G***G
发帖数: 16778 | 1 your experiments after two days are biological replicate.
five replicates at the same times are technical replicate.
最近做luciferase assay，就是把含有疾病相关的snp的promoter跟luciferase连接在
一起，看snp对promoter的影响。假设snp的ref allele是A, 疾病相关的allele是B
我首先在hek293里做，用lipofectamine3000转染。铺板子（12-well）铺了10个孔，5
个用于allele-a, 5个用于allele－b。先把lipo和dna混合好，然后把mix分给每一个孔。
以上是第一次实验，过两天我会一模一样的重复上述实验。
我想请问这里的5个孔中的每一个孔算是biological replicate还是technical
replicate呢？
而这两次实验算是biological replicate还是technical replicate呢？
之所以问这个问题是因为后来面对转染neur... 阅读全帖 |
|
z***y
发帖数: 7151 | 2 First of all, only replicating changes from a specific user is doable.
Though replication is not your first choice. There are a few COTS can do
this. Like Golden Gate and Attunity.
Second, replication is not trigger. Replication uses agent to transfer data.
Replication coupled tightly within other SQL SERVER executables and DLLs.
Trigger is schema level concept.
Last, if you really want to use replication to replicate changes from one
specific user. You can consider the using filters. You had... 阅读全帖 |
|
发帖数: 1 | 3 最近做luciferase assay，就是把含有疾病相关的snp的promoter跟luciferase连接在
一起，看snp对promoter的影响。假设snp的ref allele是A, 疾病相关的allele是B
我首先在hek293里做，用lipofectamine3000转染。铺板子（12-well）铺了10个孔，5
个用于allele-a, 5个用于allele－b。先把lipo和dna混合好，然后把mix分给每一个孔。
以上是第一次实验，过两天我会一模一样的重复上述实验。
我想请问这里的5个孔中的每一个孔算是biological replicate还是technical
replicate呢？
而这两次实验算是biological replicate还是technical replicate呢？
之所以问这个问题是因为后来面对转染neuron的实验。neuron比较难转染所以用的是
nucleofection，一次性用cuvette转染100万个细胞，然后分到12-well的3个孔中。
同样的实验过两天我也会一模一样的再重复一次。
对于neuron的转染，分到3个孔中的... 阅读全帖 |
|
s****l
发帖数: 600 | 4 【 以下文字转载自 Database 讨论区 】
发信人: sugirl (Xixi), 信区: Database
标 题: 请问MySQL的replication不通过应用程序能达到strong consistence吗？
发信站: BBS 未名空间站 (Tue Jun 2 01:55:30 2015, 美东)
MySQL cross data center replication，现在需要在一个master写，而slave或另一个
master马上（一秒以内）读出写入的值。请问如何做到？我听说mysql的replication不
是strong consistent，也就是replication是async，需要一定的backlog才replicate
到其他data center。 |
|
z*********8
发帖数: 1203 | 5 请问在DESeq里面怎么对partially replicated的sample分析啊，我有4个组，2组有
replicates，2组没有，建完dataframe后，可以直接对没有replicate的组分析，用“
blind“，也可以对有replicates的组分析，现在我想放在一起分析，人家就报error
>cds=estimateDispersions(cds)
Error in .local(object, ...) :
None of your conditions is replicated. Use method='blind' to estimate across
conditions, or 'pooled-CR', if you have crossed factors.
请问用什么command实现啊？我看了下manual，没有看懂啊！非常感谢 |
|
Z******5
发帖数: 435 | 6 我觉得对你这个实验来说，在293细胞中做的只有technical replicate，没有
biological replicate，实验本身也不需要biological replicate。
在neuron中做的话，如果只有一个来源的细胞系，也是只需要technical replicate即
可；如果在不同细胞系或者原代细胞中重复的话，才叫biological replicate吧？ |
|
C***m
发帖数: 120 | 7 1. I prefer static replication to dynamic replication, as long as it is
possible. If one cannot perfectly replicate the final payoff using simple
static replication, then go dynamic.
