W***n
发帖数: 11530 | 1 How Index Funds Prevailed
Sep 6, 2016 3:52 PM EDT
By
Stephen Mihm
Forty years ago last week, Vanguard’s John Bogle created the first index
mutual fund, offering investors a guarantee: They would never outperform the
market, but nor would they underperform it. The Index Investment Trust (now
the Vanguard 500 Index Fund) simply tracked the performance of the S&P 500.
Index funds are now a huge business, accounting for trillions of dollars of
mutual fund money. And with good reason: Even though the... 阅读全帖 |
|
w*r
发帖数: 2421 | 2 Most rdbms like oracle db2 SQL server optimizer is designed in the way to
shorten the io path cost to accommodate oltp. oracle index join hint will
bring both index into consideration. Bitmap index provided a faster
operation using and/or operation on the single index, however I doubt it
will benefit index join performance . Bitmap index probably more cost
effective when the distinct values in the index is low.
For low cardinality columns , partition is a better choice comparing to
index in mos... 阅读全帖 |
|
b**********r
发帖数: 3861 | 3 Features
A flexible premium and adjustable equity indexed life insurance policy is a
universal life insurance policy with an equity-indexing feature. The
insurance company accepts your premium payment and deposits the premium into
your cash value account, which is set up with the policy. Then, the
insurance company deducts the cost of insurance from the cash value account
and invests the premiums into long-term bonds. The interest generated from
the bonds is then invested into index call options... 阅读全帖 |
|
y****t
发帖数: 10233 | 4 问题三: index? what index??? index fund? what index fund???
Only 20% chance to beat index? What index are you talking about?
Not all indices were created equal, and not all index funds were created
equal.
Total market index? A composite index? What composite index? who made the
composite index? you mix and create that index? Have you back tested it? Who
back tested it? What are the suckers you are comparing to? 30 years? what
about the suckers without 30 year history? What about the sucker which
ch |
|
s******d
发帖数: 323 | 5 ☆─────────────────────────────────────☆
yourtt (rainmaker) 于 (Fri Nov 16 02:14:51 2007) 提到:
问题三: index? what index??? index fund? what index fund???
Only 20% chance to beat index? What index are you talking about?
Not all indices were created equal, and not all index funds were created
equal.
Total market index? A composite index? What composite index? who made the
composite index? you mix and create that index? Have you back tested it? Who
back tested it? What are the suckers you are compa |
|
|
w******1
发帖数: 154 | 7 我看到了这个bogleheads principle:
http://www.bogleheads.org/wiki/Fidelity
这个里面有下面的 stocks:
US Stocks
Fidelity Spartan Total Market Index Fund (FSTVX)
Fidelity Spartan 500 Index Fund (FUSVX)
Fidelity Spartan Extended Market Index Fund (FSEVX)
Fidelity Spartan Mid Cap Index Fund (FSCKX)
Fidelity Spartan Small Cap Index Fund (FSSVX)
Fidelity Spartan Real Estate Index Fund (FSRVX)
International Stocks
Fidelity Spartan Global ex U.S. Index Fund (FSGDX)
Fidelity Spartan International Index Fund (FSIIX/FSIVX... 阅读全帖 |
|
E******w
发帖数: 2616 | 8 其实我并不觉得buy & hold index fund不好。但是它只是投资策略之一（当然是一个
很重要的策略），我只是反对把它教条化。
新手在没有很多知识和技巧的时候，坚定地buy & hold index fund，可以轻松拿到
market average return rate。而且华尔街操纵市场的各种策略，对buy & hold index
fund基本上都没用。打个比方来说，这好比是下围棋的时候因为棋盘对称，你跟着对
手下模仿棋。你如果自己水平不行，下模仿棋一般不会输很多。重要的是，哪怕对面碰
上的是9段高手，模仿棋也照样是能下的。因此，buy & hold index fund对普通人来说
是一个非常重要而且实用的投资策略。
而且，buy & hold index fund的策略可以有很多简单的衍生策略。稍微做点研究就不
难掌握。比如，在mortgage rate低的时候，你可以做cash out refinance，把钱拿出
来去买index fund。如果贷款利率是4%，你冒的风险只是几年的market average年平均
收益可能低于4%。其实就算真是这样，你的损... 阅读全帖 |
|
s*****e
发帖数: 21415 | 9 股市里的人一般都健忘，特别inpressionable。类似小鸭子，基本上生下来看到
谁就认谁当妈。
