2018/1/27 12:42
之前宏碁在上漲時，1~10張的打死都不會增加，現在主力拉了一個錨點33.75元然後跌下來到27.6元，這些散戶突然間就覺得宏碁好便宜啊，
之前同事在20，22時一直求我賣，他說你賣個3張也好啦，都從14漲到22了，都漲8元了，不可能這樣一直漲上去啦，我沒有理他，結果股價從22.x跌到20.x後，他就說，你看吧，跌了吧，股價漲上去你沒有賣都是假的啦，我也沒有理會他。直到宏碁一直漲，狂漲到33.75元後，跌到30.x元，他說他要買了，我聽了差點沒有昏倒，為什麼22你叫我賣，30.x你卻要買？宏碁基本面有改變嗎？似乎沒有，所以他這症狀就是大部散戶的症狀，中了錨定效和加一點後悔效應，覺得宏碁從14漲到33.75一路漲沒賺到，心理一直一直很恨，結果33.75跌到30，甚至是27.3元，他們就會覺得這是從天上掉下來的禮物，趕快加碼買進。
這周1~10張的散戶狂增9113人，買了股票24990張，算是非理性的買法，這就是成交量很大所會造成的現象，不要天真的以為什麼10萬張籌定籌碼可以避免這周從33.75跌到27.6,我說過，這叫做狗犬火車頭，不要以為火車是被你吠走的，這是市場的趨勢，必然會發生的事，這人性弱點叫"代表性原則錯誤"，常常看我的文章，或聽到某個人說賣出，就以為整件事是因他而發生的，之前ptt有一個人賣了800張，結果ptt就一片拜神說，原來宏碁的大跌是他引發的。一位宏碁股東賣出了9xx張，他後來寫信來和我說對不起，覺得宏碁大跌是他引發的，有一些人就在怪罪別人賣了1張，2張拿錢出來買產品，所以引發宏碁大跌，也有人怪罪有人高檔調節，所以引發宏碁大跌…等，說真的，那就是"代表性原則錯誤"，市場上會發生的事就是會發生，這周宏碁的成交量是974304張，意思就是這周每一天宏碁成交量是20萬張，你沒有看錯就是20萬張，區區的幾百，幾仟，一兩萬張，你以為火車頭真的是被狗吠走的，會因為別人做了一些調整後，覺得宏碁就是被那些人搞下來的，無知的程度可比古代人處理天狗食月的想法，"老百姓見月亮沒瞭，就拿起銅鑼、盆子、鍋蓋，紛紛跑出屋子，邊敲邊喊：天狗吃月瞭，快打快打呀 打死它，打死它，咱們要月亮！"
要看出市場怎麼走，你要研究、觀察、記錄出真實的情況，最終自已心理有所準備，難道我該睜著眼說瞎話的去記錄我看到的東西，這文章是我未來自已要看的，為什麼我要睜著眼說瞎話，記錄下錯的東西，單純要讓大家開心，這樣未來我回頭自已看的時候，看到的就是沒有意義的東西。如果覺得我寫的不喜歡，自已可以去研究，觀察，記錄嘛，然後一周後回頭看看，是不是和你推測的一樣，反正股市中我們有的是時間，一周開盤五次，自已慢慢驗證，如果自已寫了老半天，一周回頭看，和你自已寫的完全是不符合的，走勢和自已推測的也差很多時，你就會知道誰研究，記錄、觀察的才是對的，這不是很公平嗎？
之前宏碁在上漲時，1~10張的打死都不會增加，現在主力拉了一個錨點33.75元然後跌下來到27.6元，這些散戶突然間就覺得宏碁好便宜啊，
之前同事在20，22時一直求我賣，他說你賣個3張也好啦，都從14漲到22了，都漲8元了，不可能這樣一直漲上去啦，我沒有理他，結果股價從22.x跌到20.x後，他就說，你看吧，跌了吧，股價漲上去你沒有賣都是假的啦，我也沒有理會他。直到宏碁一直漲，狂漲到33.75元後，跌到30.x元，他說他要買了，我聽了差點沒有昏倒，為什麼22你叫我賣，30.x你卻要買？宏碁基本面有改變嗎？似乎沒有，所以他這症狀就是大部散戶的症狀，中了錨定效和加一點後悔效應，覺得宏碁從14漲到33.75一路漲沒賺到，心理一直一直很恨，結果33.75跌到30，甚至是27.3元，他們就會覺得這是從天上掉下來的禮物，趕快加碼買進。
