Sunday, August 24, 2008
For example, suppose you have a cube that analyzes sales transactions. It has dimensions that describe customers (geography, education, income level, gender), products (classification, color, size), time, and the sales rep through the organizational structure. The measures include information about revenue, quantity, and discounts.
One of the most common questions would be, "How many customers bought a specific product?" An even better and more general question might be, "How many customers are buying each product?"
Although this last question seems simple, it is not. A regular COUNT measure will not provide correct results because double counts may occur. If a single customer buys a product more than once, a regular COUNT with the measure will count the product sale by customer twice. In order to get the correct results, each customer needs to be counted only once. This is the classic DISTINCT COUNT problem, and it requires a fairly complex resolution in the online analytical processing (OLAP) environment.
The problem may become even more interesting if the question becomes, "How many customers bought a specific basket of products?" Take the "Diapers & Beer" example, "How many customers bought both diapers and beer?" This type of question falls under the Basket Analysis problem category.
This document discusses the techniques to solve these two classic problems, DISTINCT COUNT and Basket Analysis. It assumes that the reader has a basic understanding of the concepts of OLAP in general, OLAP Services in particular, and MDX.For the complete document you have to visit the Microsoft Technet site.
The howto guides can be found on johnturner.com
Friday, August 22, 2008
In this post I explained that initially we had some problems with the PXE booting of machines to install Linux via the network. The problem is that we have redundant network links to ensure network uptime in case of a cable failure. To make sure that network traffic is looping in redundant network loops we have configured a Spanning Tree Protocol (which is a OSI layer-2 protocol) to ensure that no looping is done.
If your switch is doing spanning tree calculations, you’re not forwarding any traffic at all within the first 50 seconds and that may exceed the timeouts in your PXE setup. On your switch’s ports, try using "spanning-tree portfast" to jump immediately to a forwarding state.
To configure your switch (in our case a Cisco WS-C3548-XL) you have to do some settings. I will explain the way to do this here:
Login to the Cisco switch go to enable. To do so you need to enter a password twice:
Connected to Cisco C3548XL switch pubswt11 at colo [ STORAGE ] PUBCAB06
User Access Verification
After this you are logged in in the switch in such a way you are able to do some settings. Now we will have to make some changes to port number 21. We have to set the spanning-tree setting to portfast and we like to set the speed to 100 and duplex to full. This is done by going to the configuration mode for the interfaces.
pubswt11(config)#interface FastEthernet 0/21
Now we can set the needed setting by entering the following commands:
You leave the configuration mode by a key combination of Cntrl+z. now you can save the configuration with a wr command and your done. Now PXE boot should be successful.
The reason I use this strange formatting of my e-mail address is because I do not like it to be picked up by some bot crawling the internet to find e-mail addresses.
If you feel you are in need to send me a encrypted messages you can send it to an other e-mail address and use my public key to encrypt your message. On http://pgp.mit.edu/ you can search for the latest key if you like. However it is also here. You can send messages using this key to suntacXX@XXdds.nl, just remove the XX’s from the mail address.
-----BEGIN PGP PUBLIC KEY BLOCK-----
Version: GnuPG v1.4.7 (Darwin)
-----END PGP PUBLIC KEY BLOCK-----
Thursday, August 21, 2008
- IOScheduling/VRscheduler: an implementation of the V(R) scheduling algorithm;
- IOScheduling/FIFOscheduler: a pure FCFS scheduler.
A disk scheduler is the part of the Linux kernel that reorders, delays, and merges requests for disk I/O to achieve better throughput and lower latency than would happen if all the requests were sent straight to disk. The scheduler aims to minimize disk head movement as far as possible. You can think of the scheduler as a queue of requests waiting for service at the disk.