牛市火了几年就觉得股票市场或者房市一直牛；熊市熊了几年就觉得经济很快
要崩溃，世界很快要末日；黄金火了几年就有人叫嚣要回复金本位；如此种种
奇谈怪论，不一而足。
现而今有index至上论。平心而论，index是一种不错的投资方式。不过do better
也不像很多人说的那样near impossible。
前几年世界经济动荡的很，从美国金融危机开始，到后来的欧债危机一会儿一抽
风，基本上整个市场correlation很大，要涨一起涨，要跌一起跌。这种状况持续
了几年，所以看起来index投资是最靠谱的，单个股票的分析和选股价值不大。这
也是现在index至上论的根源。
不过，如果说我们对市场足够了解的话，things always change，现在correlation
大，不代表将来correlation大。其实index投资惨的时候也不是没有过，2000年
internet泡沫破裂的时候，talk to the indexers! 包括08年危机的时候，你要是
手握index满仓... 阅读全帖 |
|
w******1
发帖数: 154 | 10 我刚学习了很多index fund的东西，刚把fidelity 里的fund从freedom 2045换成了
index fund,我们学校就要把fidelity里的funds 换成vangard 的。vanguard 2045
expense ratio 只有0.18%。我是不是白折腾了？直接把所有刚换的index funds 换成
vanguard 2045？
这是vanguard index funds: er
Vanguard 500 Index Fund Admiral Class 0.05%
Vanguard extended market Index Fund 0.1%
Vanguard total international stock Index admiral shares 0.14%
PIMCO total return fund administrative class 0.71
网上看到有人说，自己配比的index funds portfolio不可能好过life cycle fund，是
指的我这两种选择吗vanguard... 阅读全帖 |
|
z***y
发帖数: 7151 | 11 When you alter clustered index, all non-clustered indexes will repointed to
new clustered index. Also since we only alter clustered index database
engine will not spend time inventory the existing constraints--they will be
left untouched.
However, when you drop the clustered index, all indexes(clustered and non
clustered) are gone. All constraints will be modified/removed.
Actually my first statement is not 100% correct I just realized... it is
true that non clustered indexes will not be dropped |
|
a*******t
发帖数: 891 | 12 According to BOL, TechNet etc,
Alter Index Rebuild = Drop index and Create index
But after I drop an index, it takes 2 hours to create index
whereas Alter Index Rebuild take 15 mins.
Why??? |
|
v*****r
发帖数: 1119 | 13 Very good blog, but the blogger didn't actually answer the question he
raised:
"So how is it you can have good perf in Oracle, w/o IOTs, but in SQL Server,
everyone says need CIs for good perf?"
My answer is Oracle simply has more options to achieve the same that
SQLServer clustered-index could achieve.
Let's first think about what kind of table will benefit most from using
clustered-index/IOT. One typical example is table storing time series data
will benefit a lot. Using clustered-index/IOT ... 阅读全帖 |
|
z****e
发帖数: 54598 | 14 今年年度语言是java应该无悬念了
December Headline: Java's popularity is going through the roof
There is hardly any doubt about it, Java will become TIOBE's programming
language of the year 2015. The steep fall of Objective-C of almost 8% this
year seems to be completely absorbed by the most popular language at the
moment. More about Java next month. Another interesting move concerns the
rise of Python. It is currently at its all time high position. It is easy to
learn, available everywhere and embraced by ind... 阅读全帖 |
|
A********h
发帖数: 2789 | 15 2014 Car Complaint Index
Thanks to consumer advocate Jack Gillis, author of the 2014 Car Book, and
the efforts of the Center for Auto Safety, we are able to provide you with
the Top 20 vehicle complaints on file with the National Highway Traffic
Safety Administration (NHTSA). Each year, thousands of Americans call their
government to register complaints about their vehicles. The complaint index
is based on a ratio of the number of complaints for each vehicle to the
sales of that vehicle. The num... 阅读全帖 |
|
s*****s
发帖数: 94 | 16 是不是大约这个意思， 问的不是数据库里面如何做index而是kafka内部如何做index
0.8中增加了逻辑offset，那么就需要做逻辑offset和物理地址间的转化
简单的方法，直接用hashmap，cache所有offset，问题就是这样空间耗费比较大
所以kafka的方式，是分段索引，用offset通过二分查找中index中找出段的起始地址，
然后再去file里面遍历找出精确的地址， 时间换空间的设计
1. LogSegment.translateOffset
首先是从index文件中找到近似的物理地址
前面说了，index中从效率考虑并不会为每个offset建立索引entry，只会分段建立
offset索引， 所以从index中直接可以找到精确物理地址的概率不大，但是可以找到最
接近的那个物理地址
如果你觉得index的粒度比较粗，可以直接给出开始查找的startingFilePosition
所以精确的物理地址需要到MessageSet文件里面去继续找 |
|
|
c*****t
发帖数: 1879 | 18 It depends.