這周1~10張的散戶狂增9113人，買了股票24990張，算是非理性的買法，這就是成交量很大所會造成的現象，不要天真的以為什麼10萬張籌定籌碼可以避免這周從33.75跌到27.6,我說過，這叫做狗犬火車頭，不要以為火車是被你吠走的，這是市場的趨勢，必然會發生的事，這人性弱點叫"代表性原則錯誤"，常常看我的文章，或聽到某個人說賣出，就以為整件事是因他而發生的，之前ptt有一個人賣了800張，結果ptt就一片拜神說，原來宏碁的大跌是他引發的。一位宏碁股東賣出了9xx張，他後來寫信來和我說對不起，覺得宏碁大跌是他引發的，有一些人就在怪罪別人賣了1張，2張拿錢出來買產品，所以引發宏碁大跌，也有人怪罪有人高檔調節，所以引發宏碁大跌…等，說真的，那就是"代表性原則錯誤"，市場上會發生的事就是會發生，這周宏碁的成交量是974304張，意思就是這周每一天宏碁成交量是20萬張，你沒有看錯就是20萬張，區區的幾百，幾仟，一兩萬張，你以為火車頭真的是被狗吠走的，會因為別人做了一些調整後，覺得宏碁就是被那些人搞下來的，無知的程度可比古代人處理天狗食月的想法，"老百姓見月亮沒瞭，就拿起銅鑼、盆子、鍋蓋，紛紛跑出屋子，邊敲邊喊：天狗吃月瞭，快打快打呀 打死它，打死它，咱們要月亮！"
要看出市場怎麼走，你要研究、觀察、記錄出真實的情況，最終自已心理有所準備，難道我該睜著眼說瞎話的去記錄我看到的東西，這文章是我未來自已要看的，為什麼我要睜著眼說瞎話，記錄下錯的東西，單純要讓大家開心，這樣未來我回頭自已看的時候，看到的就是沒有意義的東西。如果覺得我寫的不喜歡，自已可以去研究，觀察，記錄嘛，然後一周後回頭看看，是不是和你推測的一樣，反正股市中我們有的是時間，一周開盤五次，自已慢慢驗證，如果自已寫了老半天，一周回頭看，和你自已寫的完全是不符合的，走勢和自已推測的也差很多時，你就會知道誰研究，記錄、觀察的才是對的，這不是很公平嗎？
ffmpeg
https://sonnati.wordpress.com/2011/08/08/ffmpeg-%E2%80%93-the-swiss-army-knife-of-internet-streaming-%E2%80%93-part-ii/
python 3 pandas matplotlib.pyplot
寫程式這麼久，python真的有夠難寫的，網路上一堆範例都有語法上的錯誤或不完整的，後來經由大陸某教學網終於成功
https://morvanzhou.github.io/tutorials/data-manipulation/np-pd/3-8-pd-plot/
1、donwload python3 from python website
2、create aaa.py put context
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
ts = pd.Series(np.random.randn(1000), index=pd.date_range('1/1/2000', periods=1000))
ts = ts.cumsum()
print(ts)
ts.plot()
plt.show() //This is best important!!