2. (a) market liquidity,可能找不到足够多的量来复制，比如ATM digital option
（b）dynamic 只能replicate 一些greeks. 比如delta hedge,可能会有gamma risk.
(c) dynamic 多少需要模型来决定。模型的risk
胡说几句，大牛们轻拍。
that
the |
|
b******d
发帖数: 115 | 8 I have a question about the replication method with SQL7 or
SQL2000 (I'm using SQL2000 actually). I know how to set up
publication for replicating certain tables, but need to
figure out how to capture the content of a publication and
send a message to an MSMQ queue. The default for
transactional replication in SQL server is to replicate the
change to a subscriber's table. My guess is during the
process of getting the content from publisher to subscriber,
there's got to be a place (maybe internal |
|
S*****0
发帖数: 538 | 9 I have not used merge replication so far yet, but here is what I thought,
from the transactional replication standpoint, about your scenario, which
the publisher and subscriber are in different status.
1. Disable the Snapshot. During the replication configuration, you can
delay the snapshot, then even disable the snapshot job, so the snapshot will
never happen for this replication. This way, you can keep them in its own
status.
2. Another way is to use snapshot first, get both publisher a |
|
s****l
发帖数: 600 | 10 MySQL cross data center replication，现在需要在一个master写，而slave或另一个
master马上（一秒以内）读出写入的值。请问如何做到？我听说mysql的replication不
是strong consistent，也就是replication是async，需要一定的backlog才replicate
到其他data center。 |
|
发帖数: 1 | 11 技术重复是看你操作有没有失误；生物重复是看你实验本身（这里是细胞）有没有有统
计意义的差异。
[在 flareon () 的大作中提到：]
:最近做luciferase assay，就是把含有疾病相关的snp的promoter跟luciferase连接在
:一起，看snp对promoter的影响。假设snp的ref allele是A, 疾病相关的allele是B
:我首先在hek293里做，用lipofectamine3000转染。铺板子（12-well）铺了10个孔，5
:个用于allele-a, 5个用于allele－b。先把lipo和dna混合好，然后把mix分给每一个
孔。
:以上是第一次实验，过两天我会一模一样的重复上述实验。
:我想请问这里的5个孔中的每一个孔算是biological replicate还是technical
:replicate呢？
:而这两次实验算是biological replicate还是technical replicate呢？
:之所以问这个问题是因为后来面对转染neuron的实验。neuron比较难转染所以用的是
:nucleofection... 阅读全帖 |
|
l********n
发帖数: 200 | 12 公司准备在另一个州搞几个服务器分流网站流量，网站数据库是MSSQL 2005
Enterprise，所以原计划用MSSQL Replication。但现在又觉得MSSQL太贵不买了，想找
一种便宜的替代，比如说用MySQL换掉MSSQL，但MySQL Replication好像只有Master-
Slave模式，不过网上也有人捉磨出两个方向的。
显然全部换成MySQL，还要捉磨Replication，都是比较麻烦的。
不知各位有没有什么比较好的办法？软件硬件都可以，只要不是很贵。
谢谢！ |
|
a*******t
发帖数: 891 | 13 We want to out source our .com server hosting, and need to upload data from
our database server to .com database server nightly.
Possible methods:
Transactional replication (log shipping) - problem is if .com database needs
to be offline for server maintenance, or they have to switch to a "staging"
server, log shipping is broken and we'll have to reset the log shipping.
Merge replication - it's a two way sync. we don't want that.
Snapshot replication - seems the most promissing. the database is |
|
l*****b
发帖数: 82 | 14 Hi, friends,
In SQL server 2005, I have case to build new DB in another location and have replication with
existed DB. They have same schema. My approach is:
1. backup the existed DB, restore to new DB
2. delete some old data in new DB - existed DB keeps 10 years data, new DB
maintain the recent 3 year data.