For primary clustering index, there isn't a need of a separate index
file since the records are already in that order.
PostreSQL doesn't use primary clustering index, so all indexes are
secondary.
The details of how indexes are stored are also implementation specific.
For example, there are hash, b+ tree, r-tree, gist indexing etc. Some
implementation basically treat index entries as another relational
table.
Typically, the entire value of the column(s), or the result coming
from ev |
|
p*****1
发帖数: 128 | 19 HP申请一个create partially sorted index的专利 (US 2008/0104102)。
"To provide an index for a table in a database system, the index is
partially sorted in an initial phase of building the index. Subsequently, in
response to accessing portions of the index to process a database query,
further sorting of the accessed portions of the index is performed."
it is using B-Tree, divid index into different internal nodes, will sort the nodes (but not the entire tree). where each node may overlap with each other. |
|
j****y
发帖数: 1714 | 20 我想自己设计一个stock index 假设叫 Joy Index, 这个index 基于S&P 500 index (
或者Russell 3000)， 但是根据一些简单条件筛除一些公司的股票。
筛选后需要模拟这个index相对 S&P 500 index的过去1, 3, 5, 10, 20, 30年的收益，
风险。 最终的目的是想模拟一个mutual fund（track 这个 Joy Index）的风险和回报
如何。
请问这样的模拟一个人可以搞定否？工作量多大？如何搜集股票的相关历史数据？
我背景是CS, 编程应该没问题。
回帖的前10个都发一个包子表示感谢！ |
|
l**********t
发帖数: 5754 | 21 indexer managers have a design decision to make -- lower tracking errors at
higher cost (thus lower expected return), or higher TE with lower cost(thus
higher expected return). so there is a trade-off between TE & expected
return.
For example, typical indexes follow mechanic membership/rebalancing rules
known to everyone. Active managers front run passive indexers based on these
rules, and benefit at the cost of the passive, faithful indexers (these are
some of the hidden costs of passive inves... 阅读全帖 |
|
w******1
发帖数: 154 | 22 谢谢。这个index fund怎么样？Exp Ratio 0.24%
Fidelity Four-in-One Index Fund" (FFNOX)
https://fundresearch.fidelity.com/mutual-funds/composition/31634R109
The composition is:
Domestic Equity Funds 59.77%
Spartan 500 Index Fund - Investor Class 48.10%
Spartan Extended Market Index Fund - Investor Class 11.67%
International Equity Funds 25.18%
Spartan International Index Fund - Investor Class 25.18%
Investment Grade Bond Funds 15.03%
Spartan U.S. Bond Index Fund - Investor Class 15.03%
Cash & Ne... 阅读全帖 |
|
P*****e
发帖数: 313 | 23 我现在在学习投资，考古板上推荐的three-fund protfolio。比如说：
33% Vanguard Total Bond Market Index Fund VBMFX
34% Vanguard Total Stock Market Index Fund VTSMX
33% Vanguard Total International Stock Index Fund
但是现在international market不是很好，相对来说US market好一点。是不是可以考
虑只买
Vanguard Total Bond Market Index Fund VBMFX
Vanguard Total Stock Market Index Fund VTSMX
大家怎么看？现在投资total world stock index 风险大吗？ |
|
n*****r
发帖数: 159 | 24 我觉得能跑赢大盘的fund是有一些的，比如Buffet，还有一些high frequency trading
的。但是有很大一批active managed fund的历史并不长，一两年干的好就可以浑水摸
鱼。
还有就是很好的fund一般都被大机构投资者独占了，一般投资者也很难进去。08年金融
危机又给打着对冲避险基金旗号的很多fund一记闷棍。基于这样的考虑，index ETF就
开始受欢迎了。
[在 youngie88 (a4youngie) 的大作中提到：]
:最近研读了几本bogleheads.org推荐的投资的入门书籍。有一个体会就是每个作者都
在说，没有人可以time the market，所以呢只要买被动跟踪指数的Index Mutual Fund
就好了。理由大致如下：index MF的费用低，长期下来你的本金增长就快。actively
:managed MF因为要养一堆人来研究，做高卖低买的操作，故而费用就高了。更重要的
是这些主动管理的基金可以一段时间跑赢大盘，但是长期几乎没有做到的。只是不同的
:actively managed MF轮流做庄罢了。
:我对上述观... 阅读全帖 |
|
发帖数: 1 | 25 在thinkorswim中经常看到这个指标，以前没引起过注意，今天仔细看了一下还是可用
的指标：
What is the volume Index (VI) formula and how is it calculated?
