My coworker's note reminded me that new SAS adminstrators join the ranks every month so it seems like a good time to compile a list of online resources.  If I've overlooked your favorite, please share with everyone by commenting on this post. 

For connecting with other SAS administrators, there are several good options:

• SAS Deployment Community on support.sas.com
• Several LinkedIn groups but a couple of familiar ones are SAS Administration Team and SAS Architects/Administrators/Implementation Specialist
• SAS Blogs tagged for administrators such as the Wednesday series you're reading on SAS Users Groups as well as the SAS administration topics on The SAS Dummy.
• Blogs hosted by SAS users such as Paul Homes' platformadmin.com and Andrew Ratcliffe's NOTE: blog.

If you're responsible for installing and maintaining a SAS installation, you will find most of the information necessary to accomplish your tasks on these support.sas.com sites:

• System Requirements
• Install Center
• Third-Party Software Support
• System Administration topic in Samples & SAS Notes
• Migration Focus Area

 Prefer videos?  SAS Talks On Demand offers several videos for the SAS administrator. 

•  SAS® 9.3 Features and Functionality for Platform Administration
• Best Practices in SAS^® Administration 
• Best Practices for Configuring Your IO Subsystem for SAS®9 Applications

 And, of course, don't forget training and documentation:

•  Courses and training paths for the administrator
• SAS Global Certification program for platform administrators
• Product documentation for the SAS Intelligence platform

 Please let us know how SAS can continue to support this important role.

SAS Global Forum is one of the largest user-based conferences, and it has amassed a lot of treasure over the past three decades. Each year, the locale and the conference theme change to meet the current needs of SAS users. While the 80’s "large volumes" may have been associated more with the "big hair" trend, the 2013 conference was fashionably bigger and interested in large volumes of another kind--Big Data and Big Analytics!

To reflect the title of Dr. Dave Dickey’s paper, SAS conferences are all about finding the gold in your data. And like the information in Dr. Dickey’s paper, there are numerous conference treasures in the form of Paper Presentations, trend-setting interviews with industry leaders on Live Reports, geek wisdom on Tech Talks, on-site interviewers with SAS attendees and many other cool videos captured on demand for later viewing.

SAS Global Forum Take-Out is another treasure that offers a selection of some of the best audio presentations from the conference. For those of you primarily interested in reading technical papers and posters, we have it all! Proceedings for 38 conferences from 1976 to 2013 have been archived for your reading pleasure.

The meetups and networking events at the conference inspire each one of us differently. If you are inspired and interested in learning more about your local users groups and participating at similar engagements, take a look at the Happenings page. It has all the current information on customer events, SAS Talks, users groups, newsletters, blogs and much more. It’s a goldmine. The more you dig, the more gold you find!

We hope you find these SAS resources helpful and find creative ways to reap the benefits of this "pot of gold" by reading, commenting and sharing with your network.

 Image provided bySin Amigos//attribution by creative commons

To borrow a quote from Ron Burgundy in the movie Anchorman -  "I don’t know how to put this, but SAS 9.4 is kind of a big deal."   So why is SAS 9.4 a big deal?  

Presenters at Global Forum shared some of the SAS 9.4 story and there are many, many new features I could include here, but these are some of my favorites so far: 