3、python aaa.py then see error. Usually is package not install. So just pip install
import matplotlib.pyplot as plt # pip install matplotlib
import pandas as pd # pip install pandas
https://morvanzhou.github.io/tutorials/data-manipulation/np-pd/3-8-pd-plot/
1、donwload python3 from python website
2、create aaa.py put context
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
ts = pd.Series(np.random.randn(1000), index=pd.date_range('1/1/2000', periods=1000))
ts = ts.cumsum()
print(ts)
ts.plot()
plt.show() //This is best important!!
3、python aaa.py then see error. Usually is package not install. So just pip install
import matplotlib.pyplot as plt # pip install matplotlib
import pandas as pd # pip install pandas
see screen have notis, just follow install.
windows docker go
1、go
mkdir directory aaa
create file go_httpserver.go
package main
import (
"fmt"
"net/http"
)
func handler(w http.ResponseWriter, r *http.Request) {
fmt.Fprintf(w, "Welcome to my Website!!!\n %s", r.URL.Path[1:])
}
func main() {
http.HandleFunc("/", handler)
http.ListenAndServe("0.0.0.0:80", nil)
}
set GOARCH=amd64
FROM scratch
ADD go_httpserver /
ENTRYPOINT ["/go_httpserver"]
mkdir directory aaa
create file go_httpserver.go
package main
import (
"fmt"
"net/http"
)
func handler(w http.ResponseWriter, r *http.Request) {
fmt.Fprintf(w, "Welcome to my Website!!!\n %s", r.URL.Path[1:])
}
func main() {
http.HandleFunc("/", handler)
http.ListenAndServe("0.0.0.0:80", nil)
}
set GOOS=linux
go build go_httpserver.go => get linux go program
2、create docker image
2.1 install docker toolbox Link
2.2 run kitematic then see left-down DOCKER CLI , click it, then
2.2.1 create directory bbb
2.2.2 create Dockerfile
ADD go_httpserver /
ENTRYPOINT ["/go_httpserver"]
docker buid -t testdockergoweb:v1 .
2.2.3 use kitematic left-up +NEW -> My Images -> CREATE just crate image
2.2.4 now Containers list have testdockergoweb running. see right-up Settings
-> Hostname / Ports -> change
DOCKER PORT 80
PUBLISHED IP:PORT port->3111
SAVE
then
click PUBLISHED IP:PORT list have bule link.
3、export docker image
in DOCKER CLI -> docker export testdockergoweb > testdockergoweb.tar
眼光費
http://stockresearchsociety.blogspot.tw/2017/12/1211414.html
- 做短線、想做差價前，就下虛擬單 你寫下來進入點， 要出時寫下賣出點，十次實驗，對的機率有7成以上，可以做差價
- 如果你的公司值20元，一年能賺3元，目前股價60元，你能說，一年賺3元，本益比20倍，所以60元很便宜，說這種話，我聽了都覺得很好笑。一年能賺3元，公司值20元，你要年年賺3元賺13.x年才賺到到60元，結果你竟然在一季賺了0.75元，你就開始幻想這樣的賺法13年後你就回本。
- 巴非特之所以不敗，原因很簡單，公司值20元，他在股價打7折、8折時，14~16元然後看出這家公司體質改善了，未來會賺錢，走上正確的方向機率很大，巴非特就大力的買在那個價格，然後用時間去換報酬，這樣的投資才是最安全的，沒有被一些人拿走未來15年，20年的獲利當眼光費。