3. make two DB replication in sync - bi-direction transactional or merge
type
Does this work in not-identical data DB replication?
Thank you |
|
g***l
发帖数: 18555 | 15 昨天加了13个TABLE UPDATABLE TRANSACTIONAL REPLICATION，COMPILES/S就从昨天的
20跳到50了，CPU的USAGE也明显增加，到底什么在RECOMPILE，难道是REPLICATION自己
的SP?还是REPLICATION CALL的那些TRIGGERS |
|
s**********o
发帖数: 14359 | 16 显然你对REPLICATION理解不够，REPLICATION就是原封不动的把目标DATABASE的TABLE
做一个备份而已，你选择性的备份就不是REPLICATION了，很快就OUT OF SYNC.你自己
写TRIGGER好了 |
|
c******n
发帖数: 7263 | 17 显然不是啊，sqlserver 可以设filter的，可以选一些特定的row或者column的
[在 smallburrito (smallburrito) 的大作中提到：]
：显然你对REPLICATION理解不够，REPLICATION就是原封不动的把目标DATABASE的
TABLE
：做一个备份而已，你选择性的备份就不是REPLICATION了，很快就OUT OF SYNC.你自己
：........... |
|
s**********o
发帖数: 14359 | 18 显然你对REPLICATION理解不够，REPLICATION就是原封不动的把目标DATABASE的TABLE
做一个备份而已，你选择性的备份就不是REPLICATION了，很快就OUT OF SYNC.你自己
写TRIGGER好了 |
|
c******n
发帖数: 7263 | 19 显然不是啊，sqlserver 可以设filter的，可以选一些特定的row或者column的
[在 smallburrito (smallburrito) 的大作中提到：]
：显然你对REPLICATION理解不够，REPLICATION就是原封不动的把目标DATABASE的
TABLE
：做一个备份而已，你选择性的备份就不是REPLICATION了，很快就OUT OF SYNC.你自己
：........... |
|
s*r
发帖数: 2757 | 20 http://www.nature.com/news/reproducibility-the-risks-of-the-rep
Isn’t reproducibility the bedrock of the scientific process? Yes, up to a
point. But it is sometimes much easier not to replicate than to replicate
studies, because the techniques and reagents are sophisticated, time-
consuming and difficult to master. In the past ten years, every paper
published on which I have been senior author has taken between four and six
years to complete, and at times much longer. People in my lab often need... 阅读全帖 |
|
m****d
发帖数: 331 | 21 risk-neutral probabilities: q_i,
real-world probabilities: p_i
let c=dQ/dP which is a vector with entires c_i=q_i/p_i
Let x be the portfolio that replicates c.
what does "Let x be the portfolio that replicates c." mean?
anything related to market portfolio?
I know that if a portfolio can be replicated, it means that Ax=y can be
solved, where A is the payoff matrix. Thanks. |
|
l*******3
发帖数: 186 | 22 question:
1. given the choice between two different static replication portfolios
that
match an option's payoff, what criteria would you use to decide between
the
two?
2. what are some of the practical problems of dynamic replication?
transaction cost (so we cannot replicate too frequently).
any other concerns?
Thank you! |
|
C******g
发帖数: 2930 | 23 Square has not one, but two Nier games in the works: one for Xbox, and one
for PS3. Nier Replicant is being developed exclusively for PS3, and looks
somewhat different from its Gestalt counterpart. Both games feature a white-
haired protagonist named Nier, but Replicant's lead is a younger, slimmer
and a bit more effeminate than the 360's lead character. Both stories have
our hero(es) on a journey to save a family member: in Replicant, it's a
sister, while in Gestalt, it's a daughter.
We're not |
|
a***h
发帖数: 29 | 24 merge replication: error 14010, after the step "bulk copied
data into table 'sysmergesubsetfilter'"
i setup merge replication via RAS. i encounter some
problems.