• Clustering  support for the metadata and middle-tier servers enabling scalability and higher availability.  Through server clustering, where multiple servers manage copies of the same data, the threat of data loss or whole-system downtime is drastically reduced. Clustering also allows users and tasks to be spread across the clustered environment to distribute the workload and improve performance.
• Introduction of SAS Environment Manager, which provides SAS administrators with a deeper understanding of their SAS deployments  and increased capabilities to monitor and manage their SAS servers. Through a plug-in interface, SAS Environment Manager deploys software agents on each managed SAS server and interacts with these agents to gather health and availability information, perform resource control actions, collect server resource usage, and more.
• For mobile  delivery of SAS reports, there are the new options in Base SAS Output Delivery System (ODS) EPUB support. With ODS EPUB you can output SAS reports as e-books that can be read with iBooks on the iPad and iPhone.  Additionally, support is provided for output to HTML5 and Microsoft PowerPoint files.
• DS2 sessions and demos were highly attended as SAS programmers were interested in hearing about the new SAS programming language that allows code to be submitted from Base SAS sessions to run in-database to perform advanced data manipulation without moving the data out of the database.
• In the Opening Session, SAS CEO Jim Goodnight explained, “SAS 9.4 allows your IT team to deploy SAS with confidence that it meets the requirements for security, authentication, scale and resiliency.”  Attendees were interested to learn more about middle-tier and metadata server clustering, scalability support and increased authentication support in SAS 9.4.
• Goodnight’s statement during the Opening Session, “SAS 9.4 delivers cloud-friendly architecture,” added to considerable interest in the new SAS Cloud tools and technologies, including SAS App Central, SAS App Engine, SAS vApp technology and SAS Web Editor.  Cloud-friendly SAS enables rapid deployment of SAS in cloud environments, provides simple management tools and processes, and promotes innovation through a SAS cloud development platform.
• The SAS Web Application Server embedded in SAS 9.4 reduces overall cost and complexity of SAS deployments. By eliminating the cost to acquire, integrate,  maintain and support third-party software, the SAS Web Application Server saves you money and time.  By reducing integration complexity and embedding software optimized for SAS, the SAS Web Application Server provides right-sized, integrated technology that simplifies IT management and makes your SAS environment cloud-ready.
• Availability of High-Performance Analytics procedures (HP PROCs) to customers running on a single server. Customers will be able to leverage the performance benefits of multi-core computing as they take advantage of these new HP PROCs. An added benefit of the HP PROCs is that when you grow your SAS deployment from a single server to multiple servers, your high-performance code automatically scales to run in your multi-server, distributed environment.
• A Decision Management user experience that combines data management, business rules management, data lineage, orchestration and more – all from within one unified user interface.

Which SAS 9.4 topic is the "biggest deal" to you? Which topics would you like to hear more about prior to the June release?

There is plenty to say about SAS 9.4.  Be sure to check back for more information!

Why should you pre-assign SAS libraries?

As outlined in the SAS 9.3 Intelligence Platform: Data Administration Guide, there are two ways you can make sure that your server is aware of your library reference: 

• pre-assigning the library
• letting the client application define the library reference

Even if you use a LIBNAME statement in your program, it is useful to register SAS libraries and tables in SAS metadata so that they can be used by some of the SAS clients (for example, SAS Data Integration Studio). We saw this demonstrated in our last article when we assigned a libref through a LIBNAME statement versus one assigned in SAS metadata.  Different clients may assign libraries differently, such as SAS Data Integration Studio or SAS OLAP Cube Studio where the library can be automatically generated based on the metadata. 

When you pre-assign a library, you are making explicit which engine will be used to control the library (such as BASE or ORACLE). Pre-assigning the library has a number of benefits:

• Pre-assigning librefs helps maintain consistency across users and applications
• Libraries are always available to the server, regardless of how the program is run (batch versus interactive)
• Libraries are easier to migrate to new storage since the physical location can be abstracted from the user
• Pre-assigning libraries makes it easier for developers to eliminate redundant code (for example, not having to manage library references in stored processes)

As discussed last time, if you don’t pre-assign our library and try to use that library in SAS Enterprise Guide, you get an error.

PROC Print data=metaref.postassessments;
run;
ERROR: Libname METAREF is not assigned.

To see this libref as “Assigned” when connecting to the server, you would need to pre-assign the library.  This action tells SAS to run the assignment code on the server whenever the server is started. If you select “Library is Pre-assigned” in the Advanced options (SAS Management Console→Data Library→ Manager→<Library> Properties), you see three Pre-Assignment Types listed:

• By native library engine
• By metadata library engine
• By external configuration

Understanding three methods for pre-assigning libraries

So which of the three options are right for your libraries? Here is a nice graphic and description of what really happens behind the scenes when you access a library governed by one of these processes.  

Key to understanding these choices is to knowing the difference between the Metadata Library Engine method (option 2 above) versus the other two approaches that use the native engine. 

• In the case of the native engine, the system does not check the SAS metadata to see if the user has read access on the table; however, it does check the physical security and whether the user has permission to ReadMetadata. When you use a native engine, the data-level authorizations of Read, Write, Create, and Delete are not checked.
• If you want to use the metadata authorization layer to control Read, Write, Create, and Delete permissions, then you must pre-assign the library and use the Metadata Library Engine method. But remember, operating system level file permissions should always be considered whether you use the SAS Metadata Libname Engine or the underlying native engine.
• Additionally, as we have noted, it is not always efficient to pre-assign every library that you might want to use. (In fact, a little birdie reminded me that too much of good thing may affect performance.) You should determine which libraries will actually be needed before you choose to pre-assign them. You should also set READMETADATA privileges to limit which libraries a group of users sees, as not everyone needs every library.
• One related area you may want to consider is the notion of read-only access. You will note that in the properties for a library, you have the option of specifying the library as READONLY. This capability augments the metadata permissions on the library or table in the fact that it eliminates the possibility of anyone writing to or destroying elements contained in the library – even if they have metadata permissions to do so.