- 投資為什麼常常會賠錢？因為你沒有估價的能力，也沒有耐心，也不會為你的投資做長久的設想，所以很容易一時亂了手腳
- 眼光費
加回匯率因素，XXX淨值約19.7元，目前看一年能賺1~1.2元，我20投資XXX，我等一年，XXX賺1~1.2元，我等兩年XXX賺2~2.4元，兩年後，三年後XXX獲利穩定後，一般世俗的眼光費套在XXX上，大家再選一個氣氛很好，如財報很好或是大盤氣氛很好的時間點，隨隨便便也是一個很大的報酬率，除了本身公司獲利你穩賺的股息外，總是能等到別人給你10~15年獲利當眼光費的一天，一等到你就賺了10~15年的獲利了嗎？要最大化你在XXX最大的報酬，絕對不會是20，21，21元去賣掉覺得自已賺好多，你要最大化XXX的獲利，一定是在一年後，兩年後的某一天，而那一天只要XXX持續優化的情況下，一定會到來，那時才會是你得到最大報酬率的時候
Azure function
headers: {
'Content-Type': 'application/json'
},
body: {
ip: (
req.headers['x-forwarded-for'].split(":")[0]
//req.headers['x-forwarded-for'] //||
//req.connection.remoteAddress ||
//req.socket.remoteAddress //||
//req.connection.socket.remoteAddress
)
}
[轉]IMF GDP
https://www.youtube.com/watch?v=4ExDkvchC6Q 3:06開始
GDP2萬多，但實質購買力是5萬
有錢人存了很多錢
國外被掏空，經濟競爭力沒了，老闆不想產業升級
GDP2萬多，但實質購買力是5萬
有錢人存了很多錢
國外被掏空，經濟競爭力沒了，老闆不想產業升級
[轉]第一次在 GCP Storage 放置靜態網頁就上手
https://coder.tw/?p=7626
1. install gcloud
2. gcloud auth login `email` #login auth email, get rights for change 公開
2.1 gcloud auth revoke `email` #remove auth email
1. install gcloud
2. gcloud auth login `email` #login auth email, get rights for change 公開
2.1 gcloud auth revoke `email` #remove auth email
https://cloud.google.com/sdk/gcloud/reference/auth/login
設定 Bucket 新上傳的檔案預設公開，只能針對 Bucket 設定，不朔及既往
設定某資料夾（or Bucket）底下所有檔案、目錄為公開
設定 Bucket 新上傳的檔案預設公開，只能針對 Bucket 設定，不朔及既往
gsutil defacl set public-read gs://s.ntust.me
設定某資料夾（or Bucket）底下所有檔案、目錄為公開
gsutil acl set -r public-read gs://s.ntust.me/iatp
股
- 每季報告
- 第四季：因為明年3月才會公布財報，所以第三季公佈後到明年3月，近半年時間，是沒有任何依據，第3季公佈成績好，自然可以撐到明年3月，公佈成績不好，也是同樣不好到明年3月，除非有特殊情況，否則沒人看第四季的。
- 外資
- 在大洗盤，把散戶洗走：外資看好這支股票，為了增加獲利，壓低股價，嚇散戶自願出場，買下低成本股票，增加持有股數，等待最終獲利。
- 這段期間不可能讓成本上升，就算上升也會想盡辨法壓低，如果壓不下來，外資持有成本也同樣壓不下來，獲利就不能提高。
- 一般散戶受不了壓力，就直接出場，因為這是長時間，非短期操作。
- 要補單要在大盤情況也不好大跌跟補單
nginx proxy pass [ best practices ] 2 for sysctl tcp
．https://blogs.dropbox.com/tech/2017/09/optimizing-web-servers-for-high-throughput-and-low-latency/
http://www.queryadmin.com/1654/tuning-linux-kernel-tcp-parameters-sysctl/
https://blogs.dropbox.com/tech/2017/09/optimizing-web-servers-for-high-throughput-and-low-latency/
Don't USE
net.ipv4.tcp_tw_recycle=1
—don’t use it—it was already broken for users behind NAT, but if you upgrade your kernel, it will be broken for everyone.
net.ipv4.tcp_timestamps=0
—don’t disable them unless you know all side-effects and you are OK with them. For example, one of non-obvious side effects is that you will loose window scaling and SACK options on syncookies.