The trade volume index (TVI) measures the amount of money flowing in and out
of a security or the market. The TVI depends on the direction of the
security and whether securities are accumulated or distributed. The TVI
generally uses a security's intraday price data.
To calculate the TVI, the minimum tick value of the security must be known.
Next, the change in price must be calculated... 阅读全帖 |
|
z***o
发帖数: 32 | 26 注意到WF index 12月11号狂跌。新闻报道有说 Trump 说 wells fargo 罚款不会取消
。我想这是导致 index 狂跌的原因，可是 WF index 基本与 SP 500 类似的配置，关
于 wells fargo 的负面消息为什么会影响 index 那么大呢？
https://www.google.com/search?source=hp&ei=axdLWsyqMIamjwPvhaKQCg&q=wells+
fargo+index&oq=wells+fargo+index&gs_l=psy-ab.3..0l6j0i203k1.450.12467.0.
20080.18.12.0.5.5.0.133.1117.9j3.12.0....0...1c.1.64.psy-ab..1.17.1183.0..
35i39k1j0i131k1j0i20i264k1.0.0SATXxdwbUw |
|
F****n
发帖数: 3271 | 27 【 以下文字转载自 Java 讨论区 】
发信人: Foxman (今狐冲), 信区: Java
标 题: 为啥RDBMS只用一个Index?
发信站: BBS 未名空间站 (Sat Mar 28 15:46:49 2015, 美东)
不久前有个项目要比较Lucene和主要RDBMS的search performance。因为老板想知道如
果把RDBMS当成NoSQL（就是Table Design的时候坚持denormalize) 用效果如何。经过
一段时间研究,
发现在其他条件不变的情况下单个column的search二者差不多，但多个fields/columns
的 query
RDBMS要慢的多 (e.g., select * from users where last_name='xxx' and email='
yyy")
进一步我发现RDBMS对一个Table在一个Query中居然一直只使用一个Index, 即使所有的
Column都有Index！ 一开始以为是Query Planner根据selectivity的选择，但后来发现
根本不是这样：
1. 几乎所有的Quer... 阅读全帖 |
|
t***s
发帖数: 163 | 28 偶出于服务new cop的目的，疯狂打字，版主mark一个，看客送个包子吧，呵呵
上回我们说了
Asset allocation, buy & hold, low cost最重要
什么是index investing
Why index investing works？
结论是散户beat index的几率很小
你看完以后嗤之以鼻，你坚信自己是天才，或者通过自己的努力可以成为beat index的
散户
（Emulex会叫这种人为sucker，我从来都是很尊敬有这种想法的人的，呵呵）
现在我们的论题变了：
How to improve you odd to beat index?
按照本版的特殊情况，我分成两种解决方案：
A. 选active manage fund （bibbub大师的做法，前提Asset allocation, buy & hold
, low cost）
B. Do-it-yourself, 包括 选股，market timing，sector bet
要分析这两个方案，必须说到alpha （beta, R-squared, sharp ratio），new co |
|
s******d
发帖数: 323 | 29 ☆─────────────────────────────────────☆
tonys (基因决定人生) 于 (Thu Nov 15 12:42:23 2007) 提到:
偶出于服务new cop的目的，疯狂打字，版主mark一个，看客送个包子吧，呵呵
上回我们说了
Asset allocation, buy & hold, low cost最重要
什么是index investing
Why index investing works？
结论是散户beat index的几率很小
你看完以后嗤之以鼻，你坚信自己是天才，或者通过自己的努力可以成为beat index的
散户
（Emulex会叫这种人为sucker，我从来都是很尊敬有这种想法的人的，呵呵）
现在我们的论题变了：
How to improve you odd to beat index?