In the discussion that follows, I will focus only on the mainstream servers (Workspace Servers, Pooled Workspace Servers, Stored Process Servers, SAS/SHARE servers and OLAP servers) as these servers automatically read metadata when they start and assign the libraries. You will have to edit the server configuration files if you want these to be read by other servers such as SAS/CONNECT, SAS Data Step Batch Server or SAS/IntrNet.

So let’s summarize the three assignment options in SAS metadata and explore how each one works and why we would select one over another.

By Native Library Engine

When you select the native library engine as the Pre-Assignment Type, you will notice that, in SAS Enterprise Guide, the client immediately has access to the library and the defined metadata tables.

The library is assigned when the SAS session starts and referenced through a SAS System Option called METAAUTORESOURCES. SAS uses the library engine defined for the library in SAS metadata. As noted above, the data-level authorizations are not checked in the SAS metadata. To illustrate this, we will set the metadata permissions in this example to read only.

Then we will try to replace the table.

Here, you notice that despite the metadata permissions of Read and ReadMetadata, the user was able to delete the dataset. As we mentioned above, if the native engine is used to access the table, then metadata permissions are not evaluated. 

Note: For SAS 9.3, SAS has added a feature called metadata-bound libraries, which cannot be overridden. Metadata-bound libraries offer better protection than do other metadata-based approaches to access control because the enforcement begins with the physical data. The enforcement occurs regardless of how the user requests access to the data (metadata clients such as SAS Web Report Studio or direct (through a LIBNAME statement submitted from SAS Enterprise Guide). For more information, see the documentation for metadata-bound libraries.

By Metadata Library Engine

When you select the Metadata Library Engine as the Pre-Assignment Type, the library is similarly assigned through METAAUTORESOURCES options; however, this option uses the Metadata Library Engine that ensures access controls that are placed on the library and its tables and columns are enforced. Here, we submit a simple DATA step that tries to replace an existing dataset. Note that our metadata permissions did what they were supposed to do.

By External Configuration

The third and final option for pre-assigning libraries is to use an external definition or an autoexec file. This method essentially makes the library available to client applications but moves the code out of the user interface to an external file. This option is particularly useful if you want to manage your configuration outside of SAS metadata or share the configuration among multiple services such as SAS/Connect.

Four questions to help choose the best method

Since I am a visual thinker, I also thought it might helpful to show these comparisons in a flow chart. Admittedly, this is a simple set of decisions, but the chart should help you decide which pre-assigned library option might be best for your case.

Hopefully, this post was a useful discussion, and you know a bit more about what it means to pre-assign libraries and how the various options affect whether users can read or both read and write data.

Happy Data!

--greg

SAS users from all over the globe were accustomed to meet-ups and team building at conferences to talk about SAS, software and solutions. But the team that came together this time had a different agenda in mind. A team that believed in giving back to the local community as much as it got from the opportunity to gather in San Francisco to reconnect and learn from each other. The Build-a-Bike charity event was one such moment that brought back childhood memories and the carefree joys of riding a bike.

Attendees scattered into smaller groups armed with wrenches and excitement to assemble bikes for underprivileged children in the San Francisco area. . “I love bicycles, I cycle to work” said one such enthusiastic attendee stating that every time she comes to the conference, she meets new people and learns a lot. They built 50 bikes for young boys and girls in an hour. The mood was upbeat and it gave everyone involved a rare opportunity to connect outside of work, re-live their childhood and, most importantly, bring big smiles on the faces of some very lucky children. Many thanks to the participants for making this a memorable event and certainly one of the highlights from the conference.

What was your memorable moment from the conference and how would you like to give back to the community? Please share.

According to Berryhill, there are two very important things you can do to make the best of your time and those who look at your code later: Make sure your code is clear and correct. "To me, the main thing is correct. If the answers are wrong, it doesn't matter how you got there," he says.