https://read01.com/zh-tw/KBgmj7.html
Don't USE
net.ipv4.tcp_tw_recycle=1
net.ipv4.tcp_tw_reuse=1 /*Only you know, sometime can do*/
***********sysctl**************
net.ipv4.tcp_mtu_probing=1
# Increase number of max open-files
fs.file-max = 150000
# Increase max number of PIDs
kernel.pid_max = 4194303
# Increase range of ports that can be used
net.ipv4.ip_local_port_range = 1024 65535
# https://tweaked.io/guide/kernel/
# Forking servers, like PostgreSQL or Apache, scale to much higher levels of concurrent connections if this is made larger
kernel.sched_migration_cost_ns=5000000
# https://tweaked.io/guide/kernel/
# Various PostgreSQL users have reported (on the postgresql performance mailing list) gains up to 30% on highly concurrent workloads on multi-core systems
kernel.sched_autogroup_enabled = 0
# https://github.com/ton31337/tools/wiki/tcp_slow_start_after_idle---tcp_no_metrics_save-performance
# Avoid falling back to slow start after a connection goes idle
net.ipv4.tcp_slow_start_after_idle=0
net.ipv4.tcp_no_metrics_save=0
# https://github.com/ton31337/tools/wiki/Is-net.ipv4.tcp_abort_on_overflow-good-or-not%3F
net.ipv4.tcp_abort_on_overflow=0
# Enable TCP window scaling (enabled by default)
# https://en.wikipedia.org/wiki/TCP_window_scale_option
net.ipv4.tcp_window_scaling=1
# Enables fast recycling of TIME_WAIT sockets.
# (Use with caution according to the kernel documentation!)
#net.ipv4.tcp_tw_recycle = 1
# Allow reuse of sockets in TIME_WAIT state for new connections
# only when it is safe from the network stack’s perspective.
#net.ipv4.tcp_tw_reuse = 1
# Turn on SYN-flood protections
net.ipv4.tcp_syncookies=1
# Only retry creating TCP connections twice
# Minimize the time it takes for a connection attempt to fail
net.ipv4.tcp_syn_retries=2
net.ipv4.tcp_synack_retries=2
net.ipv4.tcp_orphan_retries=2
# How many retries TCP makes on data segments (default 15)
# Some guides suggest to reduce this value
net.ipv4.tcp_retries2=8
# Optimize connection queues
# https://www.linode.com/docs/web-servers/nginx/configure-nginx-for-optimized-performance
# Increase the number of packets that can be queued
net.core.netdev_max_backlog = 3240000
# Max number of "backlogged sockets" (connection requests that can be queued for any given listening socket)
net.core.somaxconn = 50000
# Increase max number of sockets allowed in TIME_WAIT
net.ipv4.tcp_max_tw_buckets = 1440000
# Number of packets to keep in the backlog before the kernel starts dropping them
# A sane value is net.ipv4.tcp_max_syn_backlog = 3240000
net.ipv4.tcp_max_syn_backlog = 3240000
# TCP memory tuning
# View memory TCP actually uses with: cat /proc/net/sockstat
# *** These values are auto-created based on your server specs ***
# *** Edit these parameters with caution because they will use more RAM ***
# Changes suggested by IBM on https://www.ibm.com/developerworks/community/wikis/home?lang=en#!/wiki/Welcome%20to%20High%20Performance%20Computing%20%28HPC%29%20Central/page/Linux%20System%20Tuning%20Recommendations
# Increase the default socket buffer read size (rmem_default) and write size (wmem_default)
# *** Maybe recommended only for high-RAM servers? ***
net.core.rmem_default=16777216
net.core.wmem_default=16777216
# Increase the max socket buffer size (optmem_max), max socket buffer read size (rmem_max), max socket buffer write size (wmem_max)
# 16MB per socket - which sounds like a lot, but will virtually never consume that much
# rmem_max over-rides tcp_rmem param, wmem_max over-rides tcp_wmem param and optmem_max over-rides tcp_mem param
net.core.optmem_max=16777216
net.core.rmem_max=16777216
net.core.wmem_max=16777216
# Configure the Min, Pressure, Max values (units are in page size)
# Useful mostly for very high-traffic websites that have a lot of RAM
# Consider that we already set the *_max values to 16777216
# So you may eventually comment these three lines
net.ipv4.tcp_mem=16777216 16777216 16777216
net.ipv4.tcp_wmem=4096 87380 16777216
net.ipv4.tcp_rmem=4096 87380 16777216
# Keepalive optimizations
# By default, the keepalive routines wait for two hours (7200 secs) before sending the first keepalive probe,
# and then resend it every 75 seconds. If no ACK response is received for 9 consecutive times, the connection is marked as broken.