If you like to check the latest kernel go to kernel.org
The SmartApps server farm consists out of an enormous number of Linux servers and a portion of them was in need of upgrading the operating system due to various reasons. Before we were running Red-Hat servers however now Oracle has its own Linux distribution. Because standardization is key when you operate large datacenters the decision to switch to Oracle Unbreakable Linux was a easy to make.
Because some of the servers who are in the first set of upgrades are production servers so the need was there to have an upgrade done in the fastest way possible to ensure not more than 30 minutes downtime per server. This would ensure that we would stay inside the maintenance time windows which are in the customer SLA’s.
Because of this the decision was made to kickstart all the servers. Instead of starting with CD1 we did a PXE boot and the servers would connect a central server which contains an image of Oracle Unbreakable Linux, via the fiber network the servers would install without (much) human intervention, some information files where used to bind specific server information and incorporate it in the install. Things like which IP address is bind to which network interface, which network mounts need to be made, which users should be created and which Oracle instance needs to be started on which server.
The generation of the kickstart script and creating it in such a way that it is so generic that we could use it on all the servers in the first upgrade batch of the server farm has been the most time consuming of all. However after the script was created and tested installation of a new server was indeed a <30 minute job.
The biggest problem we found when creating the script and started testing it was that PXE boot cannot handle in all cases some network router settings of Cisco routers. This was the case for non fiber optic servers. The problem is that we have redundant network links to ensure network uptime in case of a cable failure. To make sure that network traffic is looping in redundant network loops we have configured a Spanning Tree Protocol (which is a OSI layer-2 protocol) to ensure that no looping is done.
If your switch is doing spanning tree calculations, you’re not forwarding any traffic at all within the first 50 seconds and that may exceed the timeouts in your PXE setup. On your switch’s ports, try using "spanning-tree portfast" to jump immediately to a forwarding state. This solved the problem, for the rest we were able to do a very fast and successful upgrade of the first set of operating systems and are now planning to do the second very shortly.
If you like to know more about the working of the IEEE Standard 802.1D have a look at http://ieee802.org/1/ .If you like to know more about the algorithm behind the spanning tree protocol you can have a look at the this document http://www1.cs.columbia.edu/~ji/F02/ir02/p44-perlman.pdf
Monday, August 18, 2008
As you pay per click and your add is placed on a large number of site’s you will have a good click ratio in horizontal ad networks however the people who will be clicking will only be the people who are on mainstream website’s. The mainstream websites have the largest number of visitors and because of this the biggest success of having a visitor clicking on your add.
Having a large number of people clicking on your add may seems like a good thing, the thing which however makes the money is what they do after that. Because of the fact that those ‘ad clickers’ are visitors are visitors of main stream websites indicates that they are not the “true believers”, the real searches. When you have your add on a very niche website who is dedicated to for example only “fiber optic NAS equipment” the visitors will be for 90% people who are interested in fiber optic NAS equipment so placing an add with a Vertical Ad Network on this website will attract more people to you new fiber optic NAS server website.
If you place the same add on NAS equipment website you will have loads of people who will be clicking on the link because they like to find out about NAS equipment however as soon as they find out yours is only fiber optic enabled they will navigate to another website.
The vertical ad networks are trying to find all the corners of the internet where there is discussion about your specific network, they are looking for the thought leaders and the real buzz makers. Because they specialize in only one or two area’s they will know how to market and place your add’s and that way give you more value for your money.
Instead of having your add shown a thousand times on all kinds of websites you can now better target your audience and make sure your add is resulting in more real-world actions like having customers buying your products. This trend is started in the travel market and is now spreading to other market segments. If you are looking for a Vertical Ad Network in your market and you cannot find one,…start one. Starting a small company in a entrepreneur way in Vertical Ad Networking is a small investment and is paying off quickly as we have learned from other entrepreneurs. This is the reason we will be seeing more and more Vertical Ad Networks in the upcoming time.