Here are three of his tried-and-true tips:

1. "I like to use the %LET, particularly at the top of my code where variables are going to change," says Berryhill. He use a series of the %LET at the top of the program to remind him of changes he needs to make.
2. The NOBS option tells you how many observations you have in a dataset. According to Berryhill, the option is most useful when NOBS is 0. "If you try to do a PROC PRINT or a global dataset, you set that empty dataset, and Boom you're gone. (This only works for disk files, not views or tapes.)"
3. Try your luck. When he isn't on a tight deadline, Berryhill says that he likes to experiment with code just to see if it  works. Recently, he found that the double question mark will suppress expected errors in the input function. "I was surprised to find out that it is also supported in a statement."

Read 27 more of Berryhill's SAS tricks in his paper, "30 in 20 things you may not know about SAS."

Koketso Moeng says that Statistics South Africa has allowed its users to access internal sites from their smartphones for some time, but reading text heavy reports and tables on an iPhone can be frustrating. And even in the areas on the website where they were using visualization, the information still wasn't as accessible as they would have liked.

Choosing a platform

Moeng says the team had to carefully consider the level of security that would be offered. "We have a lot of very sensitive information that should not be released before it's time," he said.

Additionally, Statistics South Africa had to ensure that the platform integrated with its existing infrastructure.  "We have invested a lot of money in our SAS environment and our operating environment," Moeng said. "So whatever we bring into that environment must integrate - you don't want to have to reshuffle things just to bring an additional tool into your environment."

Finally, Moeng's team wanted to make sure their users were happy with the platform so they would use it. They conducted focus groups to test the mobile BI on several platforms. "We said, 'Play around with these tools and give us feedback. Which ones do you like? Why do you like them? and Which one is most appealing to you?'"

Pinch, swipe, tap

They decided to go with Roambi because of the easy interface and many of the end-users were already using the iPad and iPhone. Moeng says there are "two flavors" of Roambi: Roambi Analytics produces beautiful graphics and with Roambi Flow, you can embed those graphics. This gives the end-user an interactive document that gives context to the data visualization.

"Configuration is very easy," says Moeng. In a SAS Enterprise Business Intelligence environment, it takes only a matter of minutes to configure. (Roambi runs off TomCat and MySQL.) Roambi uses the URL from your SAS Web Report Studio environment to interact with SAS.

As you can imagine, Roambi is touch-enabled - pinch, swipe and tap to access and manipulate the reports. Moeng did his entire presentation using Roambi Flow on an iPad.

In Moeng's paper, "Extending SAS Reports to your iPhone," you can read more about publishing reports and giving access to your end-users. Also check out what SAS Visual Analytics can do on the mobile.

Each year, 8 million people visit the Alberta Parks. They're bird watchers. Hikers, bikers and canoers. They're searching for adventure or a romantic fire lit evening. Since 2002, Alberta Parks has been using a paper survey to learn about the experiences and expectations of those visitors.

According to Alberta Tourism, Parks and Recreation analyst Jared Prins, the structured data from the surveys is easy to analyze. But, what about all of that unstructured data - write-in comments like "Rangers need skimpier uniforms." The old way of doing this was to hand-type the comments into a spreadsheet and then convert that into a SAS data set for analysis.

This can be a long, painful process. Prins said manually coding the surveys would take him three weeks locked in his office so that he could concentrate only on the data. Now, the process is largely automated and takes just a few minutes. That is a significant time savings, but Prins says that's only part of it. Text mining gives Alberta Parks a deeper understanding of its data and provides consistency.

If you only wanted to know how many times a word appears in the data, a simple word count would do the trick. What happens when you run into homonyms or words that can have two meanings? The parsing and filter nodes process natural language. The text parsing node understands parts of speech so that it can distinguish between river bank and international bank. Look at this example:

"There was nowhere to park my car."

"This is a fantastic park to visit."

And the noise is further reduced by reducing words to their root form (stems, stemming and stemmed all become stem) and using Stop and Start lists. A Stop List can be set to  ignore words such as - the, and, is.

"At Alberta Parks, we're becoming more focused on discovering actionable intelligence from our text data," he said.  Text mining helps them do that. It also helps them get a deeper understanding of customer issues by making it possible to handle more open-ended questions.

Read Jared Prins' SAS Global Forum 2013 paper to learn more about how Alberta Parks is using SAS Text Miner. He also gives great ideas for setting it up for yourself. Read "Replace Manual Coding of Customer Survey Comments with Text Mining: A Story of Discovery with Text as Data in the Public Sector."