# The default values are: tcp_keepalive_time = 7200, tcp_keepalive_intvl = 75, tcp_keepalive_probes = 9
# We would decrease the default values for tcp_keepalive_* params as follow:
net.ipv4.tcp_keepalive_time = 600
net.ipv4.tcp_keepalive_intvl = 10
net.ipv4.tcp_keepalive_probes = 9
# The TCP FIN timeout belays the amount of time a port must be inactive before it can reused for another connection.
# The default is often 60 seconds, but can normally be safely reduced to 30 or even 15 seconds
# https://www.linode.com/docs/web-servers/nginx/configure-nginx-for-optimized-performance
net.ipv4.tcp_fin_timeout = 7
***********sysctl**************
==PS==
．net.ipv4.tcp_slow_start_after_idle & net.ipv4.tcp_no_metrics_save
https://github.com/ton31337/tools/wiki/tcp_slow_start_after_idle---tcp_no_metrics_save-performance
http://www.queryadmin.com/1654/tuning-linux-kernel-tcp-parameters-sysctl/
https://blogs.dropbox.com/tech/2017/09/optimizing-web-servers-for-high-throughput-and-low-latency/
Don't USE
net.ipv4.tcp_tw_recycle=1
—don’t use it—it was already broken for users behind NAT, but if you upgrade your kernel, it will be broken for everyone.
net.ipv4.tcp_timestamps=0
—don’t disable them unless you know all side-effects and you are OK with them. For example, one of non-obvious side effects is that you will loose window scaling and SACK options on syncookies.
https://read01.com/zh-tw/KBgmj7.html
Don't USE
net.ipv4.tcp_tw_recycle=1
net.ipv4.tcp_tw_reuse=1 /*Only you know, sometime can do*/
***********sysctl**************
net.ipv4.tcp_mtu_probing=1
# Increase number of max open-files
fs.file-max = 150000
# Increase max number of PIDs
kernel.pid_max = 4194303
# Increase range of ports that can be used
net.ipv4.ip_local_port_range = 1024 65535
# https://tweaked.io/guide/kernel/
# Forking servers, like PostgreSQL or Apache, scale to much higher levels of concurrent connections if this is made larger
kernel.sched_migration_cost_ns=5000000
# https://tweaked.io/guide/kernel/
# Various PostgreSQL users have reported (on the postgresql performance mailing list) gains up to 30% on highly concurrent workloads on multi-core systems