按照本版的特殊情况，我分成两种解决方案：
A. 选active manage fund （bibbub大师的做法，前提Asset allocation, buy & hold
, low cost）
B. Do-it-your |
|
g********t
发帖数: 168 | 30 Now investing in Schwab Funds gets you more for your money
We’ve taken significant steps to reduce expenses and simplify the share
classes in our Schwab Equity Index Funds:
Schwab Equity Index Funds now have lower expenses. We’ve reduced fund
operating expenses, making Schwab Equity Index Funds some of the lowest-
cost index funds you can find anywhere. Our S&P 500 Index Fund, for
example, has an expense ratio of just 0.09%, well below the industry
average.
Every investor gets our lowest fund ex |
|
b*****e
发帖数: 1125 | 31 May ROR Number of funds reporting YTD
throughMay
Barclay Hedge Fund Index 4.94% 60 9.86%
Hedge Fund Industry Money Under Management
SUB INDICES
Convertible Arbitrage Index N/A 0 15.20%
Distressed Securities Index 6.04% 2 9.12%
Emerging Markets Index 9.18% 3 19.13%
Equity Long Bias Index 17.13% 6 |
|
t***n
发帖数: 546 | 32 说index并不是只持有一种index，而是用index来配出自己理想的asset allocation，
并且不停的rebalance。用index是因为fee低，而且一般在那个asset的efficiency
frontier。正真的risk 和return由asset allocation决定。
如果真的坚持固定的asset allocation并且及时rebalance的话，在2000年的internet
泡沫前已经卖了不少到bond了吧。在08年也会卖了不少bond来填股票，到现在股票
index都涨了一倍了吧。
关键要rebalance，这样才能take advantage of各个asset class之间的low
correlation。有涨有跌才能赚钱嘛。 |
|
b*****o
发帖数: 715 | 33 这些是由index fund组成的fund。
比如， target retirement fund就是由total us stock index, total international
stock index, total us bond index, total international bond index按某种比例
混合而成的。 |
|
M********r
发帖数: 278 | 34 Key to index investing is not index per se. Index investors normally have a
diversified and balnced portfolio with a basket of indexes that cover
equities and bonds, domestic and foreign. some may even include real estate
and commodities. Risk adjusted returned is then determined by the
composition of the portfolio. Yes, at times some index in the portfolio may
be overvalued, but the risk associated with these overvalued assets are
mitigated by other non-correlated assets. As a passive investor,... 阅读全帖 |
|
r*****3
发帖数: 570 | 35 你去问经理，他不推荐index fund，因为index fund他没钱赚。稍有点股票知识的人都
知道，index fund费用最低，而且index fund长线来说beat 90%的基金。要是看长线，
根本没有基金可以年复一年的beat the market.你说有些mutual fund可以胜过市场，
但只是短期，你怎么知道那基金以后的几年怎样呢？所以我个人认为还是买index fund
比较划算。 |
|
y*******8
发帖数: 162 | 36 最近研读了几本bogleheads.org推荐的投资的入门书籍。有一个体会就是每个作者都在
说，没有人可以time the market，所以呢只要买被动跟踪指数的Index Mutual Fund就
好了。理由大致如下：index MF的费用低，长期下来你的本金增长就快。actively
managed MF因为要养一堆人来研究，做高卖低买的操作，故而费用就高了。更重要的是
这些主动管理的基金可以一段时间跑赢大盘，但是长期几乎没有做到的。只是不同的
actively managed MF轮流做庄罢了。
我对上述观点能够理解，但是觉得不能完全赞同。第一个想到的问题就是：在现今美国
的mutual fund这个投资标的下面，我不知道究竟是actively managed和index mutual
fund各占多少份额？不知道哪里可以查到相关信息？第二个想法就是每个投资者的情况
不一样，有些人喜欢自己钻研，有一些就是不要折腾。在很多行业，大家都说让专业的
人做专业的事情，术业有专攻。如果大家都觉得投资界的专业人士都是带来不了额外收
益的？那么大家为何愿意付钱给这些人提供高薪？第三个想法就是... 阅读全帖 |
|
l******o
发帖数: 279 | 37 实在是很穷，140过了也没多少包子可发，就干脆不发了（Sorry！)，分享一下我的
Exhibits Index 和 Outline of Contributions 作为对本版众多热心ID的回报吧。如
果能顺便挣些包子，485过了定当散尽家财，希望到时粮仓充实了点 ：－）