Error 14010:
Last action: The subscription to publication "t1" is invalid
Last command: {call sp_MSgetreplicainfo(?,?,?,?,?,?,?)}
Error information:
Catagory: Data source
Source: 192.168.22.3
Number: 14010
Full error: The remote server is not defined as a
subscription server.
i always encounter error 14010 when i try to setup merge
repl |
|
l********n
发帖数: 200 | 25 Standard Edition 是不是只有Pubisher-Subscriber 单方向的？
我们原计划采用Peer to Peer Replication，所以得买Enterprise。
据说有很多软件提供 Database Replication，不知这里是否有人用过？ |
|
l*****b
发帖数: 82 | 26 Hi, friends,
I will make some table replications between two DBs with same structure. The
main problem so far is how to prevent the PK conflict when replicating as
the current PKs are auto increased by sequence. My first thinking is the
have composite PK keys: instance id + auto increased, e.g. 01 and 1234 as a
PK pair, 01 is instance id and 1234 is current increased PK. I heard there
are some drawback in composite PK key, e.g. in ORM, etc. However, I am not
using ORM so far. Is my solution ok? |
|
l*****b
发帖数: 82 | 27 I would like to make replication between two tables in different location
servers. But their existed data are not the same. Could I still make these
two tables replicated?
Thanks |
|
g***l
发帖数: 18555 | 28 看哪个SERVER忙吧，DISTRIBUTOR尽量放在不忙的SERVER上，但如果NETWORK SPEED不行
的话，还是放PUBLISHER上。STANDARD 2005-2008是可以做REPLICATION的.你这个CASE比较简单，做个TRASACTIONAL REPLICATION就可以了，注意TABLE都要有KEY。 |
|
s********e
发帖数: 78 | 29 Urgent, need everyone’s help, any feedback will be very much appreciated.
SQL 2008 R2 Replication VS Mirroring Or other technology
Client need push data from each satellite server(10 regional server) to a
single repository in real time.
Requirement
1. Data transfer cross internet, low latency.(data volume low, shouldn’t
be a problem)
2. Each satellite server has only one db with identical name/layout, so
eventually the result should looks like
SourceServer SourceDB TargetServer ... 阅读全帖 |
|
s**********o
发帖数: 14359 | 30 打个比喻吧，REPLICATION就好比是复印机，从SERVER A复印到SERVER B
比如你有一篇打印好的论文，做个REPLICATION，拆了订书钉，扔到复印机里，
自动复印一份，但你要说，我只要复印引用的短路，这就难了，你自己
慢慢去一段一段的印吧 |
|
s**********o
发帖数: 14359 | 31 打个比喻吧，REPLICATION就好比是复印机，从SERVER A复印到SERVER B
比如你有一篇打印好的论文，做个REPLICATION，拆了订书钉，扔到复印机里，
自动复印一份，但你要说，我只要复印引用的短路，这就难了，你自己
慢慢去一段一段的印吧 |
|
w***w
发帖数: 1552 | 32 【 以下文字转载自 shopping 讨论区 】
发信人: wxwxw (kk), 信区: shopping
标 题: Lenovo outlet的那个ThinkPad USB Port Replicator with Digita
发信站: BBS 未名空间站 (Thu Jun 24 14:04:56 2010, 美东)
ThinkPad USB Port Replicator with Digital Video 这个性能怎么样？好像就只是通
过usb口连接，不知道性能怎么样。
考虑进一个 |
|
a********2
发帖数: 2561 | 33 我觉得是尺寸不同吧。port replicator 跟 ultrabase 从概念上说都是 dock。port
replicator 小，只能当成接口扩展用，多几个USB啊，串口、并口、显示器输出口啥的
。ultrabase个头大，最大的不同是一般都带个光驱，接口也可能更多些。
具体接口得看spec。
要接27寸显示器，只要有DVI，hdmi，displayport 之一就行。VGA口带宽不够，推不动
。 |
|
w**z
发帖数: 8232 | 34 slave replicates 会有delay的。不能保证一秒内完成。
replicate |
|
M*****n
发帖数: 2301 | 35 有些时候1秒其实算是蛮大的latency了
这里面有两个问题，一个是Consistency，the C in CAP，
一个时Durability，the D in ACID
有些系统，靠牺牲D来达到C，用memory-to-memory replication来号称
sub-second XDC consistency, 这样是能做到 <1000ms 甚至<100ms
的latency，不过也是cut corners就是了，replicated data is not even
locally committed， 这种要是出了问题recover也挺麻烦。
所以看XDC，latency不够，还要看它是具体怎么做的。
的， |
|
c****y
发帖数: 3592 | 36 用不同的risk factor来replicate一个fund's return, 比如用spx, vix etc...