I have pre-registered to see it as one of the first adapters and somewhere today the people who have registered will get a login send. Because I am a LinkedIn fan I will be looking very close to talkbiznow.com and inform you on all the things I encounter and I will invite you to join the site if I feel it has something to offer. However, a community site has to grow to show some use. An unpopular social networking site has no use; the more people attend the more useful it will get and the more the market value will grow.
Wednesday, August 13, 2008
A new download for Oracle jDeveloper is available from the Oracle website. On the change document the following new features are listed:
- JDeveloper Core IDE
- Java Coding and Agile Development
- Database Development
- UML Development
- Web Services Development
- Web / Ajax Development
- EJB 3.0 Development
- Oracle TopLink Development
- ADF Framework
- ADF Faces Rich Client
- ADF Databinding
- ADF Business Components
- ADF Task Flow
- ADF Swing
- ADF Data Visualization
- ADF Mobile
- WebCenter Development
- SOA Suite: SCA composite applications development
A nice new feature is the AJAX support now available in Oracle jDeveloper. To check the more detailed change log check the changes page at oracle.com
Monday, August 11, 2008
The end is near, PHP4 is at the end of the official support cycle. Newer versions have made sure that PHP4 will no longer be supported. However, there are so many systems that are depending on PHP4, coded by developers who have left the company for some time already. What will happen to all those unsupported code? No security fixes anymore to keep those systems clean and up to date?
No, PHP security expert Stefan Esser will continue to provide security patches for PHP4. To get security patches for PHP4 you will have to go in the future to the Suhosin website.
Urban art, art in the streets of your city, sometime just as a decoration and sometimes it has a functional use. A three-dimensional directional system for a parking garage, yes we can consider this art in my opinion. Stereoscopic paintings to show you where you have to go in the parking garage, it looks really cool and only when you look at it from a different angle you really see how it is done.
We already have seen things like this as an art form however I have never seen it put into a usable way.
Wednesday, August 06, 2008
This concept is working two ways, Google is plugging their application on more and more websites and if they ever will start playing adds before showing you the content of a clip on youtube they will be on millions of websites the moment they flip the switch. On the other hand developers are benefiting from the coding done by Google and the backend servers which the API is talking to.
Some examples of websites powerd by google API’s:
TOTLOL is a video website where you can place a kid behind without any problem, the website is showing video content from youtube for 6 month olds till 6 year olds.
Thematicmapping is used to generate visual representations of data on the world globe in Google earth. You can quickly see your data represented geographically.
You can find more project on this google weblog: http://google-code-featured.blogspot.com/
Tuesday, August 05, 2008
• 25% of respondents said that the main focus of strategic discussions between top management and IT is purely financial;
• less than 25% believe that IT projects will significantly contribute to business growth in the next three years (i.e. projects related to product innovation, take-overs, mergers and a thorough go-to-market strategy);
• 40% of business respondents are rather pessimistic about the success ratio of IT projects and state that one in three projects is not finished within the deadlines or budget;
• 71% of respondents believe that their senior management is more or less comfortable with outsourcing certain activities. This figure is higher than in 2007. However, 38% of companies limit themselves to body shopping and insourcing;
• 30% of respondents state they were never involved in IT-related incidents. One can wonder whether this reflects the reality and whether it is not rather linked to a lack of transparency around incidents.
You can download the report from the Deloitte page.
Finding files in a UNIX system can be difficult sometimes. The find command will help you find all the files you need, however the command is not that easy to use if you like to use it to its full strength. Oracle has released a paper on the command “Guide to Linux Find Command Mastery” done by Sheryl Calish.
also a very good guide is this about page: http://linux.about.com/od/commands/a/blcmdl1_findx.htm done by Juergen Haas or you can take a look at the man page on my website.
“The find utility recursively descends the directory hierarchy for each path seeking files that match a Boolean expression written in the primaries given below.
find will be able to descend to arbitrary depths in a file hierarchy and will not fail due to path length limitations (unless a path operand specified by the application exceeds PATH_MAX requirements).