我的case他引一项很弱，所以我Exhibit Index中有意突出了一些闪光点，在文章（18
－65）、引用（69－79，141－145）和审稿（80－87）方面和网上看到的模板以及律师
提供的都有所不同，有意突出了自己认为比较强的方面：由于是跨学科，杂志分布比较
广，SJR 排名里找到了最适合自己的用上了。希望对和我情况类似的有所帮助。
Exhibits Index是我自己准备的，分享应该没问题。PL、RL 有很多律师的努力在里面
，就不共享了。下一片介绍 Outline of Contributions，是在律师给基础上大幅修改
了的，只列出 Outline。
INDEX OF EXHIBITS
Letters of Recommendation
Exhibit 1: Expert Testimony ... 阅读全帖 |
|
g******4
发帖数: 6339 | 38 [1] FHFA/OFHEO
The US Federal Housing Finance Agency (formerly Office of Federal Housing
Enterprise Oversight aka OFHEO) publishes the HPI inx, a quarterly broad
measure of the movement of single-family house prices.
The HPI is a weighted, repeat-sales index, meaning that it measures average
price changes in repeat sales or refinancings on the same properties in 363
metropolises. This information is obtained by reviewing repeat mortgage
transactions on single-family properties whose mortgages ha... 阅读全帖 |
|
g******4
发帖数: 6339 | 39 [2] S&P/Case-Shiller Indices
The Case-Shiller index prices are measured monthly and tracks repeat sales
of houses using a modified version of the weighted-repeat sales methodology
proposed by Karl Case and Robert Shiller and Allan Weiss. This means that,
to a large extent, it is able to adjust for the quality of the homes sold,
unlike simple averages.
As a monthly tracking index, Case-Shiller Index has long lag time. Typically
, it takes about 2 months for S&P to publish the results, as opposed ... 阅读全帖 |
|
g******4
发帖数: 6339 | 40 [1] FHFA/OFHEO
The US Federal Housing Finance Agency (formerly Office of Federal Housing
Enterprise Oversight aka OFHEO) publishes the HPI inx, a quarterly broad
measure of the movement of single-family house prices.
The HPI is a weighted, repeat-sales index, meaning that it measures average
price changes in repeat sales or refinancings on the same properties in 363
metropolises. This information is obtained by reviewing repeat mortgage
transactions on single-family properties whose mortgages ha... 阅读全帖 |
|
g******4
发帖数: 6339 | 41 [2] S&P/Case-Shiller Indices
The Case-Shiller index prices are measured monthly and tracks repeat sales
of houses using a modified version of the weighted-repeat sales methodology
proposed by Karl Case and Robert Shiller and Allan Weiss. This means that,
to a large extent, it is able to adjust for the quality of the homes sold,
unlike simple averages.
As a monthly tracking index, Case-Shiller Index has long lag time. Typically
, it takes about 2 months for S&P to publish the results, as opposed ... 阅读全帖 |
|
u******u
发帖数: 595 | 42 For any relational database (Oracle, MS SQL, MySQL, DB2...):
Primary key is a table constraint, it means Unique and NOT NULL.
alter table T add primary key (id); will create a primary key on id column,
in the meantime, it will also creat a uniq index on that id column.
You can create a unique index for id column on table T as below:
create unique index idx_t on t(id);
However, unique index column can be NULL.
Therefore, Unix index doesn't equal to primary key.