kernel.sched_autogroup_enabled = 0
# https://github.com/ton31337/tools/wiki/tcp_slow_start_after_idle---tcp_no_metrics_save-performance
# Avoid falling back to slow start after a connection goes idle
net.ipv4.tcp_slow_start_after_idle=0
net.ipv4.tcp_no_metrics_save=0
# https://github.com/ton31337/tools/wiki/Is-net.ipv4.tcp_abort_on_overflow-good-or-not%3F
net.ipv4.tcp_abort_on_overflow=0
# Enable TCP window scaling (enabled by default)
# https://en.wikipedia.org/wiki/TCP_window_scale_option
net.ipv4.tcp_window_scaling=1
# Enables fast recycling of TIME_WAIT sockets.
# (Use with caution according to the kernel documentation!)
#net.ipv4.tcp_tw_recycle = 1
# Allow reuse of sockets in TIME_WAIT state for new connections
# only when it is safe from the network stack’s perspective.
#net.ipv4.tcp_tw_reuse = 1
# Turn on SYN-flood protections
net.ipv4.tcp_syncookies=1
# Only retry creating TCP connections twice
# Minimize the time it takes for a connection attempt to fail
net.ipv4.tcp_syn_retries=2
net.ipv4.tcp_synack_retries=2
net.ipv4.tcp_orphan_retries=2
# How many retries TCP makes on data segments (default 15)
# Some guides suggest to reduce this value
net.ipv4.tcp_retries2=8
# Optimize connection queues
# https://www.linode.com/docs/web-servers/nginx/configure-nginx-for-optimized-performance
# Increase the number of packets that can be queued
net.core.netdev_max_backlog = 3240000
# Max number of "backlogged sockets" (connection requests that can be queued for any given listening socket)
net.core.somaxconn = 50000
# Increase max number of sockets allowed in TIME_WAIT
net.ipv4.tcp_max_tw_buckets = 1440000
# Number of packets to keep in the backlog before the kernel starts dropping them
# A sane value is net.ipv4.tcp_max_syn_backlog = 3240000
net.ipv4.tcp_max_syn_backlog = 3240000
# TCP memory tuning
# View memory TCP actually uses with: cat /proc/net/sockstat
# *** These values are auto-created based on your server specs ***
# *** Edit these parameters with caution because they will use more RAM ***
# Changes suggested by IBM on https://www.ibm.com/developerworks/community/wikis/home?lang=en#!/wiki/Welcome%20to%20High%20Performance%20Computing%20%28HPC%29%20Central/page/Linux%20System%20Tuning%20Recommendations
# Increase the default socket buffer read size (rmem_default) and write size (wmem_default)
# *** Maybe recommended only for high-RAM servers? ***
net.core.rmem_default=16777216
net.core.wmem_default=16777216
# Increase the max socket buffer size (optmem_max), max socket buffer read size (rmem_max), max socket buffer write size (wmem_max)
# 16MB per socket - which sounds like a lot, but will virtually never consume that much
# rmem_max over-rides tcp_rmem param, wmem_max over-rides tcp_wmem param and optmem_max over-rides tcp_mem param
net.core.optmem_max=16777216
net.core.rmem_max=16777216
net.core.wmem_max=16777216
# Configure the Min, Pressure, Max values (units are in page size)
# Useful mostly for very high-traffic websites that have a lot of RAM
# Consider that we already set the *_max values to 16777216
# So you may eventually comment these three lines
net.ipv4.tcp_mem=16777216 16777216 16777216
net.ipv4.tcp_wmem=4096 87380 16777216
net.ipv4.tcp_rmem=4096 87380 16777216
# Keepalive optimizations
# By default, the keepalive routines wait for two hours (7200 secs) before sending the first keepalive probe,
# and then resend it every 75 seconds. If no ACK response is received for 9 consecutive times, the connection is marked as broken.
# The default values are: tcp_keepalive_time = 7200, tcp_keepalive_intvl = 75, tcp_keepalive_probes = 9
# We would decrease the default values for tcp_keepalive_* params as follow:
net.ipv4.tcp_keepalive_time = 600
net.ipv4.tcp_keepalive_intvl = 10
net.ipv4.tcp_keepalive_probes = 9
# The TCP FIN timeout belays the amount of time a port must be inactive before it can reused for another connection.
# The default is often 60 seconds, but can normally be safely reduced to 30 or even 15 seconds
# https://www.linode.com/docs/web-servers/nginx/configure-nginx-for-optimized-performance
net.ipv4.tcp_fin_timeout = 7
***********sysctl**************
==PS==
．net.ipv4.tcp_slow_start_after_idle & net.ipv4.tcp_no_metrics_save
https://github.com/ton31337/tools/wiki/tcp_slow_start_after_idle---tcp_no_metrics_save-performance
訂閱：
文章 (Atom)