或者说来replicate一个股票的return.
最简单的就是用regression,但基本上没啥用,tracking error巨大,beta没啥参考价值
请问还有啥别的办法么? 有个问题是那个fund如果是采用hedge的策略,e.g. sell
covered call.那index的参考价值就不大了
有人有经验么? |
|
h*****y
发帖数: 218 | 37 一般说
data is replicated at N nodes.
那么这个数据到底是有N个copy还是N＋1个，也就是N到底是total还是指纯粹的冗余？ |
|
w***w
发帖数: 1552 | 38 ThinkPad USB Port Replicator with Digital Video 这个性能怎么样？好像就只是通
过usb口连接，不知道性能怎么样。
考虑进一个 |
|
h****g
发帖数: 1 | 39 After a routine replication from data server A to B, the
index in a paticular
table is always lost. Why? |
|
a*****a
发帖数: 438 | 40 Has anyone ever run SQL replication (MS SQL 2000) on a cluster (Active/Passive)
and offload the distribution agent on passive node?
I searched the whole world and figured I am the only one:(run into some
problems and want to find help.. |
|
t**i
发帖数: 9 | 41 Can anybody shed some light on the knowledge of SQL replication? What's it?
And how to do it?
Thanks a lot! |
|
w*******e
发帖数: 1622 | 42 干嘛买Enterprise Edition呢?? Standard Edition就能Replication, 价格还便宜 |
|
w*r
发帖数: 2421 | 43 唉，dataase里面mssql算是最最便宜的了吧，你说的replication如果M$自己没有解决
方案的话那………………
2008不是号称有CDC很强大吗？具体偶也不明白…………我知道informatica power
exchange可以capture
data change at record level on oracle.M$的支不支持就不太清楚了 |
|
w*******e
发帖数: 1622 | 44 CDC?? 俺咋觉得也不怎么样呢??---(因为多出了些CDC table(s)..)
CDC在运作时先在source server里生成CDC table(s), 然后用Query functions把
changed data给移到另一个server上....
(看着跟replication差不多, 只是多了个CDC table(s)...) |
|
a*******t
发帖数: 891 | 45 need to test if snapshot replication puts the target database off line |
|
i****a
发帖数: 36252 | 46 Ms sql server? Replication takes care of pk automatically. In fact it
requires a pk for repl to work |
|
v*****r
发帖数: 1119 | 47 I guess your concerns is that your target db is inserted not only by the
replication process, so the the target db sequence might clash with the
source db sequence.
The workaround is simple, modify your target db sequence to start with a
much higher value to ensure the source/target db sequences are using
different range.
Your thought will work but the problem is you are making your application un
-scalable by "instance id + auto increased". |
|
g***l
发帖数: 18555 | 48 replication sp_addpushsubscription_agent 用户名错
用的是DOMAIN SERVICE ACCOUNT,PASSWORD放进去也不行 |
|
i****a
发帖数: 36252 | 49 it's better to use SQL account for replication especially if the servers
are not in the same domain. |
|
i****a
发帖数: 36252 | 50 In mssql you have to reset replication by forcing a ... Forget the term now.
.. something like re-publish articles. |
|