Send me email for more database q |
|
g***l
发帖数: 18555 | 43 INDEX在三个COLUMN上？这样对查询帮助不大吧。为什么不能一个COLUMN上一个INDEX呢
，你有PK么或者CLUSTERED INDEX么，如果没有的话，NONCLUSTED INDEX也不快。
240M太多了，应该SPLIT成多个TABLE,上面套个VIEW，用VIEW给用户查询,多个TABLE的话，一个一个加INDEX快多了 |
|
y****9
发帖数: 144 | 44
As far as I know, there are following types of tables in Oracle
- Heap organized tables ( default one, > 99% tables are this type in oracle
applications, in fact I did not see any of my production databases use any
other type of tables yet except for temporary tables - I use it, developers
have no idea about temporary tables in my work place)
- Index organized tables (IOT)
- Index clustered tables
- Hash clustered tables
- Sorted hash clustered tables
- Nested tables
- temporary tables
- Object... 阅读全帖 |
|
s**********o
发帖数: 14359 | 45 BJ，那个COLUMNSTORE INDEX是不是ORACLE BITMAP INDEX啊，记得以前就说INDEX在
GENDER上的时候，用BITMAP INDEX比较好。 |
|
c*****d
发帖数: 6045 | 46 第一点，当然可以创建bitmap然后and/or操作。但是bitmap和btree index存储方式完
全不同，并不是所有的字段都适合bitmap。你这两个字段都不适合bitmap
第二点，“composite index一般不index单个column，浪费很大”不知道你在说什么。
composite index本来就是在两个字段上作index
cost |
|
u**d
发帖数: 211 | 47 在你这个案例里，关键在于你的 query 包含了 SELECT *
根据语义 * 就要把每个 record 里所有字段都返回。就意味着任何 plan 最终都要访
问 base table。所以 optimizer 会选择，在查询第一个 index 之后，直接访问 base
table，再对剩下的 predicates 做过滤。
要想达到你想象的效果，首先 base table 要有 primary key (类似 Lucene 里 Doc
ID 的概念)。其次，SELECT 语句只选择 primary key。这样 optimizer 只会访问
indexes，因为 indexes 包含了处理改查询的所有信息，而不会再访问 base table 了
。在对所有 indexes 做 intersect 的时候，有些系统未必会一定选择 merge join，
也可能选择 hash join。这也说的通，因为 hash join 有很多特殊的优化，比 merge
join 好，即使看起来还要建 hash table 似乎多余。如果你坚信 merge join 一定会
更好，那就用 hint ... 阅读全帖 |
|
F****n
发帖数: 3271 | 48 不久前有个项目要比较Lucene和主要RDBMS的search performance。因为老板想知道如
果把RDBMS当成NoSQL（就是Table Design的时候坚持denormalize) 用效果如何。经过
一段时间研究,
发现在其他条件不变的情况下单个column的search二者差不多，但多个fields/columns
的 query
RDBMS要慢的多 (e.g., select * from users where last_name='xxx' and email='
yyy")
进一步我发现RDBMS对一个Table在一个Query中居然一直只使用一个Index, 即使所有的
Column都有Index！ 一开始以为是Query Planner根据selectivity的选择，但后来发现
根本不是这样：
1. 几乎所有的Queries, Query Planner都只会选一个Index，根本没有啥优化可言--像
上面那种简单的例子，即使用第二个Index可以提高几千倍的速度，也不会用。而
Lucene总是使用所有的Index然后sort-and-merge。Pe... 阅读全帖 |
|
t***s
发帖数: 48 | 49 【 以下文字转载自 CS 讨论区 】
发信人: truss (truss), 信区: CS
标 题: 【求教】Text Indexer for Large Volume of ASCII files【先谢】
发信站: BBS 未名空间站 (Wed Oct 17 14:02:57 2012, 美东)
我需要index大概四百万个ascii文件。哪位高人给推荐个好用的text indexer。多谢。
简单的说，就是个类似于text search engine里面indexer那一部分，但是提供比较方
便的command line access。最好在windows上。
具体点，就是能够index这四百万文件，把结果放在一个可以从command line方便读取
的repository里。可以是relational database或者其他proprietary的格式，只要能够
从command line或者perl之类的scripts读，结果能输出到ascii文件就可以。当然如果
可以从数据库里直接用SQL读更好。
安装越简单越好。最好都是command line。
我试过微软的sear... 阅读全帖 |
|
t***r
发帖数: 4 | 50 很多东西，做成了slides格式，需要一个index方便自己。
发现slides下index支持不太好，问题如下：
1。 index输出的标题为
[
index
]
我想去掉这个，自己写标题
2。 index不让双栏
问题不大，可以忍受，但是觉得很奇怪
各位有何高见？ Thanks |
|
